A History Of Ancient And Early - Singh Upinder.pdf

A HISTORY OF ANCIENT AND EARLY MEDIEVAL INDIA FROM THE STONE AGE TO THE 12TH CENTURY

Upinder Singh

Delhi • Chennai • Chandigarh Upper Saddle River • Boston • London Sydney • Singapore • Hong Kong • Toronto • Tokyo

Brief Contents Photographs, Maps, and Figures About the Author Preface Acknowledgements A Reader’s Guide Introduction: Ideas of the Early Indian Past 1. Understanding Literary and Archaeological Sources 2. Hunter-Gatherers of the Palaeolithic and Mesolithic Ages 3. The Transition to Food Production: Neolithic, Neolithic–Chalcolithic, and Chalcolithic Villages, c . 7000– 2000 BCE 4. The Harappan Civilization, c. 2600–1900 BCE 5. Cultural Transitions: Images from Texts and Archaeology, c. 2000–600 BCE 6. Cities, Kings, and Renunciants: North India, c. 600–300 BCE 7. Power and Piety: The Maurya Empire, c . 324–187 BCE 8. Interaction and Innovation, c. 200 BCE –300 CE 9. Aesthetics and Empire, c. 300–600 CE 10. Emerging Regional Configurations, c. 600–1200 CE A Note on Diacritics Glossary Further Readings References Credits

Contents Photographs, Maps, and Figures About the Author Preface Acknowledgements A Reader’s Guide Introduction: Ideas of the Early Indian Past THE MAIN PHYSIOGRAPHIC ZONES OF THE SUBCONTINENT WAYS OF DIVIDING THE INDIAN PAST CHANGING INTERPRETATIONS OF EARLY INDIAN HISTORY NEW HISTORIES, UNWRITTEN HISTORIES 1 Understanding Literary and Archaeological Sources READING ANCIENT TEXTS FROM A HISTORICAL POINT OF VIEW Ancient palm leaf manuscripts THE CLASSIFICATION OF LITERARY SOURCES: LANGUAGE, GENRE, AND CONTENT THE VEDAS THE TWO SANSKRIT EPICS: THE RAMAYANA AND MAHABHARATA Archaeology and the Mahabharata The chronological layers in the Ramayana THE P URANAS THE DHARMASHASTRA Theory and practice in the Dharmashastra BUDDHIST LITERATURE Songs of Buddhist nuns

JAINA LITERATURE SANGAM LITERATURE AND LATER TAMIL WORKS The stories of the two Tamil epics EARLY KANNADA AND TELUGU LITERATURE OTHER ANCIENT TEXTS, BIOGRAPHIES, AND HISTORIES Banabhatta and his royal biography THE NATURE OF ANCIENT INDIAN HISTORICAL TRADITIONS THE ACCOUNTS OF FOREIGN WRITERS Al-Biruni on the writing of the Hindus ARCHAEOLOGY AND THE EARLY INDIAN PAST SCIENTIFIC TECHNIQUES IN ARCHAEOLOGY Radiocarbon dating INTERPRETING ARCHAEOLOGICAL EVIDENCE ETHNO-ARCHAEOLOGY The social and cultural aspects of technology P ROTECTING SITES EPIGRAPHY: THE STUDY OF INSCRIPTIONS ANCIENT AND EARLY MEDIEVAL SCRIPTS LANGUAGES OF ANCIENT AND EARLY MEDIEVAL INSCRIPTIONS Deciphered and undeciphered scripts DATING THE INSCRIPTIONS How to convert ancient era dates into modern ones THE CLASSIFICATION OF INSCRIPTIONS Memorializing death in stone INSCRIPTIONS AS A SOURCE OF HISTORY NUMISMATICS: THE STUDY OF COINS A BRIEF HISTORY OF INDIAN COINAGE COINS AS A SOURCE OF HISTORY Counter-struck coins of the Kshatrapas and Satavahanas

CONCLUSIONS 2 Hunter-Gatherers of the Palaeolithic and Mesolithic Ages THE GEOLOGICAL AGES AND HOMINID EVOLUTION What does it mean to be human? HOMINID REMAINS IN THE INDIAN SUBCONTINENT PALAEO-ENVIRONMENTS CLASSIFYING THE INDIAN STONE AGE THE PALAEOLITHIC AGE LOWER PALAEOLITHIC SITES Typical lower palaeolithic tools Isampur: a centre of stone tool manufacture MIDDLE PALAEOLITHIC SITES The Levallois technique UPPER PALAEOLITHIC SITES Upper palaeolithic tools P ALAEOLITHIC ART AND CULTS Ostrich eggshell beads THE LIFE-WAYS OF PALAEOLITHIC HUNTER-GATHERERS Food resources—now and then THE MESOLITHIC AGE MESOLITHIC SITES Microliths Animal bones at mesolithic sites Graves, subsistence, and settlement patterns The journey to get chalcedony THE MAGNIFICENCE OF MESOLITHIC ART CONCLUSIONS 3 The Transition to Food Production: Neolithic, Neolithic–Chalcolithic, and Chalcolithic Villages, c . 7000– 2000 BCE

THE NEOLITHIC AGE AND THE BEGINNINGS OF FOOD PRODUCTION WHY DOMESTICATION? THE IDENTIFICATION OF DOMESTICATION AND FOOD PRODUCTION IN THE ARCHAEOLOGICAL RECORD The analysis of ancient plant remains THE TRANSITION TO FOOD PRODUCTION IN THE INDIAN SUBCONTINENT THE EARLIEST VILLAGE SETTLEMENTS IN THE INDIAN SUBCONTINENT, C. 7000–3000 BCE The north-west The Vindhyan fringes and other areas NEOLITHIC, NEOLITHIC–CHALCOLITHIC, AND CHALCOLITHIC COMMUNITIES, C. 3000–2000 BCE The north and north-west Did people actually live in the Burzahom pits? Rajasthan The Malwa region The western Deccan The middle Ganga plain and eastern India South India The mystery of the ash mounds Community feasting at neolithic Budihal THE LIFE OF EARLY FARMERS CHANGES IN CULTIC AND BELIEF SYSTEMS Female figurines—ordinary women or goddesses? CONCLUSIONS 4 The Harappan Civilization, c . 2600–1900 BCE CIVILIZATION AND URBANIZATION: DEFINITIONS AND IMPLICATIONS The 10 characteristics of cities, according to Childe RECENT DISCOVERIES AND CHANGING PERSPECTIVES HARAPPAN, INDUS, OR SINDHU–SARASVATI CIVILIZATION? ORIGIN: THE SIGNIFICANCE OF THE EARLY HARAPPAN PHASE

The problems with diffusionist theories THE RELATIONSHIP BETWEEN THE EARLY AND MATURE HARAPPAN PHASES THE GENERAL FEATURES OF MATURE HARAPPAN SETTLEMENTS PROFILES OF SOME HARAPPAN CITIES, TOWNS, AND VILLAGES THE DIVERSITY OF THE HARAPPAN SUBSISTENCE BASE Animal bones at Shikarpur HARAPPAN CRAFTS AND TECHNIQUES Sculpture in stone and metal The making of long carnelian beads NETWORKS OF TRADE Shortughai—a Harappan trading post in Afghanistan THE NATURE AND USES OF WRITING RELIGIOUS AND FUNERARY PRACTICES The ‘fire altars’ THE HARAPPAN PEOPLE How healthy were the Harappans? THE RULING ELITE Defining a state THE DECLINE OF URBAN LIFE THE SIGNIFICANCE OF THE LATE HARAPPAN PHASE CONCLUSIONS 5 Cultural Transitions: Images from Texts and Archaeology, c . 2000–600 BCE PERSPECTIVES FROM TEXTS USING THE VEDAS AS A HISTORICAL SOURCE The date of the Rig Veda WHO WERE THE INDO-ARYANS? THE CULTURE REFLECTED IN THE FAMILY BOOKS OF THE RIG VEDA SAMHITA Tribes and wars Hymn to arms ( Rig Veda Samhita 6.75)

Lineage, clan, tribe Pastoralism, agriculture, and other occupations Varna in the Rig Veda Women, men, and the household The family and the household Religion: sacrifices to the gods Hymn to Indra ( Rig Veda 2.12) The soma plant and its juice THE HISTORICAL MILIEU OF LATER VEDIC AGE TEXTS Aspects of everyday life The emergence of monarchy The ceremony of the jewel offering The varna hierarchy The Purusha-sukta ( Rig Veda 10.90) Gender and the household Religion, ritual, and philosophy The Nasadiya hymn ( Rig Veda 10.129) The sacrificial arena The atman , according to Uddalaka Aruni Popular beliefs and practices Atharva Veda spells ARCHAEOLOGICAL PROFILES OF DIFFERENT REGIONS OF THE SUBCONTINENT, c . 2000–500 BCE NEOLITHIC–CHALCOLITHIC AND CHALCOLITHIC CULTURES The north-west and north The Indo-Gangetic divide, the upper Ganga valley, and the doab The Sanauli cemetery The copper anthropomorph Black and Red Ware Western India

The middle Ganga valley Eastern India The North-east The cultural sequence in central India The chalcolithic farmers of the Deccan The Daimabad bronzes Food, nutrition, and health among the people of Inamgaon Goddesses with and without heads Neolithic–chalcolithic sites of South India Pictures on stone FROM COPPER TO IRON: EARLY IRON AGE CULTURES OF THE SUBCONTINENT A clarification about the Indian megaliths The north-west The Indo-Gangetic divide and the upper Ganga valley: the Painted Grey Ware culture Painted Grey Ware The evidence from Rajasthan The middle and lower Ganga valley Central India The Deccan South India The enigma of the megalithic anthropomorphs THE IMPACT OF IRON TECHNOLOGY THE PROBLEM OF CORRELATING LITERARY AND ARCHAEOLOGICAL EVIDENCE CONCLUSIONS 6 Cities, Kings, and Renunciants: North India, c . 600–300 BCE THE SOURCES: LITERARY AND ARCHAEOLOGICAL Panini and his Ashtadhyayi Northern Black Polished Ware THE 16 GREAT STATES

The identification of Taxila THE GANAS OR SANGHAS The conflict between the Sakyas and Kosalans Vassakara seeks the Buddha’s advice on how to defeat the Vajjis POLITICAL CONFLICTS AND THE GROWTH OF THE MAGADHAN EMPIRE The chronology of the early dynasties of Magadha THE PERSIAN AND MACEDONIAN INVASIONS The storming of the Malla citadel LAND AND AGRARIAN EXPANSION FROM VILLAGE TO TOWN: THE EXAMPLE OF ATRANJIKHERA THE EMERGENCE OF CITY LIFE Perceptions of the forest ARCHAEOLOGICAL AND LITERARY PROFILES OF EARLY HISTORICAL CITIES THE NORTH-WEST THE INDO-GANGETIC DIVIDE, THE UPPER GANGA VALLEY, AND THE DOAB THE MIDDLE AND LOWER GANGA VALLEY CENTRAL INDIA AND THE DECCAN URBAN OCCUPATIONS, CRAFTS, GUILDS, AND MONEY THE NEW SOCIAL ELITES: THE GAHAPATI AND SETTHI TRADE AND TRADERS CLASS, KINSHIP, VARNA , AND CASTE Activities in times of adversity Varna and jati GENDER, FAMILY, AND HOUSEHOLD Marriage, according to the Grihyasutras THE RENUNCIATORY TRADITION The Samannaphala Sutta THE AJIVIKAS EARLY BUDDHISM

THE LIFE OF THE BUDDHA THE BUDDHA’S TEACHINGS The analogy of the raft THE BUDDHIST SANGHA AND THE LAITY THE SOCIAL IMPLICATIONS OF THE BUDDHA’S TEACHINGS The Ambattha Sutta BUDDHISM AND WOMEN Patachara’s song The eight conditions imposed on nuns The seven kinds of wives3 EARLY JAINISM THE JAINA TIRTHANKARAS, VARDHAMANA MAHAVIRA THE JAINA UNDERSTANDING OF REALITY THE JAINA DISCIPLINE The liberated man On not killing earth bodies THE SOCIAL COMPOSITION OF THE JAINA SANGHA AND LAITY The true Brahmana Malli or Mallinatha? CONCLUSIONS 7 Power and Piety: The Maurya Empire, c . 324–187 BCE THE MAJOR SOURCES FOR THE MAURYA PERIOD KAUTILYA’S ARTHASHASTRA The statistical analysis of word frequencies in the Arthashastra MEGASTHENES’ INDICA The Greeks on Megasthenes ASHOKA’S INSCRIPTIONS The different categories of Ashokan inscriptions and their location References to famine relief in the Mahasthan and Sohgaura inscriptions

ARCHAEOLOGICAL AND NUMISMATIC EVIDENCE THE MAURYA DYNASTY Legends of Ashoka The stone portrait of Ashoka at Kanaganahalli LITERARY AND ARCHAEOLOGICAL PROFILES OF CITIES Pataliputra and the palace, according to Arrian and Aelian SOME ASPECTS OF RURAL AND URBAN LIFE THE NATURE AND STRUCTURE OF THE MAURYA EMPIRE Kautilya’s timetable for a king The life of a king, according to Megasthenes (via Strabo) Rock edict 6 (Girnar version) The Maurya state and forest people ASHOKA AND BUDDHISM Minor rock edict 1 (Rupnath version) ASHOKA’S DHAMMA The 5th pillar edict (Delhi–Topra pillar) The 13th rock edict (Shahbazgarhi version) Ashoka’s assessment of his success: the Shar-i-Kuna Greek–Aramaic inscription SCULPTURE AND ARCHITECTURE Ancient and modern quarries at Chunar The medieval and modern histories of Ashokan pillars The discovery of an Ashokan stupa at Deorkothar The Parkham yaksha , then and now THE DECLINE OF THE MAURYA EMPIRE CONCLUSIONS 8 Interaction and Innovation, c . 200 BCE –300 CE THE POLITICAL HISTORY OF NORTH INDIA THE SHUNGAS The Besnagar pillar inscription of Heliodorus

THE INDO-GREEKS Coins of the Indo-Greeks THE SHAKA-P AHLAVAS OR SCYTHO-P ARTHIANS THE KUSHANAS The Rabatak inscription THE SHAKA KSHATRAPAS OF WESTERN INDIA A lake, a storm, and a king THE SATAVAHANA EMPIRE IN THE DECCAN The royal portrait gallery in the Naneghat cave KINGS AND CHIEFTAINS IN THE FAR SOUTH: THE CHERAS, CHOLAS, AND PANDYAS The royal drum VILLAGES AND CITIES Plant remains from Sanghol CITIES OF THE NORTH-WEST THE INDO-GANGETIC DIVIDE AND THE UPPER GANGA VALLEY THE MIDDLE AND LOWER GANGA VALLEY AND EASTERN INDIA Chandraketugarh CENTRAL AND WESTERN INDIA CITIES AND TOWNS OF THE DECCAN CITIES OF THE FAR SOUTH Madurai in the Maduraikkanchi CRAFTS AND GUILDS Guilds as bankers TRADE AND TRADERS Ancient travellers LONG-DISTANCE TRADE Kaveripattinam in the Pattinapalai TRADE WITH EAST AND SOUTHEAST ASIA INDO-ROMAN TRADE

Periplus Maris Erythraei (The Periplus of the Erythraean Sea) Recent excavations at Arikamedu THE WIDER ROLES OF TRADE AND TRADERS ASPECTS OF SOCIAL CHANGE IN NORTH INDIA AND THE DECCAN: VARNA , CASTE, GENDER The Jatakas as a source of social history SOCIETY IN EARLY HISTORICAL SOUTH INDIA An ancient Tamil love poem A heroic death PHILOSOPHICAL DEVELOPMENTS: ASTIKA AND NASTIKA SCHOOLS The Bhagavad Gita LOOKING AT THE HISTORY OF RELIGIONS BEYOND THE FRAMEWORK OF ‘ISMS’ THE WORSHIP OF YAKSHAS AND YAKSHIS, NAGAS AND NAGIS GODDESSES, VOTIVE TANKS, AND SHRINES VEDIC RITUALS P URANIC HINDUISM Shivaism The formation of the Vaishnava pantheon Krishna and Balarama on Agathocles’ coins Shakti worship THE EMERGENCE OF MAHAYANA BUDDHISM Monastic and lay practices in texts versus inscriptions THE DIGAMBARA–SHVETAMBARA SCHISM IN JAINISM RELIGIOUS ARCHITECTURE AND SCULPTURE EARLY HINDU TEMPLES AND SCULPTURE BUDDHIST ARCHITECTURE AND SCULPTURE Stupa-monasteries of the north-west Central Indian stupas—Sanchi and Bharhut Stupas of Andhra Pradesh Early relief sculpture at Buddhist stupa sites

Buddhist caves in the Western Ghats THE JAINA CAVES AT UDAYAGIRI AND KHANDAGIRI THE GANDHARA SCHOOL OF SCULPTURE EARLY STONE SCULPTURES FROM VIDISHA AND MATHURA TERRACOTTA ART THE PATRONAGE OF RELIGIOUS ESTABLISHMENTS Gifts of water pots from ancient Gandhara Pious donations at Bandhogarh CONCLUSIONS 9 Aesthetics and Empire, c . 300–600 CE POLITICAL HISTOR THE GUPTA DYNASTY Ramagupta—did he exist? The inscription of Chandra and the legend of the unsteady pillar THE VAKATAKAS OF THE DECCAN A queen’s grant OTHER DYNASTIES OF PENINSULAR INDIA THE ADMINISTRATIVE STRUCTURE OF THE GUPTA AND VAKATAKA KINGDOMS An ancient panchayat? REVENUE RESOURCES OF STATES LAND OWNERSHIP TYPES OF LAND, LAND MEASURES, AND LAND TENURE ROYAL LAND GRANTS The terms of the Vakataka grants PATTERNS OF URBAN HISTORY The lifestyle of the nagaraka CRAFT PRODUCTION, GUILDS, AND TRADE ASPECTS OF SOCIAL STRUCTURE: GENDER, FORMS OF LABOUR, SLAVERY, AND UNTOUCHABILITY

Faxian’s account The ganika and kulastri in Sanskrit kavya PATTERNS OF RELIGIOUS DEVELOPMENTS THE EMERGENCE OF TANTRA THE EVOLUTION OF THE VAISHNAVA PANTHEON SHIVAISM THE CULT OF THE GREAT GODDESS THE WORSHIP OF OTHER DEITIES BUDDHISM Kumarajiva (343–413 CE) JAINISM A CLASSICAL AGE OF ART? RELIGIOUS ARCHITECTURE SCULPTURE SANSKRIT LITERATURE The cloud messenger The Natyashastra ASTRONOMY AND MATHEMATICS Ancient mathematical and medical manuscripts MEDICAL KNOWLEDGE The ideal hospital, according to Charaka CONCLUSIONS 10 Emerging Regional Configurations, c . 600–1200 CE SOURCES, LITERARY AND ARCHAEOLOGICAL New evidence regarding Wang Xuance’s missions to India POLITICAL NARRATIVE AND POLITICAL STRUCTURE The image of the ideal king in inscriptions of Orissa Rudramadevi, the female king THE DECCAN

The Aihole inscription of Pulakeshin THE FAR SOUTH Religious and political symbolism in the Tanjavur temple NORTH INDIA: THE P USHYABHUTIS, HARSHVARDHANA The life and travels of Xuanzang EASTERN INDIA Some origin myths of the dynasties of Orissa THE RAJPUT CLANS The Tomaras and Delhi in legends and inscriptions KASHMIR AND THE NORTH-WEST Didda ROYAL LAND GRANTS BRAHMANA BENEFICIARIES THE NATURE OF BRAHMADEYA SETTLEMENTS Kara-shasanas and kraya-shasanas THE IMPACT OF BRAHMANA SETTLEMENTS ON AGRARIAN RELATIONS LAND GRANTS AS PART OF LARGER SOCIAL AND CULTURAL PROCESSES RURAL SOCIETY: REGIONAL SPECIFICITIES Popular agricultural sayings of early medieval Bengal URBAN PROCESSES IN EARLY MEDIEVAL INDIA HISTORICAL PROCESSES IN EARLY MEDIEVAL SOUTH INDIA THE NATURE OF SOUTH INDIAN STATES The segmentary state, according to Southall and Stein ADMINISTRATIVE STRUCTURES RURAL SOCIETY The history of a Karnataka village AGRICULTURE AND IRRIGATION Irrigation devices in early medieval Tamil Nadu Betel leaves and areca nuts

URBAN PROCESSES Weavers and weaving in early medieval Tamil Nadu TRADE AND TRADERS Aihole and the Ayyavole THE RELIGIOUS SPHERE BUDDHISM IN EARLY MEDIEVAL INDIA A letter from Xuanzang to Prajnadeva MAJOR CENTRES OF JAINISM SHANKARA AND ADVAITA VEDANTA THE HINDU CULTS Vishnuism and Shivaism The Shakti cult The Goddess as killer of the demon Mahisha SOUTH INDIAN BHAKTI: THE ALVARS AND NAYANMARS Songs of the Nayanmar saint Appar Andal’s songs Karaikkal Ammaiyar—her life and songs THE PHILOSOPHICAL UNDERPINNINGS OF SOUTH INDIAN BHAKTI AND LATER DEVELOPMENTS The vachanas of Basavanna P ATRONAGE TO TEMPLES Temple women in Chola inscriptions THE ARCHITECTURE AND SCULPTURE OF EARLY MEDIEVAL INDIA THE NAGARA, DRAVIDA, AND VESARA STYLES OF TEMPLE ARCHITECTURE WESTERN INDIA AND THE DECCAN The discovery of an early medieval quarry site near Pattadakal THE P ALLAVA KINGDOM THE CHOLA TEMPLES CHOLA METAL SCULPTURE Archaeometric analysis of Nataraja images

CONCLUSIONS A Note on Diacritics Glossary Further Readings References Credits

What the reviewers say about this book ...

Professor Singh seems to have given us a singularly learned, well-written, and detailed introduction to the study of ancient India... It is possible to have disagreement with Professor Singh on various issues, but that, in fact, lends charm to what she writes because what would a book like this be worth unless it can generate debates in the class-room? —Dilip K. Chakrabarti, University of Cambridge

[The book’s] unusual format consists of not only a narrative text, but boxed information from original sources and research works, and on key concepts, which the students will find instructive.... The website for further references and reading makes a supplement to the narrative. The list of further readings is impressive.... Due attention is paid to regional histories, especially South India and sources in regional languages.... —R. Champakalakshmi, The Hindu , 13 October 2008

... a well illustrated, marvellously produced textbook covering the vast history from the Stone Age to the 12th century.... Singularly impressive for its make-up and appearance, this textbook is the first of its kind in the country....

Each chapter of the book contains a critical reappraisal of sources and the development of historical knowledge ... helping students understand the rigorous methodology that underlies the process.... ‘[U]nsettled’ issues have been dealt with through the debates without losing their complexity and thus creating awareness of various scholars’ valuable contributions towards the construction of historical knowledge....

Singh’s book ... educates its readers as to how history can stake claims on various areas of knowledge in the domain of interdisciplinary studies like gender studies, environmental history, human geography, landscape archaeology and human ecology. —Rajan Gurukkal, The Book Review , October 2008

Professor Singh has succeeded in her venture of producing a balanced and stimulating textbook on the early Indian past. She has followed recent trends in historiography, incorporating into her book new theoretical perspectives, scientific technologies, and the enormously growing archaeological data. Often neglected South Indian history is also adequately represented. —N. Karashima, University of Tokyo

With its in-depth assessment of the literary and archaeological sources and theoretical discourses, [this book] provides a unique and long overdue introduction to the study of Indian history to the 12th century, which gives full coverage also to peninsular India. —Hermann Kulke, University of Kiel

This is the first work on ancient India where the text has been constructed at different levels. Ten chapters pan across the whole canvas, from prehistory and protohistory to ancient and early medieval history. The panorama is interspersed with inset capsules where some themes are picked out to illustrate larger elements in the chapters....

Singh’s deep affection for all kinds of ancient Indians has ensured that ... she does not lose sight of ordinary people, or for that matter, their eating habits, or even their pets.

Early India is not merely humanised through such capsules and sources, it is also illuminated by the roughly 450 illustrations that accompany the text. —Nayanjot Lahiri, India Today , 11 August 2008

[T]his up-to-date, lavishly illustrated, and thoughtfully-designed volume is clearly the new standard against which future texts will be measured.... Singh’s overview of early Indian history deftly integrate[s] archaeological data in a way few, if any, other reviews have achieved or ventured....

[Singh] stresses the complexity and diversity of experience ... while also crafting a composite image, a mosaic, of a unified Indian past. That she is able to do justice to regional specificity, occupational diversity, and cultural complexity is a testament to [her] powerful historical vision....

The most enduring value of Upinder Singh’s new synthesis is the way in which it aims to create not simply consumers but producers of historical thought. —Kathleen D. Morrison, Seminar , 593, January 2009

Singh ... writes with a refreshing openness, and her constant aim is to communicate clearly, without simplifying the complex subject matter before her. [T]his is the major contribution of the book....

In an era when most historians are torn between different and contending theories, Singh remains rooted to facts and analysis without ever committing the error of claiming that she has said the last word on the subject. —Rudrangshu Mukherjee, The Telegraph , 14 November 2008

... a fascinating and up-to-date account of South Asia’s past, from the dim beginnings of the hunterforager way of life to the early medieval period. It is based on an objective assessment of both literary and archaeological sources ... the book will be useful to students of history and archaeology at all levels and to all educated laymen who desire to know about South Asia’s past. —K. Paddaya, Deccan College, Pune

The language is refreshingly gender-sensitive and direct. The visuals are chosen with care and several of them are spectacular. Access to primary sources (both visual and textual) enriches the book enormously. It is more than apparent that the author has carefully deliberated over each sentence in order to create a text that is comprehensive. —Kumkum Roy, IIC Quarterly , Autumn 2008

The Author

Upinder Singh is Professor in the Department of History at the University of Delhi. She studied history at St. Stephen’s College, Delhi, and went on to receive her M.A. and M.Phil. from the University of Delhi, specializing in ancient Indian history. She obtained her Ph.D. from McGill University, Montreal.

She taught history at St. Stephen’s College, Delhi, from 1981 until 2004, after which she joined the faculty of the Department of History at the University of Delhi. Professor Singh’s wide range of research interests and expertise include the analysis of ancient and early medieval inscriptions, social and economic history, religious institutions and patronage, the history of archaeology, and the modern history of ancient monuments. Her research papers have been published in various national and international journals. She is the author of several books—Kings, Brāhmanas, and Temples in

Orissa: An Epigraphic Study (AD 300–1147) (1994); Ancient Delhi (1999; 2nd edn., 2006); a book for children, Mysteries of the Past: Archaeological Sites in India (2002); The Discovery of Ancient India: Early Archaeologists and the Beginnings of Archaeology (2004); and Delhi: Ancient History (edited, 2006).

Professor Singh lives and teaches in Delhi. She is married and has two sons.

Photographs, Maps, and Figures PHOTOGRAPHS

The ruins at Bhita R. C. Majumdar D. D. Kosambi A 12th century manuscript of the Prajnaparamita Purana Qila excavations in progress, 1954 The mound of Hastinapura Marine archaeologist at work Ancient ship anchor, Bangaran Island Harappan carnelian beads Nagarjunakonda salvage operations in progress J. F. Fleet D. C. Sircar A Pala period image with a donative inscription A copper plate inscription Hero stone from Khanapur, Karnataka Ratti seeds Silver punch-marked coin of Magadha Uninscribed cast copper coin of Kaushambi Silver coin of Indo-Greek king Demetrius Gold coin of Kushana king Vima Kadphises Gold coin of Gupta king Kumaragupta I Silver Gurjara-Pratihara coin Copper Pallava coin Cowrie shells Re-struck silver coin of Nahapana Arun Sonakia The Bhimbetka rock shelters Quartzite handaxe from the Narmada valley H. D. Sankalia Lower palaeolithic tools from Attirampakkam Borer from Nellor district Middle palaeolithic scraper, Attirampakkam Upper palaeolithic chert blades, Narmada valley Burin from Mukat Manipur Microliths from various sites A pot from Nal, Baluchistan Neolithic stone tools, Burzahom Bone tools, Burzahom Bone arrowhead, Burzahom

Perforated harvester, Burzahom Burnished globular jar with long neck, Burzahom Decorated stone harvester, Gufkral Celts from Nayapur and Kuchai Shouldered celt, Kuchai Female figurine, Mehrgarh View of Mohenjodaro (Sindh, Pakistan) John Marshall Rakhaldas Banerji Daya Ram Sahni Painted designs on early Harappan pottery, Nal and Kulli Early Harappan pottery, Zangian and Shahi Tump Well flanked by house walls, Mohenjodaro Main street, Mohenjodaro Narrow lane between house walls, Mohenjodaro Great Bath, Mohenjodaro Main street and house walls, Kalibangan Eastern gate, Banawali Cross-section of defence wall, Banawali Apsidal structure, Banawali Well and drains, Lothal Lothal dockyard Tank and northern gate, Dholavira Eastern gate, Dholavira citadel Well and massive drain, Dholavira citadel Miniature perforated pot, Dholavira Beaker, Dhalovira Pot with pointed base, Dholavira Ring stand, Dholavira Pottery designs Terracotta human and animal figurines Terracotta mask Terracotta circular and triangular cakes Chert blades Stone gamesmen Copper arrowhead and celt Stone sealing and seal The ‘dancing girl’ Shell ladle, Lothal Jewellery and beads Stone weights, Dholavira Terracotta cart, Harappa Harappan seals Female figurine with fan-shaped headdress Female figurine, Banawali The ‘Pashupati seal’ Harappan seals with depictions of tiger and elephant Terracotta figurine Terracotta games and dice Terracotta perforated bird-shaped rattle Terracotta bull with moveable head

Terracotta cart A ‘unicorn’ seal Megalithic burial, Hire-Benkal Pottery from late Harappan levels, Bhorgarh Copper harpoons from Shishupalgarh and Hastinapura Marine archaeologists, Dwarka Diver measuring submerged structure, Dwarka Circular stone structure in the inter-tidal zone, Dwarka Bone knife, Daimabad Pottery from different phases, Daimabad Daimabad bronzes Inamgaon artefacts Pottery from different periods, Prakash Period III (late Jorwe) pottery, Inamgaon Period III (late Jorwe) terracotta figurine, Inamgaon Neolithic celt, Brahmagiri Pottery from different periods, Maski Topikal, Cochin Sarcophagus in dolmenoid cist, Sanur PGW sherds from Hastinapura and Ahichchhatra PGW sherds from various sites Chamber tomb with port hole, Brahmagiri Close-up of chamber, Brahmagiri Megalithic cist, Brahmagiri Silver punch-marked coins NBPW from various sites Silver punch-marked coins of Kashi, Kosala, and Magadha Gandhara punch-marked coin Alexander Cunningham Panel showing Ajatashatru’s visit to the Buddha Excavated section, Hastinapura Pottery of different periods, Hastinapura Excavated eastern fortifications, Kaushambi Excavated monasteries and mound, Shravasti Excavations in progress, Piprahwa Relic casket, Piprahwa Pottery of different periods, Ahichchhatra Excavated brick structures, Ujjain The lion capital of Ashoka’s Sarnath pillar Inscription on Delhi–Topra pillar Rocks bearing the Bahapur/Srinivasapuri edict The Delhi–Meerut pillar Stone portrait of Ashoka at Kanaganahalli Ring wells and storage jar, Purana Qila The Bhita mound Panoramic view of Kaushambi The Rummindei pillar incription The Vaishali pillar Sarnath capital The Delhi–Topra pillar Elephant capital, Sankissa

Bull capital, Rampurva Dhauli elephant Façade of Lomash Rishi cave Stupa no.1, Sanchi Stone sculpture, Lohaniganj The Parkham yaksha Carved ring stones Red sandstone yakshi, Sanghol Kanishka image from Mat, Mathura Copper coins of Yaudheyas, Ayodhya, and Kunindas The Besnagar pillar inscription of Heliodorus Coins of the Indo-Greeks Silver coin of Appollodotus I Gold coin of Huvishka Copper coin of Soter Megas Gold coin of Kanishka III Copper coin of the Yaudheyas Local coin of Ujjain Coin of Nahapana Silver coin of Rudrasimha I Copper coin of Vasishthiputra Pulumavi Copper coin, Satavahana dynasty Copper coins of Satakarni I Punch-marked coins from Andhra and Pandya country Uninscribed copper coins of Cheras, Cholas, and Pandyas Walls of different periods, Purana Qila Terracotta plaque, Purana Qila Stamped and incised pot-sherds, Purana Qila Anthropomorphic pot, Purana Qila Red spouted vessel and sprinkler, Sarnath Panchachuda Chandraketugarh terracottas Yaksha Rishyashringa Sandstone Nagaraja from Chhargaon, Mathura Terracotta figurine, Mathura Terracotta tank Winged creatures worshipping linga, Mathura Nagarjunakonda reliefs Debala Mitra Yakshi on pillar, Bharhut Bharhut railing medallion Sanchi Stupa no.1, gateway and railing details Buddha image, Nagarjunakonda Remains of stupa with ayaka pillars, Nagarjunakonda Stadium, Nagarjunakonda Stupa with spoked-wheel plan, Nagarjunakonda ‘Scythian figure’, Nagarjunakonda Maya’s dream, Amaravati The Buddha’s birth, Nagarjunakonda The Buddha’s birth, Gandhara School Great Departure, Nagarjunakonda First sermon, Nagarjunakonda Ornamented stupa , Nagarjunakonda

Chaitya halls: Karle, Bedsa, Kanheri Chaitya hall entrance, Bhaja View of Bhaja caves Nashik Cave 18 Udayagiri–Khandagiri, Cave 1, Ranigumpha Verandah of Cave 10, Udayagiri–Khandagiri Gandhara head Buddha, Gandhara style Standing figure, Gandhara style Fasting Siddhartha, Gandhara school Buddha image from Govind Nagar, Mathura Nagaraja, Mathura Seated Tirthankara , Kankali Tila, Mathura Surya, Kankali tila, Mathura Karttikeya, Kankali Tila, Mathura Terracotta female figurine, Mathura Terracottas plaques, Chandraketugarh A tiger striding out of a Bandhogarh cave Flautists, Mahajanaka Jataka , Ajanta, Cave 1 Copper plates found in a pot Copper plate seals ‘King and queen type’ coin of Chandragupta I ‘Tiger slayer type’ coin, Samudragupta Brahmi script, Allahabad prashasti ‘Ashvamedha type’ coin, Samudragupta ‘Lyrist type’ coin, Samudragupta ‘Lion slayer type’ coin, Chandragupta II ‘Archer type’ coin, Kumaragupta I A set of copper plates, with ring and seal Sarnath: ‘KushanaGupta’ red ware pot, bowl, and lids Hari-Hara in the Badami Caves Krishna Govardhana, Varanasi Vishnu resting on Sheshanaga, Deogarh Gaja-Lakshmi Ekamukhalinga, Khoh (MP) Mahadeva in the Elephanta Cave Buddha, Kanheri Buddha and bodhisattva figures, Cave 2, Kanheri Colossal Buddha, verandah of Cave 3, Kanheri Bodhisattva, Nalanda View of structures, Nalanda Corner of stupa, Nalanda Tirthankara, Kankali Tila, Mathura Dashavatara temple, Deogarh Bhumara temple Nachna-Kuthara temple Lakshmana temple, Sirpur Bhitargaon brick temple Detail of doorway, Nachna-Kuthara The Ajanta caves Cave 19 façade, Ajanta

Cave 19 interior, Ajanta Buddha figures, Ajanta Ajanta paintings Buddha head, Mathura View of Udayagiri caves Udayagiri relief Buddha in the dharmachakra pravartana mudra, Sarnath Standing Buddha, Sarnath Buddha figures on stone slab, Sarnath Dancer and musicians, Aurangabad cave Stucco head from Taxila Terracotta images of Ganga and Yamuna, Ahichchhatra Bronze image of Manikkavachakar Detail of Papanatha temple, Pattadakal Brahmi script, Aihole inscription Hero stone, Karnataka Copper coin, Pallava dynasty Gold coin of Chola king Kulottunga I Gold coin of Rajendra Chola Gold coin of Rajaraja Chola Silver coin of Gurjara-Pratihara king Bhoja I Silver Gurjara-Pratihara coin Debased gold coin of Chandella king, Madanavarma The Anangpur dam Suraj Kund reservoir Billon coin of Chahamana king, Prithviraja II Coin of Shahi king Spalapatideva The stupa at Borobudur, Java The 12th century Vishnu temple, Angkor Vat View of temple and relief scenes, Cambodia Spiti valley key monastery, Spiti valley Tabo monastery, Spiti valley Clay statues in assembly hall Painting of shrine, Alchi, Ladakh Tara, Alchi Gommateshvara at Shravana Belagola Details of the Dilwara temple, Mount Abu Shiva with Nandi bull, Aihole Varaha lifting Prithvi, Aihole ‘Durga’ temple, Aihole Varaha sculpture from Lalitapur Yogini sculpture, Chaunsat Yogini temple, Bheraghat Yogini temple, Dudhai, Lalitpur Chaunsat Yogini temple, Khajuraho Sapta-Matrika sculpture Mahishasuramardini from various sites Bronze image of Manikkavachakar Lingaraja temple, Bhubaneshwar Jagannatha temple, Puri Nagara style shikhara, Lingaraja temple, Bhubaneshwar Dravida style shikhara, Brihadishvara temple, Tanjavur

Khajuraho temple Trefoiled arches of the Martanda temple, Kashmir Kailashanatha temple, Ellora Ravana lifting mount Kailasha, Ellora Ornamental pillar, Ellora Jaina tirthankara, Ellora Goddess Ganga, Ellora Cave interior and shrine, Ellora Manushi Buddhas, Teen Thal cave, Ellora Cave exterior, ceiling bracket mithuna figures, Badami Cave interior and dancing Shiva, Badami Virupaksha temple, Pattadakal Entrance, Papanatha temple, Pattadakal Gaja-Lakshmi, Papanatha temple Rama, Sita, and Lakshmana, Papanatha temple View, Papanatha temple Pattadakal quarry site Eastern entrance, Hoysaleshvara temple, Halebid Ornamental pillar, Hoysaleshvara temple Nandi, Hoysaleshvara temple Ganesha, Hoysaleshvara temple Huntress, Keshava temple, Belur Shiva and Parvati, Keshava temple Ravana lifting Kailasha, Keshava temple Hanumana, Keshava temple Vishnu resting on Sheshanaga, Mamallapuram cave Mamallapuram temple details Mamallapuram rathas Shore temple, Mamallapuram Brihadishvara temple, Tanjavur Relief panels, Brihadishvara temple, Tanjavur Chola Nataraja bronze

1 The physical geography of the Indian subcontinent 2.1 Early hominid remains 2.2 Hominid discoveries in the subcontinent 2.3 Major palaeolithic sites 2.4 Some early mesolithic sites 3.1 Centres of agriculture 3.2 Early village settlements in the north-west 3.3 Early centres of agriculture in the subcontinent 3.4 Three major chalcolithic sites of Rajasthan 3.5 Ahar culture sites, Rajasthan 3.6 Village settlements in the middle Ganga plain 3.7 Some important neolithic sites in South India 4.1 Distribution of major Harappan sites 4.2 Some early Harappan sites 4.3 Harappan routes of internal trade 4.4 Long-distance trade routes

MAPS

5.1 Major neolithic–chalcolithic sites in the Indian subcontinent 5.2 Ochre Coloured Pottery sites 5.3 Copper hoard sites 5.4 Major chalcolithic sites in Malwa and the Deccan 5.5 Some neolithic–chalcolithic settlements in South India 5.6 Early finds of iron in the subcontinent 5.7 Some Painted Grey Ware sites 6.1 The 16 mahajanapadas 6.2 Some early historical cities of north and central India 6.3 Major trade routes of early historical India 7.1 Find-spots of Ashokan inscriptions 8.1 Dynasties of India and central Asia, c. 200 BCE 300 CE 8.2 Tamil–Brahmi and early Vatteluttu inscriptions 8.3 Cities of early historical South India 8.4 Major routes connecting Asia, Europe, and Africa 8.5 India and Southeast Asia 8.6 Distribution of Roman coins in India 8.7 The Erythraean sea, according to the Periplus 8.8 Early historical monasteries in Andhra Pradesh 8.9 Buddhist caves in the Western Ghats 9.1 The kingdoms of the Guptas, Vakatakas, and some contemporary dynasties 9.2 Important ports in Indian Ocean trade networks, c. 300–600 CE 9.3 Faxian’s route 10.1 Major dynasties of peninsular India, c. 700–1300 10.2 Some dynasties of India, c. 550–700 CE 10.3 Xuanzang’s route 10.4 Major dynasties of northern, central, and eastern India, c. 700–1100 CE 10.5 Urban centres in Tamil Nadu, c. 1000 CE 10.6 Ports and cities in Indian Ocean trade networks, c. 600–1500

1.1 1.2 1.3 1.4 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 3.1 3.2 3.3 3.4 3.5 3.6

FIGURES The period of composition of some important ancient Indian texts Languages spoken in India today Pots from Gundiyali and Lodai Kharoshthi and Brahmi scripts Skull structure of gorilla, homo erectus , homo sapiens sapiens The percussion technique of making flakes Lower palaeolithic tools Isampur tools Preparation of a Levallois flake Middle palaeolithic tools Upper palaeolithic tools Decorated ostrich eggshell objects Microliths The evolution of maize from the wild grass teosinte A flotation apparatus Burial with grave goods, Mehrgarh, Period I Nal pottery Kulli pottery from Nindowari Burzahom pottery

3.7 Hunting scene engraved on stone, Burzahom 4.1 Amri pottery 4.2 Kot Dijian pottery from various sites 4.3 Painted motifs on pre-Harappan pottery, Kalibangan 4.4 Horned deity on terracotta cake and pot, Kalibangan, Period I 4.5 Citadel and lower town, Mohenjodaro 4.6 Citadel and adjacent area, Harappa 4.7 Plan of Dholavira 4.8 Harappan pottery 5.1 Diagram of sacrificial arena 5.2 Designs on Cemetery-H pots 5.3 Gandhara grave culture burial, Loebanr 5.4 Ochre Coloured Pottery pottery from Ambakheri 5.5 Copper hoard objects 5.6 Inamgaon figurines 5.7 Different types of megalithic monuments 5.8 Black and Red Ware from megalithic sites in the Deccan and South India 5.9 Painted Grey Ware pottery 6.1 Northern Black Polished Ware 7.1 Some symbols on Magadhan punch-marked coins 7.2 Schematic plan of a fortified city based on the Arthashastra 8.1 Sirkap: plan of the great stupa-temple and neighbouring block; stone masonry of different periods 8.2 Reconstruction of the Vidisha temple; Naga temple and its southern gate, Sonkh 8.3 Plan of the Ashtabhujasvamin temple, Nagarjunakonda 8.4 Plan of monastic complex, Takht-i-bahi 8.5 Plan of Sanchi Stupa no. 1 8.6 Plan of a stupa-monastery complex, Nagarjunakonda; Thotlakonda monastery 8.7 Evolution of Buddhist chaitya architecture 9.1 Buddhist complex, Pallavaneswaram, Kaveripattinam 10.1 Plan of Keshava temple, Belur 10.2 Plans of Shiva temple at Narttamalai; Brahmapureshvara temple at Pullamangai; Nageshvarasvami temple, Kumbakonam 10.3 Plan of Brihadishvara temple, Tanjavur

Preface

From 1981, I spent over twenty years teaching the undergraduate course on ancient and early medieval India at St. Stephen’s College, Delhi. It was a daunting course, demanding coverage of many different areas and issues over enormous spans of time. I was fortunate to have students with sharp and inquisitive minds, whose questions constantly forced me to re-think my perspectives and conclusions, and who made me realize that teaching is ultimately about the quality of communication between student and teacher. Undergraduate teaching, with its enormous pressures of teaching and marking work, left very little time for research. Nevertheless, I did manage to keep my research going, and explored issues related to social and economic history, religious institutions, inscriptions, archaeology, and the modern histories of ancient sites and monuments. A History of Ancient and Early Medieval India: From the Stone Age to the 12th Century emerged from the intersection of my experiences as a teacher and researcher. Primarily a textbook and reference work for both undergraduate and postgraduate students, this book will, I hope, also appeal to the general reader. Its aim is to provide an introduction to ancient and early medieval India through a comprehensive overview of historical issues and details within a firm chronological framework; explanations of basic concepts and terminology; an exposure to the flavour of textual, material, and visual historical sources; and a highlighting of new discoveries and research. Perhaps most importantly, this book focuses on the process through which historical knowledge is formed, and the intellectual inquiry and debate that form part of this process. This book is not a mere summary of existing knowledge. Rather than offer students a smoothened narrative, which they will then be expected to absorb passively, it is necessary to expose them to the complex details and textures of history. Where there are unresolved issues, they have been presented as such, rather than conveying a false sense of certainty. Where there are debates, the different perspectives have been presented, along with my own assessment of which arguments are convincing and which ones are not. Historians and teachers invest far too much time and energy in telling students what to think, rather than how to think for themselves. Students need to learn to evaluate evidence and hypotheses, to relentlessly question and critique what they read or are told, and formulate and express their independent views. It is essential to acknowledge the valuable contributions made by various scholars towards the construction of historical knowledge and to understand the rigorous methodology that underlies this process. However I hope that this book encourages readers to think courageously and creatively beyond the current boundaries of academic discourse and debate. Since this is a macro-history of the Indian subcontinent, and in a single volume at that, it outlines broad trajectories, always aware of the fact that these are only a few of multiple trajectories. Thus, for instance, while the account of the beginnings of food production may suggest that this was the inexorable direction in which things were moving, emphasis is still placed on the fact that hunting

and gathering remained a preferred subsistence activity for many communities across the centuries. Similarly, the discussion of the early historical period may seem to suggest that everything was making way for the emergence of city life, but it must not be forgotten that most people of the subcontinent continued to live in villages. The privileging of certain processes over others is partly the result of the training and tendency of a historian to focus on what appear to be significant changes, and also due to the inherent nature and inadequacies of sources and available data. The fact is that whether we look at the archaeological or literary sources, we know much more about agricultural groups than hunter-gatherers, and much more about city-dwellers than village folk. Nevertheless, it is important to constantly remind ourselves about the partial and inadequate nature of our historical narratives. Prehistory to c. 1200 CE is an enormous span of time, and it is not possible to be exhaustive on each and every issue. The structure of this book involves breaking this vast period into broad chronological units. For earlier periods, all radiocarbon dates mentioned in this book are calibrated dates. Following current usage, BCE (Before Common Era) is used instead of BC, and CE (Common Era) instead of AD. Against the background of the controversy over the dates of the Buddha’s life, c. 480 BCE has been taken as the date of the parinibbana. Within the broad chronological units, profiles have been constructed of the various geographical regions, incorporating the range of available literary and archaeological evidence, bringing out the complex strands of historical processes within and across different regions. The coverage of regions is necessarily dependent on available information, and the gaps and inadequacies in this information should inspire young scholars to take on the challenge of addressing them. Each chapter looks at various aspects of a particular period on the basis of a critical survey of the available sources. The narrative is punctuated by boxes focusing on key concepts, primary sources, further discussion of specific issues or details, recent discoveries, and new directions in research. From the beginning of the historical period, the chapters start with a synopsis of political history and a discussion of political processes. This is not because these are necessarily the most important aspects of history, but because it is useful for students to have a basic understanding of political context and chronology. Political narrative has been accompanied, to every possible extent, with a discussion of political structures and processes. Political, social, economic, religious, and cultural history are discussed sequentially in order to bring out their inter-connectedness within a chronological and contextual frame. The discussion of social history looks at issues such as class, caste, gender, and subordinate and marginalized groups. Philosophical ideas are treated as an important part of the intellectual life of different periods. Religious doctrines and practices are discussed as important areas requiring detailed investigation, and not merely as part of an ideology reflecting existing power structures. I hope that the many excerpts from original sources and photographs create sensitivity towards the aesthetic dimensions of Indian cultural traditions reflected in literature, art, and architecture. As far as possible, references have been cited to enable the interested reader to go to the original source. Translations have often been slightly modified to make them more accessible. Punctuation has been altered to suit the style of the book, especially since diacritical marks have been dispensed with. Since historical literature generally uses such diacritics and students should understand them, the conventionally used systems of transliteration for Sanskrit and Tamil have been provided towards the end of the book.

It is a matter of great satisfaction for me that this book contains over 400 illustrations—line drawings, photographs, and maps—many of a quality and range that are not to be found in any book on ancient and early medieval India. The visual element is as important for understanding prehistoric stone tools as for appreciating art and architecture. The illustrations are much more than an adjunct or supplement to the text. In many cases they convey much more than words possibly can, illuminating the past and making it vivid, meaningful, and exciting. In spite of my best effort, I am aware that this book has certain limitations. For instance, largely because the book was already very long, the last chapter does not discuss the Delhi Sultanate or the history of Islam in the subcontinent, which are very important parts of the early medieval period. For similar reasons, the rich and varied cultural developments of this period could not be surveyed exhaustively. I have instead given a brief overview, with a focus on South India, hoping that the photographs will to some extent make up for the lack of detailed discussion. This book provides students and scholars with a foundation, encouraging them to pursue further reading, depending on their needs and interests. The historical narrative given in the book relies not only on my own research but also on a vast array of writing and research produced by others. My debt to this scholarship is acknowledged in the in-text references and the readings suggested at the end of the book. Readers are encouraged to follow these references for more detailed treatment of various issues. The Web supplement carries forward the features of this book, especially in terms of excerpts from original sources and illustrations. This resource allows a reader access to constant additions and updates to the material. This open-endedness is essential, given the fact that new data and changes in perspective are an integral part of the discipline of history. I hope that this book communicates how exciting and challenging an exploration of the history of ancient and early medieval India can be. My students, initially at St. Stephen’s College, and subsequently in the History Department of the University of Delhi, have been an important part of my own exploration of this history. That is why this book is dedicated to them. Upinder Singh

A Reader’s Guide to A History of Ancient and Early Medieval India

A first of its kind in India, this book has been developed and designed as a textbook for students of ancient Indian history. It brings together an exhaustive coverage of a large span of India’s ancient past in a lucid narrative style. Pedagogic elements built into the book make the study of history a thought provoking and enjoyable experience.

In order to help you make the best use of this book, this section provides a window into the various components of the text.

Each chapter constitutes a chronological unit within a larger framework, providing a comprehensive overview of historical issues and details, and constructing profiles of the various geographical regions in the subcontinent. The chapter outline provides a view of the broad organization of the chapter. An opening story from a variety of sources serves as an engaging start for the chapter and also presents a strand from the rich thematic core of the chapter’s discussion. Boxes

Five kinds of boxes appear throughout the book. Each kind has a separate role in helping you explore and understand different dimensions and key issues related to history learning and teaching.

Numerous important concepts and terms used by historians (sometimes drawn from different disciplines) such as state, tribe, class, and caste are explained with their specific and complex meaning in KEY CONCEPTS . This helps in using these terms and concepts with greater clarity and appropriateness, and in gaining a better idea of the inherent interdisciplinary nature of history. KEY CONCEPTS

Lineage, clan, tribe

Historians use several sociological terms and concepts while describing ancient cultures. Kinship refers to socially and culturally recognized relationships among people, commonly assumed to be based on natural or biological ties. These ties may be based on birth/descent (conlineal or agnatic. Unilineal kinship systems which recognize descent through the mother are known as matrilineal. Multi-lineal or cognatic systems are those in which descent through both the mother and father is recognized. In both patrilineal and matrilineal systems,

Learning about the original sources of history, and how they are interpreted, makes history truly exciting. Familiarity with primary sources is an integral part of the appreciation and evaluation of historical theories and arguments. The PRIMARY SOURCES boxes provide you with descriptions and illustrations of archaeological source material, interesting information about literary sources and their authors, and many translated excerpts from original texts and inscriptions. PRIMARY SOURCES

The analysis of ancient plant remains

The study of ancient plant remains is known as palaeobotany or archaeobotany. Botanical remains from ancient sites often include macro-botanical remains such as seeds or grains. These can get preserved through desiccation, waterlog- analysed under microscopes to determine what types of plants they represent and whether these were wild or domesticated. Plant remains can also take the form of micro-botanical remains. Tiny particles

History is full of debates on various issues. We intersperse our macro-level main narrative in the book with a more detailed look at specific issues. FURTHER DISCUSSION boxes enrich your understanding of the multi-layeredness of our past, and the need to be ready to move beyond generalities and on-the-surface narratives, to closer, more detailed investigations. FURTHER DISCUSSION

Female figurines—ordinary women or goddesses?

At one time, scholars tended to use the ‘Mother Goddess’ label for all female figurines found at sites. This was largely because of the belief that the worship of fertility goddesses was an important part of agricultural societ- In the light of such problems, the term ‘Mother Goddess’ should be replaced by the longer but more neutral phrase— ‘female figurines with likely cultic significance.’ This does not mean that none of these figurines might have had a reli-

Historical knowledge is constantly growing. New discoveries can often radically change our understanding of the past. RECENT DISCOVERIES boxes direct attention to new exciting discoveries, the people and circumstances related to these discoveries, and how these discoveries have made an impact on our understanding of India’s early past. RECENT DISCOVERIES

Isampur: a centre of stone tool manufacture

Isampur (Gulbarga district, Karnataka) is a village located in the north-western part of the Hunsgi valley, drained by a small seasonal stream known as the Kamta Halla. The palaeolithic site lies about 2 km north-west of the village, close to large flakes, and debitage (waste material). The main tool types were chopping tools, knives, handaxes, cleavers, and scrapers. While unfinished tools occurred in large numbers, there were relatively few finished ones. Hammer stones of dif-

While it is important for you to be aware of new historical research, this research is often not easily accessible. NEW DIRECTIONS IN RESEARCH boxes bridge the gap between students and researchers by presenting samples of interesting new research, and by explaining their methodology and results in a clear and straight-forward way. This exposes you to new trends in history writing, and provides a sense of the constantly changing understandings of the past. NEW DIRECTIONS IN RESEARCH

Pictures on stone

Pictures made on granite rocks can be seen in many places in Karnataka and Andhra at sites such as Kupgal, Piklihal, and Maski. They are difficult to date, but a rough chronology can be worked out on the basis of style, content, and weather- are also people standing in a chain-like formation, usually interpreted as dancers. Other less frequently occurring motifs include the elephant, tiger, deer, buffalo, birds, footprints, and abstract designs. In general, the scenes tend to be small

Maps, Photographs, and Figures

Moving ahead from dreary text-based history writing in India, Ancient and Early Medieval India has over 450 illustrations—maps, photographs, and sketches— that bring history alive. History becomes an exciting exploration when we can visually situate our learning, and appreciate the richness of our subcontinental past and culture.

Maps are one of the most important tools for a history student. Note the use of legends and captions, different colours for topographical and elevation details, a scale to give an idea of respective distances, and the use of latitude and longitude coordinates to show the location of the mapped area.

Over 350 photographs of various artefacts, such as stone tools, terracottas, pottery, and coins, excavated sites, temples, and sculptures, enliven the text.

Chert burin

Chert core

Period IV: stamped and incised red ware

SEE CHAPTER 8, P . 28 FOR MORE DETAILS ON THE MAHAYANA AND HINAYANA SCHOOLS. The first connected life story of the Buddha occurs in the Nidanakatha (1st century). The Pali or Sri Lankan chronicles—the Dipavamsa (4th-5th century) and the Mahavamsa

In order to help you follow a certain idea in detail, or to follow a topic dealt with in different chapters, cross references are provided in the margins. These are indicated by a cross reference icon with relevant page numbers.

www.pearsoned.co.in/upindersingh Photographs of Harappan sites and artefacts Allahdino is a small (1.4 ha) unfortified village site of the Harappan civilization, about 40 km east of Karachi. Houses made of mud-brick, often resting on stone foundations, were laid out in a west–south-west to east–north-east orientation. A large multi-roomed building on a large mudbrick platform in the north-eastern part of the excavated area seems to have had some special significance. Another building was associated with three wells. The wells at

A Web supplement available on www.pearsoned.co.in/upindersingh contains additional material such as extracts from original sources, photographs, and points for discussion. A Web supplement icon and a short caption indicate the supplementary material available in relation to the discussion in the text.

Diacritic marks, used extensively in academic writing, have been avoided to facilitate easy reading. However, the conventionally used systems of transliteration for Sanskrit and Tamil have been provided at the end of the book.

Further Readings for various chapters are provided towards the end of the book. They are meant for readers interested in acquiring more detailed information.

We hope that this book will prove to be an important contribution towards transforming the way ancient Indian history is taught and learnt. It is our endeavour to constantly improve this book, and we would be glad to receive suggestions from all our readers. Please write to us with your feedback to [email protected]

Introduction Ideas of the Early Indian Past Chapter outline THE MAIN PHYSIOGRAPHIC ZONES OF THE SUBCONTINENT WAYS OF DIVIDING THE INDIAN PAST CHANGING INTERPRETATIONS OF EARLY INDIAN HISTORY NEW HISTORIES, UNWRITTEN HISTORIES

THE RUINS AT B HITA

The Puranas describe a universe shaped like an egg, vertically divided into the celestial worlds, earth, and netherworlds. The earth is a flat disc, consisting of seven land masses ( varshas) arranged in concentric circles, alternating with seas of salt water, molasses, wine, butter, curd, milk, and fresh water. Situated in the centre of the earth is Jambudvipa, in whose southernmost part lies Bharatavarsha, the golden Meru mountain rising from its midst. One of several explanations of the name Bharatavarsha connects it with the Bharata people, descendants of the legendary king Bharata, son of Dushyanta and Shakuntala. Cosmography blends with geography in the Puranas. Bharatavarsha is said to consist of nine divisions (khandas), separated from one another by seas. But the mention of its mountains, rivers, and places—some of which can be identified—suggests that the composers of such texts were familiar with various areas of the Indian subcontinent, and perceived them as parts of a larger cultural whole.

For people of other lands, the major subcontinental landmark was the Indus, or Sindhu, the mighty river that originates in the Tibetan plateau, flowing 3,200 km south-west across fertile plains before it merges with the Arabian Sea. The words ‘India’, ‘Hindu’, and ‘Hindustan’ originate from the name of this river. Ancient Chinese sources refer to the land of ‘Shen-tu’, Greek texts mention ‘India’, and Persian inscriptions describe ‘Hidu’ as one of the subject countries of the Achaemenid king Darius. These terms initially referred only to the lower Indus valley, but their connotations expanded swiftly. For Megasthenes, who visited the court of Chandragupta Maurya in the 4th century BCE, ‘India’ meant the entire subcontinent. Many centuries later, Arabic and Persian texts used the word ‘Hindustan’ for this vast stretch of land and ‘Hindu’ for its inhabitants. While the idea of the Indian subcontinent forming a distinct geographic and cultural unit is a very old one, its nation-states—India, Pakistan, Nepal, Bhutan, Bangladesh, and Sri Lanka—emerged only in recent times. When exploring the ancient history of South Asia, it is necessary to ignore modern political boundaries and to treat the Indian subcontinent and its many regions and sub-regions as a single canvas. The history of the subcontinent is really about the historical trajectories and interactions of these regions and sub-regions, which at certain points of time—during the peak of the Maurya, Mughal, and British empires—attained some measure of political unity. The Main Physiographic Zones of the Subcontinent

The Indian subcontinent has fairly well-defined geographical frontiers but enormous ecological diversity. Its climatic patterns are similar to those prevailing in other areas on the same latitude but are significantly modified by the Himalayas and the Western Ghats. The Himalayas block the icy northern winds from sweeping across the Indo-Gangetic plains in winter as well as the rain-laden monsoon winds from the south-west in summer. The barrier of the Western Ghats similarly leads to rainfall in the western coastal strip. Most of the subcontinent gets its rains from the south-west monsoon, except for the north-west and Sri Lanka, which rely on winter rains. In the north, the subcontinent is bordered by the Himalayas, fairly young fold mountains. The process of their uplift and folding is still going on, making them geologically unstable. The Himalayas can be divided into the western, central, and eastern zones, each with their own specific characteristics. The north-western part of the subcontinent includes the arid mountainous North-West Frontier Province and the Baluchistan province of contemporary Pakistan. Leaving aside the fertile

river valleys, this area is not especially suited for agriculture, but the many routes running along its valleys and passes connect the subcontinent with areas lying to its west. Even more arid conditions prevail in the Thar desert of Rajasthan, where low hills and sand dunes rise over the underlying low, rocky plateau. Between the desert and the north-western mountains lies the Sindh province of southern Pakistan, the Indus providing precious water in an area of very low rainfall. The northern course of this river lies in Tibet and Ladakh, and along with its tributaries, it flows through the fertile plains of Indian and Pakistani Punjab. To the east of the Indus is the shrivelled course of a once mighty river, the Ghaggar-Hakra. The fertile northern alluvial plain of the Ganga and its tributaries is another major geographical zone of the subcontinent. The western part of this plain is known as the doab (literally, ‘the land between two rivers’, the Ganga and Yamuna). The middle part of the plains corresponds roughly to the state of Bihar and the eastern part of the state of Uttar Pradesh in modern India. The eastern part includes the delta of the Ganga and Brahmaputra, comprising of modern West Bengal, Assam, and Bangladesh. The Vindhyan ranges separate the northern plains from peninsular India, while the Aravalli hills divide the Thar desert from central India. The Malwa plateau, with its two major rivers, the Narmada and Tapi, lies between the Aravallis and the central Indian mountains. Peninsular India is an old and relatively stable geological formation, its landscape marked by plateaux, plains, and the fertile valleys of rivers such as the Mahanadi, Krishna, Godavari, Pennar, and Kaveri. The Deccan plateau, formed by the lava flows from very ancient volcanoes, constitutes the dominant part of the peninsula. It is bordered by the Eastern and Western Ghats, beyond which are the narrow Coromandal and Malabar–Konkan coastal plains. The Nilgiri, Annamalai, and Cardamom hills lie in the extreme south of the peninsula, which is separated from the island of Sri Lanka by the Mannar strait. The various geographical zones of the subcontinent have never been isolated units. From very early times, human interaction took place through routes cutting across mountains, rivers, and regions, dictated by geographical features and human needs. The Himalayas could be crossed at points such as the Bolan, Gomal, and Khyber passes, and a network of overland routes connected the subcontinent to China, central Asia, West Asia, and Europe. There was also the over 7,500 km long subcontinental coastline, home to numerous fishing and sailing communities from times immemorial, which linked the subcontinent to the larger Indian Ocean world and to areas such as Southeast Asia and the Persian Gulf. The natural landscape has always been an important part of human life, and has affected and influenced people’s thought and action in many ways. The topography, climate, soil, and natural resources of any land influence modes of subsistence, settlement patterns, population density, and trade. Humans have in turn transformed the environment in many ways. Situating the human past in its specific environmental context helps us understand the different rhythms and patterns of cultural development and interactions in the various regions. However, as we will see further on, ecology too has a history and the subcontinental environments of today differ in many respects from those of the past. Ways of Dividing the Indian Past

The English word ‘history’ comes from the Greek historia (inquiry or investigation). History is essentially a discipline that inquires into the experiences of people who lived in the past. Historians often classify the past by dividing it into different periods. Labels are convenient, but they should be meaningful and consistent, and it is necessary to be aware of their limitations.

MAP 1 THE PHYSICAL GEOGRAPHY OF THE INDIAN SUBCONTINENT

For a long time, historians divided Indian history into the Hindu, Muslim, and British periods. However, this classification is flawed and can be questioned on several grounds. For example, is the religious affiliation of the ruling elite the best basis for labelling a period? In that case, why is the third period described as the British and not the Christian period? From when can we start using the term ‘Hindu’ in the context of ancient India. How can it be applied to the reigns of the many ancient Indian kings who patronized Buddhism or Jainism? Did the advent of Muslim rulers create a major

rupture in the fabric of Indian society, especially when the sway of these rulers—except at the height of the Mughal empire—did not extend over all or even most of the subcontinent? Due to such reasons, most historians have discarded the Hindu–Muslim–British periodization of the Indian past in favour of a more neutral classification into the ancient, early medieval, medieval, and modern periods. The dividing lines may vary, but the ancient period can be considered as stretching roughly from the earliest times to the 6th century CE; the early medieval from the 6th to the 13th centuries; the medieval from the 13th to the 18th centuries; and the modern from the 18th century to the present. The current use of these terms shifts the focus away from religious labels towards patterns of significant socio-economic changes. The ancient or earliest parts of the human past can be further divided into prehistory and history. The enormously long period before the invention of writing and the study of that period are known as prehistory. The part of the past that comes after the invention of writing, and the study of that part of the past (i.e., of literate societies) constitute what is considered history. A language consists of spoken symbols of communication. A script, or writing, is a system of visual communication using signs or symbols associated with specific meanings or sounds, written down on some surface. Human beings used languages long before they invented scripts. The cuneiform script of Mesopotamia (ancient Iraq) was invented in c. 3400 BCE and Egyptian hieroglyphics in c. 3100 BCE. In the Indian subcontinent, the earliest substantial evidence of writing is associated with the Harappan civilization and dates from c. 2600 BCE, but recent discoveries push back the origins of the script to the second half of the 4th millennium BCE. The ancient Mesopotamians pressed letters onto moist clay tablets, while the ancient Egyptians wrote on papyrus sheets made of reeds. The Harappan script is mostly found on seals and sealings. But apart from the specimens of writing that have actually survived, it can be assumed that people must have written on perishable material as well. Writing marked a new stage in human expression and communication. It opened new possibilities for storing and transmitting ideas and knowledge across distance and time. Its impact was complex and varied. Rulers used writing to advertise and exercise power, merchants to record business transactions, priests to preserve religious texts, and poets to give permanence to their creative expression. We can speculate about the precise impulses that led to the invention of writing, but all over the world (with a few exceptions) it coincided with the emergence of cities and states. For these reasons, historians consider the beginning of writing an important watershed in the story of ancient cultures. However, in a situation where relatively few people knew how to read or write, writing gave a certain power and privilege to those who knew it and denied it to those who did not. Further, the invention of writing did not mean the end of oral transmission. The spoken word has always held a special significance in many cultural traditions, and this significance continued even after manuscripts of texts came to be made. Oral versions of many written texts continued to circulate and often had a far greater outreach and impact. The beginning of writing is also an important watershed in the study of the past because written evidence becomes available to the historian. Nevertheless, it must be remembered that such evidence covers only a very small portion of the human past. The past before writing (prehistory) and the history of non-literate people who did not leave behind written sources are also extremely important and have to be recovered. And even when written sources are available, archaeological sources continue to be important for historians.

In the Indian subcontinent, the story of writing is a bit complicated. Although the Harappans were a literate people, their script has not yet been deciphered. So historians cannot use the written material they left behind to reconstruct their history. Another mystery is: what happened to writing after the decline of the Harappan civilization in c. 1900 BCE? While it is possible that people continued to write, although on perishable material, there are hardly any surviving specimens of writing between c. 1900 BCE till we come to the 4th century BCE. The oldest script in the subcontinent is the Harappan script, but the oldest deciphered script is Brahmi, known from about the 4th century BCE, and the two scripts seem to be quite different. For these reasons, it is not easy to draw the dividing line between history and prehistory in India and the term protohistory is useful. This word carries different meanings. In the European context, it is sometimes used to refer to people who did not themselves have writing, but who are mentioned in the written records of a contemporary literate group. In the Indian subcontinent, the Harappan civilization—a literate culture with an undeciphered script—is included in protohistory. This term can also include the period c. 1500–500 BCE, for which there is an orally transmitted literature (the Vedas), but no evidence of writing. Archaeologists often use the word protohistory for the long period between the beginning of food production and the advent of iron technology. This would include neolithic and chalcolithic cultures in different parts of the subcontinent. The subcontinent is a huge geographical area, and the transition to literacy did not take place everywhere at the same time. For instance, areas outside the literate Harappan zone were inhabited by non-literate people. Going by the earliest surviving samples of deciphered writing, the beginning of the historical period in north India would have to be placed in the 4th century BCE. However, it can be presumed that this writing had a history on perishable material, one that must go back to at least the 6th century BCE. Lists of historical kings and philosophers of this century are available for parts of north India. Considering all these factors, there is a good case for placing the beginning of the historical period in north India in the 6th century BCE. The evidence of 4th century BCE Brahmi inscriptions from Anuradhapura in Sri Lanka, 2nd century BCE Tamil–Brahmi inscriptions, and the political history reflected in Sangam literature suggest that the transition to the historical period in South India occurred some time between the 4th and 2nd centuries BCE. Of course, if the Harappan script is deciphered some day, the dates for the beginning of the historical period in northern India will have to be pushed back to the 3rd millennium BCE, or even earlier. Changing Interpretations of Early Indian History

The historiography (the scholarly activity of constructing and writing history) of ancient and early medieval India reveals many significant changes over time; these can be understood against the background of the political and intellectual contexts in which they emerged and flourished. The various ‘schools’ of history writing are often presented and understood in terms of one school making way for the other in a neat, forward progression. The reality is, however, much more complex. There was considerable variety within the various schools; some of them co-existed (and still do so) in dialogue or conflict with each other, and there are many examples of writings that go against the grain and do not easily fit into the dominant historiographical trends of their time.

HARAPPAN WRITING ON SEAL

EGYPTIAN HIEROGLYPHICS

MESOPOTAMIAN CUNEIFORM

The 18th and 19th centuries were dominated by the writings of European scholars, usually referred to as the Orientalists or Indologists, although they often described themselves as ‘antiquarians’. Many of them were employees of the East India Company and later, the British

Government of India. The founding of the Asiatic Society of Bengal in 1784 provided an institutional focus for scholars working in a number of related fields such as textual study, epigraphy, numismatics, and history. A major contribution of the Indologists lay in their efforts to collect, edit, and translate ancient Indian texts. In this, they depended heavily on information provided by ‘native informants’, whose contribution was rarely acknowledged. Indology soon spread beyond the confines of the British empire and became a subject of study in European universities. Apart from the study of ancient texts, the 19th century also witnessed important developments in the field of epigraphy, numismatics, archaeology, and the study of art and architecture. The decipherment of the Ashokan Brahmi and Kharoshthi scripts were major breakthroughs. The analysis of coins contributed to the construction of a framework of political history. Officers of the Geological Survey discovered prehistoric stone tools and laid the basis of Indian prehistory. The Archaeological Survey of India was established in 1871, and over the succeeding decades, this institution made an important contribution towards unearthing and analysing the material remains of India’s past. The contributions and breakthroughs of the 18th and 19th centuries were rooted in a colonial context, and this is evident in certain features of Indological writing. The Brahmanical perspective of ancient Sanskrit texts was often uncritically taken as reflecting the Indian past. Social and religious institutions and traditions were critiqued from a Western viewpoint. Indian society was presented as static and its political systems unwaveringly despotic over the centuries. Race, religion, and ethnicity were often confused with each other and there was a tendency to exaggerate the impact of foreign influence on ancient India. This is the time when the classification of the Indian past into the Hindu, Muslim, and British periods took root.

R. C. MAJUMDAR (1888–1980), A LEADING HISTORIAN OF THE NATIONALIST SCHOOL

Indian scholars of the late 19th and first half of the 20th centuries made major contributions towards constructing a connected narrative of ancient India. Writing against the background of an emergent, and later increasingly strong, national movement, these historians are generally referred to as Nationalist historians. They were responsible for meticulously weaving together data from texts, inscriptions, coins, and other material remains to amplify the contours of the ancient Indian past.

Especially important contributions were made in the field of political history. South India was brought into the narrative and the study of regional polities progressed. The nationalist tinge in the writings of these scholars can be seen in their insistence on the indigenous roots of all major cultural developments. It is also reflected in their search for golden ages, which led to their exalting the age of the Vedas and the Gupta empire. Non-monarchical polities were discovered and were celebrated to counter the idea that India had never known anything but despotic rule. The periodization of the Indian past into the Hindu, Muslim, and British periods was, however, retained. It coalesced with a communal tendency to valorize the ‘Hindu period’ and to project the advent of the Turks and Islam as a calamity and tragedy. The 1950s saw the emergence of Marxist historiography, which went on to play an extremely influential role in the construction of the history of ancient and early medieval India. In the long run, the major achievement of Marxist historians was to shift the focus from an event-centred history dominated by political narrative to the delineation of social and economic structures and processes, especially those related to class stratification and agrarian relations. Marxist historiography also contributed towards uncovering the history of non-elite groups, some of whom had suffered centuries of subordination and marginalization. While making these valuable interventions and contributions, Marxist writings often tended to work with unilinear historical models derived from Western historical and anthropological writings. Texts were sometimes read uncritically, with insufficient attention paid to their problematic chronology and peculiarities of genre. Archaeological data was included, but the basic framework of the historical narrative remained text centric. Initially, the focus on class meant less attention to other bases of social stratification such as caste and gender. Religion and culture were often sidelined or mechanically presented as reflections of socio-economic structures.

www.pearsoned.co.in/upindersingh EXCERPTS FROM ORIENTALIST, NATIONALIST, AND MARXIST WRITINGS ON ANCIENT INDIA

Despite their important differences, the major historiographical schools also shared some similarities, for instance, in their emphasis on Brahmanical Sanskrit texts and their tendency to marginalize archaeological evidence. Certain tenets of all these schools continue to thrive in the present. Some of the fundamental premises and methods of Orientalist historiography continue to hold their ground and histories of Third World countries such as India remain Eurocentric in many respects. Appeals to the ancient and early medieval past are still often dictated by nationalist or communalist agendas. Marxist historiography continues to be an influential force in early Indian historiography.

D. D. KOSAMBI (1907–66), A PIONEER OF MARXIST HISTORIOGRAPHY

A few other aspects of the large volume of historical research of the last 50 years or so can be identified and cited here. New theoretical perspectives, scientific techniques, and a continuing growth in the volume of archaeological data have been transforming our understanding of the early Indian past, especially with regard to subsistence practices, technology, and human interaction with the environment. Palaeo-environmental studies have directed attention to the changing ecology of the different regions and its impact on human life; these important issues are likely to increasingly engage the attention of scholars. Investigations of archival material have begun to reveal in unprecedented detail the complex stories of the people, institutions, and ideas involved in the construction of archaeological knowledge. Such studies also reflect the need to break the disciplinary divides between the ‘ancient’ and the ‘modern’ (and all that lies in between) by inquiring into issues such as the modern histories of ancient sites and monuments. The research of a small group of historians (mostly women) working on gender relations has radically altered the frontiers of early Indian social history. The focus on gender has involved much more than simply inserting women into history. Breaking away from the traditional ‘position of women’ mould, it has asked new questions, broken the artificial divide between the private and political domains, and revealed the power hierarchies within the family and the household. The most important achievement of this line of research is that it has demonstrated the close relationship between gender and hierarchies based on class, caste, and political power. A significant feature of recent historiography of the early medieval period is the detailed study of the changing profiles and configurations of regions and sub-regions. Based on careful empirical examination of epigraphic and textual sources, these studies have identified changes in political, economic, and social structures, with a special focus on agrarian relations and the legitimation of political power. In doing so, they have revealed the varied historical textures and trajectories in different parts of the Indian subcontinent in early medieval times. A critical understanding of historiography, one which recognizes the contributions and limitations of past and present ideological and theoretical frameworks, is essential in order to understand where the history of ancient and early medieval India stands today. However, the major advances of the

future are likely to be the result of questioning and thinking beyond the boundaries of existing historiographical positions and methodologies. New Histories, Unwritten Histories

History is not one but many stories, only a few of which have as yet been written. The challenges to build on the advances that have already been achieved so far are many. Currently, there are two parallel images of ancient South Asia—one based on literary sources, the other on archaeology. Texts and archaeology generate different sorts of historical narratives and suggest different rhythms of cultural continuity, transition, and change. Historians generally use archaeological evidence selectively as a corroborative source when it matches hypotheses based on their interpretation of texts. Archaeologists, for their part, have not adequately explored the historical implications of the available archaeological data. Correlations between literature and archaeology tend to be simplistic and devoid of careful reflection on methodology. We need to seriously consider whether, given their inherent differences, textual and archaeological evidence can be integrated, or whether we should simply aim at juxtaposition. The old tradition of extracting supposedly self-evident ‘facts’ from literary sources needs to be replaced by an approach that is more sensitive to their genre, texture, and cadence. However, in view of the information and insights offered by rapidly growing archaeological data, historical narratives can no longer afford to remain text-centric. A more sophisticated approach towards textual study has to be accompanied by a proper incorporation of archaeological evidence. This will lead towards a more nuanced image of ancient India. It will reveal the complexities and diversities of cultural processes in the various regions, and will incorporate the ordinary and everyday into our understanding of the ancient past. Histories of early India should ideally represent the various regions and communities of the subcontinent in all their diversity. However, while the heartlands of great empires and kingdoms are well represented, many regions—for instance the North-East—are not. Such regions have to be brought into the ambit of history. Bringing more people into history also requires further initiatives towards uncovering the past of groups who have been subordinated and marginalized for centuries, such as the labouring poor, lower castes, and tribal communities. This is not easy, given the fact that a great proportion of the source material available to historians has been created by elite groups and therefore reflects their ideas and interests. Nevertheless, the past of people who have been hidden from history has to be uncovered and written, and these histories must become an integral part of the narrative of the ancient Indian past. Explorations of gender, the family, and the household need to be pushed further and have to become part of larger social histories. Issues and institutions such as the family, class, varna, and jati need to be viewed from long-term perspectives, showing how the different bases of social identity intersected and changed over time. India’s varied and complex cultural traditions are also in need of urgent attention. Interestingly, while these continue to be the focus of intensive research among scholars working in South Asian studies, religious studies, and art history departments abroad, they have in recent decades remained somewhat marginal to mainstream historical writing in India. Indian historians have often tended to treat religious cults and traditions primarily as ideologies reflecting social and political power structures of the time. It must be recognized that the many different strands of religious thought and

practice are an important aspect of history in their own right and need thorough investigation. This also applies to the history of ideas and the aesthetic dimensions of the Indian past reflected in literature, art, and architecture. Our understanding of the history of the subcontinent tends to be far too insular, and much greater attention needs to be paid to its relationships with other areas, especially East Asia and Southeast Asia. Apart from examining trade networks, there is a need to try to explore and understand the complexities of the cultural transactions between the different parts of Asia. These transactions are reflected not only in textual evidence, but also in a rich and exciting storehouse of material evidence in the form of inscriptions, sculpture, and architecture. There is a close relationship between history and identity; the past has, therefore, always been a contested terrain. In contemporary India, the ancient past is invoked in different ways in political discourse, including propaganda with chauvinistic or divisive agendas. There are debates over the state’s right to project and propagate certain interpretations of the past through school textbooks. Communities frequently take offence at things written about them in historians’ scholarly writings. In such a charged and intolerant atmosphere, there are several dangers—of the deliberate manipulation and distortion of the past to achieve political ends, of historical hypotheses being judged on the basis of their political implications rather than their academic merit, and of historians being criticized for writing objective history. The need for defining and enlarging a liberal academic space which nurtures level-headed dialogue and debate has perhaps never been greater. Aside from its role in current identity politics, ancient history is often considered distant, difficult to relate to, even irrelevant to our times and concerns. However, if we look carefully enough, we will in fact find that the roots of some of the social practices, institutions, and ideas of the present lie in the remote past. But even more interesting than the things that are familiar are those that are startlingly different. The most important thing that history can do is to teach us to think historically. It can make us realize that human experiences are diverse and complex; and it can make us aware of the many entangled threads of continuity and change that connect the present to the past. No less important is the fact that the story of the past contains much that is interesting and exciting. That in itself is enough justification for reading and writing history.

Chapter One Understanding Literary and Archaeological Sources Chapter outline READING ANCIENT TEXTS FROM A HISTORICAL POINT OF VIEW ARCHAEOLOGY AND THE EARLY INDIAN PAST EPIGRAPHY: THE STUDY OF INSCRIPTIONS NUMISMATICS: THE STUDY OF COINS CONCLUSIONS

A 12TH CENTURY MANUSCRIPT OF THE PRAJNAPARAMITA

In 1148 CE, rary and scholarly ambitions, began writing a bKalhana, a man with liteook. Kalhana belonged to a well-connected Brahmana family of Kashmir. His father Chanpaka was at one time closely associated with the royal court, but by the time Kalhana was born, the family had fallen out of favour. Kalhana worked hard for two years, recording local traditions and examining manuscripts, chronicles, inscriptions, coins, and monuments. He drew on his family members’ political experience and his personal observation of events that were unfolding in his own lifetime. The book was completed in 1150 CE and was titled Rajatarangini (River of Kings). Consisting of eight

cantos, each called a taranga (wave), it gave a connected account of the kings of Kashmir from the early ones of legend to the historical rulers of the 12th century. Kalhana is often described as India’s first historian. He asserts in the Rajatarangini that a person who recounts the events of the past must do so like a judge, without bias or prejudice. However, his book does not always distinguish between fact and legend, and often explains events by citing fate. It is not surprising that there are differences in perspective between a 12th century historian such as Kalhana and historians of more recent times. Moreover, Kalhana considered himself primarily a gifted and skilful poet, one who could make pictures of the past come vividly alive. He described the natural beauty of Kashmir with pride and feeling, wove lively character sketches, and gave dramatic descriptions of political events.

The past, like the present, is complex and can be looked at from many perspectives. There can never be a single, final, perfect history. There can never be a complete or exact picture of what happened in the past; the task of the historian is to bring us as close as possible to such a picture. Historical analysis involves carefully examining the available sources of information, searching for fresh evidence, and devising creative, innovative ways of interpreting historical data. It involves asking new questions and searching for new answers to old ones. Debate and disagreement are an important part of the growth of all forms of knowledge, and history is no exception. All historical interpretations are ultimately based on evidence derived from the sources of history, conventionally divided into two categories—literary and archaeological. From a historian’s point of view, literary sources include all texts—long or short, written or oral; archaeological sources include all tangible, material remains. But these distinctions are not absolute. All remains of the past, including literary manuscripts, are actually material in nature. And certain kinds of archaeological sources which have writing on them—inscriptions, coins, and inscribed images—can be considered both material objects and texts. The ways in which historians have used different kinds of sources to construct the history of ancient and early medieval India will become clear as you read this book. This chapter gives a broad overview of the major sources, highlighting their general features, and the important issues that have to be kept in mind while using them as windows to the past. Reading Ancient Texts from a Historical Point of View

All literary works are connected to the historical contexts in which they are produced and in which they circulate. However, an ancient text does not necessarily offer a simple or direct reflection of the society of its time. It constitutes a complex representation of that society and a refracted image of the past. Information has to be teased out with care, skill, and ingenuity to make historical inferences. Many early religious texts were not primarily meant to be read but to be recited, heard, and performed. They were passed on orally from one generation to the next, even after they were available in the form of written manuscripts. PRIMARY SOURCES

Ancient palm leaf manuscripts

Paper was invented in China in the 3rd century BCE. New techniques led to its increasing use and by the 4th century, paper had replaced bamboo strips as standard material for writing in that country. Wood block printing probably began during the rule of the Sui dynasty (581–618 CE) and became popular during the Tang period (618–907 CE). In India, on the other hand, traditional writing materials and methods continued to be used for many centuries. Ancient Indian manuscripts were often made with palm leaves. Here is a description of how such manuscripts (known as talapatra in Sanskrit, olai in Tamil) were usually made: The leaf used was either from the talipot palm (Corypha umbraculifera; tali in both Sanskrit and Tamil) or palmyra palm (Borassus flabelliformis, Sanskrit tala, Tamil panai). The talipot leaf is larger, thinner, and more flexible and durable than that of the palmyra. Talipot leaves may measure about 90 × 8–9 cm, and the palmyra ones about 50 × 3–4 cm. The selected leaves were cut to the right shape and size. They were then pierced in one, two, or three places (on the left, middle, and right top). A string was woven through these holes, and then wound around the leaves. One end of the string was knotted or was tied to a small object (e.g., a shell, wooden peg, or button) to prevent it from slipping out of the holes. The cover of the palm leaf manuscript was made of wood, dry palm petioles, or in rare cases, ivory.

The writer engraved letters on the leaf with a stylus (a pointed, pen-like object). The leaf was then smeared with soot or powdered charcoal mixed with vegetable juice, so that the black mixture filled the grooves and the writing was easy to read. The letters ran parallel to the length of the leaves. In some cases where the leaf was very long or when the text was in verse, the words were written in two or three columns. If there was a commentary, it was usually written above, below, or sometimes around the text. Page numbers were often given in the right margin.

Palm leaf manuscripts had to be stored very carefully as they were vulnerable to many natural hazards such as heat, insects, water, fungus, dust, and fire as well as the danger of destruction by human hands. Scribes kept the manuscript tradition alive by repeatedly making copies of old manuscripts. This vibrant tradition started declining around the 19th century with the coming of the printing press.

There are special techniques for treating and preserving old palm leaf manuscripts. First, the manuscript is fumigated or treated with insecticides (e.g., thymol, chloromate solution, formaldehyde, phosphene gas, or ethylene oxide). The leaves are then cleaned using solvents such as water, detergents, or ethyl alcohol. Next, any split, broken, or damaged portions are repaired. This can be done using special, thin paper and a water soluble mixture including small quantities of polyvinyl acetate and methyl cellulose. Once the repairs are complete, the leaves are oiled to make them flexible and polished gently with a soft, dry cloth. They can then be restrung and the covers attached. The repaired manuscript has to be stored carefully so that it is protected from any fresh damage. The discovery, preservation, and care of ancient manuscripts are crucial parts of the preservation of the historical heritage. There are thousands of old manuscripts in various parts of the subcontinent whose contents have not yet been studied or published. It is impossible to estimate just how many have been destroyed and how many are waiting to be discovered.

A text can be read in many different ways from a historical point of view, but certain important issues have to be addressed while doing so. Foremost among these are its age and authorship. Ancient texts are much older than their surviving manuscripts, and have had a life of their own. They have grown and changed over time and this process of growth and change—the period of composition—could in some cases have lasted for hundreds of years before they were compiled or given a more or less final shape. A text can be used as a source of historical information for the period during which it was composed, but if the composition stretched over a long period of time, it becomes essential to identify its different chronological layers and the various additions or interpolations made over time. This is not easy and requires a very careful analysis of language,

style, and content. Certain texts have been analysed in this manner, resulting in the publication of critical editions accompanied by a critical apparatus. A critical edition is prepared after a careful study of different manuscripts of a text and identifies its original core. The critical apparatus directs attention to variations across manuscripts and different commentarial interpretations. Many early texts were the work of not one, but many authors. Even if many of these authors must remain anonymous, it is important to identify their background and the perspectives and biases they reflect, such as those of class, religion, and gender. Other questions that can be asked about these texts include: Where were they composed and in which geographical area did they circulate? Who transmitted them and how did they go about doing so? Who was their target audience? What was the place of these texts within prevailing social and political power structures and cultural traditions? Analysing a text from the historical point of view does not mean mechanically plucking out selfevident ‘facts’. The information a text provides has to be carefully understood within the framework of the particular genre or type of literature it represents. In the case of poetry or drama, the analysis requires sensitivity to the literary conventions of the time and the writer’s style and imagination. In other cases, a text may represent an ideal, not an actual situation and it cannot be read as a description of what was actually happening at the time. Ancient texts often contain myths, and although myths can tell us indirectly about history, the two should not be confused with each other.

F IGURE 1.1 THE PERIOD OF COMPOSITION OF SOME IMPORTANT ANCIENT INDIAN TEXTS

THE CLASSIFICATION OF LITERARY SOURCES: LANGUAGE, GENRE, AND CONTENT

Ancient and early medieval Indian texts can be divided into categories on the basis of language, genre, content, age, and the tradition or class of literature to which they belonged. Linguists and philologists (scholars who study old languages) have divided the languages of the world into different families. Languages belonging to the same family have certain structural similarities and share a significant number of similar, related words (or cognates). For instance, Hindi, Punjabi, Marathi, Bengali, Assamese, Gujarati, Sindhi, Oriya, Nepali, and Kashmiri belong to the IndoEuropean family. So do Persian, Greek, Latin, German, French, Dutch, Italian, Spanish, Armenian, and many other languages of Europe and Asia. Languages of the Dravidian family—Tamil,

Malayalam, Telugu, Kannada, and Tulu— are today largely spoken in South India. Exceptions include Brahui, which is spoken in the Baluchistan area of Pakistan, Gondi in central India and Malto in the Rajmahal hills of eastern India. Santali, Khasi, Mundari, and some other languages of eastern India belong to the Austro-Asiatic family. Certain languages of the North-East, such as Manipuri, Bodo, Garo, and Lushai belong to the Tibeto-Burmese family. Andamanese, one of the languages spoken in the Andaman Islands, is not apparently related to any of the known language families. The oldest surviving texts in the Indian subcontinent—the Vedas—are in Sanskrit. Sanskrit belongs to the Indo-Iranian branch of the Indo-European family of languages, as do ancient Pali and Prakrit. There were various dialects of Prakrit—e.g., Maharashtri, Shauraseni, and Magadhi. Apabhramsha is a term used for the further development of Prakrit up to the end of the 1st millennium CE. Among the Dravidian languages, Tamil has the oldest literature, followed by Kannada. Many of the other Indian regional languages and dialects we are familiar with today took shape between c. 1000 and 1500. The various languages were not closed, separate worlds, but overlapping and interacting ones.

F IGURE 1.2 LANGUAGES SPOKEN IN INDIA TODAY

Languages have histories and change with the times. The pre-classical Sanskrit of the Rig Veda is different from the classical Sanskrit of Kalidasa’s poetry. The term ‘classical Sanskrit’ refers to the

language whose rules were codified by the 5th/4th century BCE grammarian Panini in his Ashtadhyayi. Another important Sanskrit grammar is Patanjali’s Mahabhashya (2nd century BCE). The oldest surviving Prakrit grammar is Vararuchi’s Prakritaprakasha, whose date is debated. The ancient Tamil of the Sangam poems is different from modern Tamil. The Tolkappiyam is the oldest surviving Tamil grammar; parts of it go back to the early centuries CE. Such grammatical texts tell us about the structure of ancient languages and they also contain incidental historical references to their time. Ancient Indian texts are sometimes divided into religious and non-religious (or ‘secular’) texts. Although this is a handy distinction, there are a few things worth keeping in mind. The English word ‘religion’ attaches great importance to belief, and suggests fixed, rigid, mutually exclusive boundaries and distinct religious identities. No ancient Indian word has such a meaning. The Sanskrit dharma or the Pali dhamma, for instance, had a broader reference to a path that people should follow or an exemplary way of life. They included many different kinds of things—codes of conduct, social practices, forms and objects of worship, ritual activity, traditions, and philosophical ideas. Ancient societies did not make the kind of distinction between the religious and the secular domains with which we are familiar in modern times. Therefore, we should not be surprised to find an interweaving of what appear to be religious and non-religious themes and content in ancient texts. Some of the major literary sources for the history of ancient and early medieval India are discussed in the following sections. As the volume of texts is considerable, these should only be considered a representative sample. The idea is to give a brief introduction to their range, with a special focus on texts frequently used and cited by historians. Most of these works were not historical texts, i.e., they were not written with the conscious aim of maintaining an account of what happened in the past. But, as we shall see in the course of this book, texts of any kind can be used as sources of history. THE VEDAS

In the Hindu tradition, the Vedas have the status of shruti (literally, ‘that which has been heard’). They are thought to embody an eternal, self-existent truth realized by the rishis (seers) in a state of meditation or revealed to them by the gods. The category of smriti (literally, ‘remembered’) texts includes the Vedanga, Puranas, epics, Dharmashastra, and Nitishastra. The word Veda comes from the root vid (literally, ‘to know’) and means ‘knowledge’. There are four Vedas—Rig, Sama, Yajur, and Atharva. The Rig Veda contains the world’s oldest surviving poetry, some of it of extraordinary beauty and philosophical depth. Each Veda has four parts, the last three of which sometimes blend into each other—the Samhita, Brahmana, Aranyaka, and Upanishad. The Rig Veda Samhita is a collection of 1,028 hymns (suktas) arranged in 10 books (Mandalas). The Sama Veda consists of 1,810 verses, mostly borrowed from the Rig Veda, arranged according to the needs of musical notation. The original melodies are, however, lost. The Yajur Veda deals with the details of the performance of rituals. The Atharva Veda is the latest Veda and contains hymns (some from the Rig Veda), but also spells and charms which reflect aspects of popular beliefs and practices. The Brahmanas (this term should not be confused with the Brahmana varna or caste) are prose explanations of the Samhita portions and give details and explanations of sacrificial rituals and their outcome. The Aranyakas (forest books) interpret sacrificial rituals in a symbolic and

philosophical way. There are 108 Upanishads, among which 13 are considered the principal ones. The Upanishads contain a great variety of philosophical ideas about sacrifice, the body, and the universe, but are most closely associated with the concepts of atman and brahman. Within the Vedic corpus as a whole, Books 2–7 (known as the family books) of the Rig Veda Samhita are considered the oldest; the later portions of this Samhita, along with all the other Vedic texts, comprise later Vedic literature. There are several recensions (shakhas) of the Vedas, associated with different schools (charanas) of Vedic study and interpretation. (The terms shakha and charana are often used interchangeably.) The Shakala shakha is the only surviving recension of the Rig Veda. The texts of the Yajur Veda are divided into those of the Shukla (White) school and Krishna (Black) school. The recensions of the Shukla (also known as Vajasaneya) Yajur Veda are the Madhyandina and Kanva. The Black school is represented by the Kathaka, Kapishthala, Maitrayani, and Taittiriya recensions. The main difference between the texts of the two schools is that the Samhitas of the White school contain only the mantras (prayers and sacrificial formulae), while in the texts of the Black school the mantras are accompanied by a commentary describing and discussing various aspects of the sacrificial rituals. The Kauthuma, Ranayaniya, and Jaiminiya (or Talavakara) are recensions of the Sama Veda, and the Shaunaka and Paippalada of the Atharva Veda. References in inscriptions mention other recensions of the Vedas that once existed but are now lost. Vedic texts comprise a religious literature, and references to possible historical events are few. For example, Book 7 of the Rig Veda Samhita refers to a battle of 10 kings, in which Sudas defeated a number of adversaries who had confederated against him. Historians have tried to reconstruct various aspects of the culture represented in the Vedas, but it is not easy to interpret this vast and complex literature. A major problem in using the Vedas as a source of history is the problem of dating the Rig Veda. The dates that have been suggested for the composition of this text range from c. 6000 BCE to 1000 BCE. Many historians take c. 1500–1000 BCE as the period of composition of early Vedic literature and c. 1000–500 BCE as that of later Vedic texts. This chronology is essentially based on the tentative dates suggested by Max Müller in the 19th century. Vedic literature forms an important part of the Brahmanical tradition—texts preserved and transmitted by a section of Brahmana males. It reflects their religious beliefs, practices, and points of view. As a source of history, these texts are used for information about life in parts of north-western and northern India during the 2nd and 1st millennia BCE. But apart from the question of dates, as we shall see later on, there are several problems in co-relating the evidence from the Vedas with archaeology. A number of supplementary texts known as Vedanga (literally, ‘limbs of a Veda’) aimed at helping the proper recitation, use, and understanding of the Vedas. These include works on phonetics (shiksha), metre (chhanda), grammar (vyakarana), etymology (nirukta), ritual (kalpa), and astronomy (jyotisha). The broad period of composition of Vedanga literature is c. 600–200 BCE. Yaska’s Nirukta, a work on the etymology of words in the Rig Veda, belongs to the 6th century BCE. THE TWO SANSKRIT EPICS: THE RAMAYANA AND MAHABHARATA

The two Sanskrit epics, the Mahabharata and Ramayana, fall within the category of smriti as well as itihasa (traditional history), although the Ramayana is sometimes classified as kavya (poetry). Similarities in language and style suggest that they emerged from a common cultural milieu. The Mahabharata refers to Valmiki and the Ramayana, and outlines the Rama story in a section called the Ramopakhyana. The Ramayana in turn mentions the Kurus, Hastinapura, and Janamejaya, although it does not mention the Mahabharata war. The two epics were clearly aware of each other, at least in their later stages of development. The composition of the Mahabharata can be placed between c. 400 BCE and c. 400 CE, and the Ramayana between the 5th/4th century BCE and the 3rd century CE. More recently, Hiltebeitel (2001: 18–20) has suggested a shorter period of composition for the Mahabharata, from the mid-2nd century BCE to the year zero. Nevertheless, the fact that the different stages in the composition and development of the epics could well have spanned many centuries, possibly even a millennium, should make it obvious why most historians no longer use the term ‘epic age’. The epics are magnificent texts with powerful stories that have captured the imagination of millions of people over the centuries. To use them as historical sources, it is necessary to identify their internal chronological layers, which is not an easy task. According to tradition, Rama lived in the treta yuga (age) and the Mahabharata war happened later, in the dvapara yuga. However, some historians argue that the events and characters associated with the Mahabharata reflect a slightly earlier period than those of the Ramayana. This is because the setting of the Mahabharata is the Indo-Gangetic divide and the upper Ganga valley, while in the Ramayana, the centre of political gravity had clearly shifted eastwards, to the middle Ganga valley. The strong women characters of the Mahabharata suggest an earlier stage of social development, when women were less subordinated to men compared to later times. The practice of niyoga (levirate; i.e., when a husband deputes his conjugal rights over his wife to another man in order to produce an heir) in the Mahabharata also suggests a social stage that is prior to that of the Ramayana, which reflects much stricter controls over women. The Mahabharata consists of 18 Parvas (books) and has two main recensions—a northern and southern. The core story concerns a conflict between two sets of cousins—the Kauravas and the Pandavas—and a great war that was fought between them at Kurukshetra. But the text also contains a huge amount of material that has little or no connection with the main story. According to tradition, it was composed by Vyasa, but in its present form, it is clearly not the work of a single individual. The Mahabharata is truly an encyclopaedic work, and it boasts of this fact. A heroic story formed the core to which many other stories, sermons, and didactic portions containing teachings, were added over centuries. The additions include the sermon on dharma given by Bhishma as he lay dying on a bed of arrows, and the stirring discourse of Krishna to Arjuna on the eve of the war, known as the Bhagavad Gita. Whether a bitter war between the Pandavas and the Kauravas ever happened cannot be proved or disproved. It is possible that there was a small-scale conflict, transformed into a gigantic epic war by bards and poets. Some historians and archaeologists have argued that this conflict may have occurred in about 1000 BCE. The Ramayana exists in the form of two main recensions—northern and southern; the northern recension can be further divided into the north-eastern, north-western, and western. The language of the northern recension is more elaborate and polished than that of the southern one. The epic consists

of seven Kandas (books), of which the first (Bala Kanda) and last (Uttara Kanda) are later interpolations. The basic story is about Rama, prince of Kosala; his banishment to the forest due to the intrigues of his wicked stepmother; the abduction of his wife Sita by Ravana, the king of Lanka; Sita’s rescue; and Rama’s return to the capital, Ayodhya, to become king. The compact vocabulary and style indicate that the core of the text was the work of a single individual, traditionally identified as Valmiki. Valmiki appears in the Balakanda, where he is inspired to compose the epic, and in the Uttarakanda, where he gives refuge to Sita who has been disowned by Rama. Excavations at the site of Ayodhya have indicated the existence of a settlement here from the Northern Black Polished Ware (NBPW) phase, which may go back at the earliest to c. 700 BCE. However, as with the Mahabharata, the archaeological evidence does not tell us whether there is any historical basis to the events or the characters of the Ramayana. The popularity and dynamism of the Rama story is indicated by the fact that apart from the Valmiki Ramayana (which seems to be the oldest version) there are numerous other tellings of the Rama story— a Jaina version (the Paumachariu of Vimalasuri, in Prakrit), a Buddhist version (the Dasharatha Jataka in Pali), a 12th century Tamil version by Kamban (the Iramavataram), and the Ramcharitmanas (16th century) by Tulsidas, to name only a few. There are also innumerable oral versions of the story. The Rama legend has enjoyed great popularity in other parts of Asia as well and there are various tellings of the story in Tibet, Myanmar, Laos, Cambodia, and Indonesia. PRIMARY SOURCES

Archaeology and the Mahabharata

Archaeological explorations and excavations at places mentioned in the Mahabharata—e.g., Hastinapura, Kurukshetra, Panipat, Tilpat, Baghpat, Mathura, and Bairat—have given evidence of a pottery called Painted Grey Ware (PGW) which goes back to c. 1000 BCE. This shows that these sites were inhabited around this time, and the nature of the remains suggests that the people who lived here shared a pastoral-cum-agricultural lifestyle. There is another sort of evidence from Hastinapura: The Matsya and Vayu Puranas state that during the reign of king Nichakshu (fifth king after Parikshit, grandson of Arjuna, who became king after the war), due to a flood in the Ganga, the capital was shifted from Hastinapura to Kaushambi. Excavations at Hastinapura gave evidence of a flood in the Ganga, after which the site was deserted for several centuries. However, it is not necessary that this was the same flood mentioned in the Puranas. There is a strong local tradition that the Purana Qila in New Delhi marks the place where Indraprastha, the capital of the Pandavas, once stood. Shams Siraj Afif’s Tarikh-i-Firuz Shahi (14th century) states that Indraprastha was the headquarters of a pargana (district). A 14th century stone inscription found in Naraina village in west Delhi also mentions Indraprastha. The 16th century Ain-i-Akbari of Abul Fazl states that Humayun’s fort was built at the place where Indraprastha, capital of the Pandavas, was located long ago. In fact, till the end of the 19th century, there was a village called Indarpat inside the fort walls.

Excavations carried out at the Purana Qila between 1954 and 1971 revealed several archaeological levels ranging from the 4th century BCE to the 19th century CE. The discovery of a few stray pieces of PGW indicated the possibility that an older settlement was located somewhere nearby. However, there is no way of knowing for sure whether this settlement had any connection with the Mahabharata legend.

P URANA QILA EXCAVATIONS IN PROGRESS, 1954

Archaeology cannot really prove or disprove the historicity of epic events or characters. The crux of the matter is that there is a qualitative difference between the nature of literary and archaeological evidence. The epic imaginatively weaves together an event-centred narrative about people and places. Archaeology, on the other hand, tells us about general patterns of material culture, and cannot easily be used to corroborate textual details about individuals or events.

PRIMARY SOURCES

The chronological layers in the Ramayana

On the basis of a careful analysis of language, style, and content, J. L. Brockington (1984) has identified five distinct chronological and cultural stages in the development of the Ramayana. The epic started taking shape as an oral composition during the 5th–4th centuries BCE (stage I). The story emphasized the heroic element and its geographical horizons were limited. The material culture and social structure represented were relatively simple, the religious ideas and practices closer to those of the Vedas than the Puranas. In stage II, dated 3rd century BCE–1st century CE, there was a shift from the heroic to the aesthetic element. The geographical awareness of the text expanded eastward into the lower Ganga valley. References to social and economic life, such as the emphasis on the chastity of women and the descriptions of cities and trade caravans, suggest increasing levels of class stratification and subordination of women. The power of the king was emphasized and warfare had become more elaborate. The story was imbued with a religious significance. Stage III belongs to the 1st–3rd centuries CE. By then urbanization had spread to many new areas. The division of society into four varnas was emphasized. The king was exalted as a protector of his people and the social order. The subordination of women had increased. Vedic gods such as Brahma and Indra were still important, but Vishnu and Shiva had appeared on the scene and were exalted. Books I and VII were added to the epic during this period. In stage IV (4th–12th centuries), the religious and aesthetic emphasis increased. Descriptions of society underlined the pre-eminence of the Brahmanas and the low position of the Shudras and out-castes. References to the inauspiciousness of widows and the practice of sati (the ritual self-

immolation of widows) reflect the increasing subordination of women. Vishnu and Shiva emerged as supreme gods in a religious milieu marked by temple worship and pilgrimage. The trends visible in the fourth stage were strengthened from the 12th century onwards (stage V). Apart from these different cultural stages, Brockington also identifies corresponding changes in the delineation of the main characters of the story. For instance, he argues that in stage I, Rama was essentially considered an exemplary human and that it was towards the end of stage II that he started being conceived of as divine. In stage III, Rama’s victory over Ravana came to be presented as a victory of dharma (righteousness) over evil. Although there are references to devotion to Rama in this stage, the divine character of Rama, his association with Vishnu, and his description as an incarnation (avatara or pradurbhava) of Vishnu are regular features of stages IV and V. Brockington talks of the transformation of a heroic epic into a religious epic. However, Pollock (1991: 52) argues that the Ramayana was pervaded by the idea of Rama’s divinity from the very outset.

The various tellings often have different beginnings and endings, and characters and events are moulded in different ways (see Richman, 1992). For instance, in the Paumachariu, Ravana is presented as a tragic hero who is killed by Lakshmana, not by Rama (who embodies all the Jaina virtues, including non-violence). Apart from written and oral versions of the story, the Ramayana has also been the subject of art and performance—sculpture, painting, plays, dance dramas, and television serials. The epics can be read in many different ways from the historical point of view. While most scholars have focused on debating the historicity of their events, some have tried to describe their many different cultural layers. Another approach is to read such texts as a response to a specific kind of historical context. For instance, James L. Fitzgerald (in Mittal and Thursby, 2005: 54) has argued that the Mahabharata was a Brahmanical response to certain specific historical developments: the increasing popularity of religious traditions such as Buddhism and Jainism, and the rise of dynasties such as the Nandas and Mauryas, who extended support to them, were perceived by a section of the Brahmanas as threatening the Brahmanical order. The Mahabharata was their response to this perceived crisis. THE PURANAS

The word ‘Purana’ means ‘old’. According to tradition, the Puranas were composed by Vyasa, but it is clear that in the form in which they have come down to us, they were not the work of one person nor of one age. There are 18 Mahapuranas (great Puranas), and many more Upapuranas (secondary Puranas). The standard list of the 18 Mahapuranas includes the Vishnu, Narada, Bhagavata, Garuda, Padma, Varaha, Matsya, Kurma, Linga, Shiva, Skanda, Agni, Brahmanda, Brahmavaivarta, Markandeya, Bhavishya, Vamana, and Brahma. The origins of the Puranas may have overlapped to some extent with the Vedas, but their composition stretched forward into the 4th– 5th centuries CE, and in some cases, even later. The Puranas are supposed to have five characteristics (pancha-lakshanas), i.e., they are supposed

to discuss five topics—the creation of the world (sarga); re-creation (pratisarga); the periods of the various Manus (manvantaras); the genealogies of gods and rishis (vamsha); and an account of royal dynasties (vamshanucharita), including the Suryavamshi and Chandravamshi kings, whose origin is traced to the sun and the moon. Actually, not all Puranas deal with all these five topics, and most of them deal with much more. The conception of time in the Puranas is mind-boggling. There are four ages or yugas—krita, treta, dvapara, and kali, all consisting of thousands and thousands of years. These four yugas make up a mahayuga, and 1,000 mahayugas constitute a kalpa. Every kalpa is divided into 14 manvantaras, each presided over by a Manu. One yuga follows the other, and the periodic destruction of the world is followed by its re-creation. This cycle of time is connected with the cyclical decline and revival of dharma. The earliest parts of the Puranic genealogies are either entirely or partly mythical. The later genealogies of kings of the kali age (which, according to tradition, began the day Krishna died, 20 years after the Mahabharata war) have historical material. The account is given in the future tense in the form of a prophecy, because Vyasa is supposed to have lived at the end of the dvapara yuga and the beginning of the kali yuga, before the events he is supposed to be describing. The Bhavishya Purana is mentioned in some Puranas as the original authority for the genealogies, but the present versions of this text have incomplete material on the subject. Although their details do not always match, the Puranas—especially the Vayu, Brahmanda, Brahma, Harivamsha, Matsya, and Vishnu—do provide useful information on ancient political history. They refer to historical dynasties such as the Haryankas, Shaishunagas, Nandas, Mauryas, Shungas, Kanvas, and Andhras (Satavahanas). They also mention certain kings, with names ending in the suffix ‘naga’, who ruled in northern and central India in the early centuries CE, about whom very little else is known. The dynastic lists end with the Guptas (4th–6th centuries), indicating that most of the Puranas were compiled at about this time. However, some are later—e.g., the Bhagavata Purana belongs to the 10th and the Skanda Purana to the 14th century, with additions made up to the 16th century. The Puranas have accounts of mountains, rivers, and places, which are useful for the study of historical geography. They also reflect the emergence of religious cults based on devotion, especially towards the gods Vishnu and Shiva and the goddess Shakti. This devotion was expressed through the worship of images of deities in temples, pilgrimage (tirtha), and vows (vrata). Some of the Puranic myths such as the stories of encounters and interactions between demons (rakshasas, asuras), gods (devas), and sages (rishis) are interpreted by historians as allegorical representations of interactions among people belonging to different cultures. The Puranas had a very important function in the Brahmanical tradition as vehicles of Brahmanical social and religious values. At the same time, they also reflect the interaction of Brahmanical and non-Brahmanical cultural traditions and the emergence and development of Hindu religious practices. THE DHARMASHASTRA

The Sanskrit word dharma (from the root dhri, meaning ‘to maintain, support, or sustain’) is very rich in meaning and difficult to translate. The concept of dharma is based on the idea that the universe is governed by a certain natural law and that the moral laws guiding people’s lives should be in consonance with that natural law.

Dharma refers to the proper, ideal conduct of a person living in society, a course of action which leads to the fulfilment of the goals of human life. These goals, known as purusharthas, are dharma (righteous conduct), artha (material well-being), kama (sensual pleasure), and moksha (deliverance from the cycle of rebirth). In this scheme of things, material gain and sensual pleasure are considered desirable goals, if pursued in accordance with dharma. The concept of dharma is closely tied up with the idea of samsara—the cycle of birth, death, and rebirth. The fruits of dharma include the acquisition of spiritual merit (punya), and its impact is supposed to be felt not only in this life but in future lives as well. The obligations of dharma are considered as applicable to and binding on everybody. Therefore, dharma also means duty. A special group of Sanskrit texts dealing specifically with dharma are collectively known as the Dharmashastra. These texts can be subdivided into three groups. The first two are the Dharmasutras (composed during c. 600–300 BCE) and the Smritis (c. 200 BCE–900 CE). The third includes brief and elaborate commentaries (Tikas and Bhashyas, respectively), collections with comments and conclusions (Nibandhas), and compendia of views from different texts (Sangrahas), all composed between the 9th and the 19th centuries. As there is little variation in language or style within a particular group of Dharmashastra texts, it is not always easy to assign absolute dates to individual works. The Dharmasutras are part of Vedanga literature as well as the Dharmashastra corpus. Vedanga literature includes the Kalpasutras (aphorisms on ritual), which are divided into Shrautasutras, Grihyasutras, and Dharmasutras. Sutra (literally, ‘thread’) refers to a style in which ideas are expressed in very short, condensed statements. The Shrautasutras deal with Vedic sacrifices that required the use of at least three fires. The Grihyasutras deal with the simpler domestic sacrifices involving the use of only one fire. The rituals they discuss include daily sacrifices to be performed by a householder, mainly involving oblations of ghee or offerings of flowers and fruits. They also describe the samskaras (literally, ‘preparation’, ‘arrangement’)—rituals marking important life stages, such as upanayana (initiation), vivaha (marriage), and antyeshti (funerary rites). The Dharmasutras deal with dharma. Dharmashastra recognizes three sources of dharma—shruti (i.e., the Vedas), smriti (i.e., the Smriti texts), and sadachara or shishtachara (good custom or the practices of the learned, cultured people). As a matter of fact, the Samhitas of the Vedas do not contain direct discussion of rules of conduct, so the second and third sources of dharma are very important. A person’s dharma depends on gender, age, marital status, varna, and ashrama. The four varnas are—Brahmana, Kshatriya, Vaishya, and Shudra. The first three of these are referred to in the Brahmanical tradition as dvija (literally, ‘twice-born’) as they alone have the right to the sacred-thread ceremony, which is considered similar to a second birth. The ashrama system went through several stages of development and ultimately divided the life of a dvija male into four stages—brahmacharya (celibate studenthood), grihastha (the householder stage), vanaprastha (partial renunciation), and sannyasa (complete renunciation). The fourth ashrama is not obligatory. The ashramas represent an ideal scheme and it should not be imagined that people in ancient India necessarily followed it in real life. Further, it was not supposed to apply, even as an ideal, to women or Shudras. Apart from norms of social behaviour, Dharmashastra deals with a number of other issues including personal, civil, and criminal law. However, the ‘laws’ of these ‘law books’ are not like the provisions of the Indian civil or penal codes. We do not know to what extent their recommendations

were actually used or applied in early times. These texts are normative and prescriptive—they talk about the way things should be, from the point of view of a section of Brahmana males who were the ‘dharma experts’ and also the implied subject for many of the rules. Although the Dharmashastra texts do not directly describe the society of their time, certain inferences about social practices can be made on their basis. Contradictions within or across texts may indicate different opinions among experts, differences in customary practices in different areas, or changes in social norms over time. The Brahmanical tradition had some amount of in-built elasticity in order to come to terms with social reality. PRIMARY SOURCES

Theory and practice in the Dharmashastra

The Dharmashastra texts reveal the tension between theory and practice within the Brahmanical tradition. They divide society into four varnas, but also refer to the more numerous jatis (castes), which they explain as the outcome of intervarna marriages (varna-samkara). Although they assert that everybody must follow the dharma of their varna, they concede that in times of emergency or acute distress, people can follow the duties of other varnas. They refer to the dharma of different regions (desha-dharma), castes (jati-dharma), and families (kula-dharma). Consider the following examples based on the Manava Dharmashastra, often referred to as the Manu Smriti, a text generally assigned to between c. 200 BCE and 200 CE (a more recent view places it in the 2nd–3rd centuries CE): A. The Manu Smriti forbids marriage between a man and the daughter of his maternal uncle or paternal aunt. Medatithi, the 10th century commentator on the text, states that such cross-cousin marriages are against dharma. But Madhava, the 14th century commentator on the Parashara Smriti, gives detailed arguments to show that there was nothing wrong with such marriages, citing Vedic passages and custom. B. The Manu Smriti condemns marriage between a dvija man and a Shudra woman. But when it talks of the division of property, it specifies the shares to be given to the sons born of a Brahmana, Kshatriya, or Vaishya father by a Shudra woman. C. The text states that a widow should not remarry. But it fixes the length of time a woman should wait for a husband who is missing, and lays down the inheritance rights of sons with one mother and two fathers (i.e., a son whose mother has married a second time). D. In one place, the Manu Smriti forbids the eating of meat. However, elsewhere, it includes meat among the items to be offered to a Brahmana invited to a shraddha (ceremonies in honour of and for the benefit of ancestors).

Example A shows that the author or authors of the Manu Smriti and the commentator Medatithi clearly disapproved of cross-cousin marriage. But Madhava apparently lived in a part of South India where such marriages were socially accepted, and so he defended them. Examples B and C indicate that the authors of the Manu Smriti disapproved of marriage between a dvija male and Shudra female, and did not approve of women, including widows, remarrying. But as such things did happen they had to regulate prevailing practice by laying down some rules. Example D

similarly shows that the authors of the Manu Smriti did not approve of meat eating among Brahmanas, but had to acknowledge the prevalence of non-vegetarianism. The authors of the Dharmashastra texts had to confront and try to regulate a wide variety of social practices. This, to a large extent, accounts for the variations in their opinions and prescriptions. BUDDHIST LITERATURE

Early Buddhist literature is generally divided into canonical and non-canonical texts. Canonical texts are the books which lay down the basic tenets and principles of a religion or sect. The various Buddhist schools classify their canonical literature in different ways, some into 9 or 12 Angas, others into 3 Pitakas. There are Pali, Chinese, and Tibetan versions of the Tipitaka (The Three Baskets/ Collections). The Pali Tipitaka of the Theravada school is the oldest of them all. Pali was a literary language which developed out of a mixture of dialects, particularly those spoken in the Magadha area of eastern India. The Tipitaka consists of three books—the Sutta, Vinaya, and Abhidhamma. In the Buddhist context, sutta (from the Sanskrit sutra) refers to texts that are supposed to contain what the Buddha himself said. The Sutta Pitaka contains the Buddha’s discourses on various doctrinal issues in dialogue form. With the exception of a few suttas, the authority of this work was accepted by all Buddhist schools. The Vinaya Pitaka has rules for monks and nuns of the sangha (monastic order). It includes the Patimokkha—a list of transgressions against monastic discipline and atonements for these. The Abhidhamma Pitaka is a later work, and contains a thorough study and systemization of the teachings of the Sutta Pitaka through lists, summaries, and questions and answers. The three Pitakas are divided into books known as the Nikayas (analogous but not identical to the Agamas of the Buddhist Sanskrit tradition). For instance, the Sutta Pitaka consists of five Nikayas— the Digha, Majjhima, Samyutta, Anguttara, and Khuddaka Nikayas. The Jatakas—stories of the previous births of the Buddha—are one of the 15 books of the Khuddaka Nikaya, and their composition can be placed between the 3rd century BCE and the 2nd century CE. The Khuddaka Nikaya also contains the Dhammapada (a collection of verses dealing mainly with ethical sayings), and the Theragatha and Therigatha (songs of Buddhist monks and nuns). The Therigatha, which describes women’s experience of renunciation, is especially important because it is one of the very few surviving ancient Indian texts composed by or attributed to women. According to Buddhist tradition, the Sutta and Vinaya Pitakas were recited at the first council of monks at Rajagriha immediately after the Buddha’s death, and 100 years later at the second council at Vaishali. But their composition must have extended over several centuries, up to the time of the third council convened in the 3rd century BCE during the reign of Ashoka. The composition of the basic core of the Pali Tipitaka can therefore be placed between the 5th and 3rd centuries BCE. The canon is supposed to have been written down in the first century BCE in Sri Lanka under the patronage of a king named Vattagamani, by which time it must have undergone further modifications. Non-canonical Buddhist literature in Pali includes the Milindapanha (1st century BCE–1st century CE) which consists of a dialogue on various philosophical issues between king Milinda—no doubt the Indo-Greek Menander—and the monk Nagasena. The Nettigandha or Nettipakarana (The Book

of Guidance) belongs to the same period and gives a connected account of the teaching of the Buddha. Commentaries on the Tipitaka include a 5th century work by Buddhaghosha. The first connected life story of the Buddha occurs in the Nidanakatha (1st century). The Pali or Sri Lankan chronicles—the Dipavamsa (4th–5th centuries) and the Mahavamsa (5th century)—contain a historical-cum-mythical account of the Buddha’s life, the Buddhist councils, the Maurya emperor Ashoka, the kings of Sri Lanka, and the arrival of Buddhism on that island. Apart from texts in Pali, there are several Buddhist works in Sanskrit, and in a mixture of Prakrit and Sanskrit that is often referred to as Buddhist Sanskrit or Buddhist hybrid Sanskrit. The trend towards the use of Sanskrit intensified in the Mahayana schools, but some non-Mahayana texts were also composed in Sanskrit or mixed Prakrit-Sanskrit. For instance, the canon of the Sarvastivada school is in Sanskrit. The Mahavastu, which has some Mahayana elements, gives a hagiography (sacred biography) of the Buddha and describes the emergence of the monastic order in mixed Sanskrit–Prakrit. The Lalitavistara (1st–2nd centuries), a hagiography of the Buddha associated with the Sarvastivada school but strongly tinged with Mahayana elements, is in Sanskrit and mixed Prakrit-Sanskrit.

SEE CHAPTER 8, PP . 440–41 FOR DETAILS OF THE MAHAYANA AND HINAYANA SCHOOLS

PRIMARY SOURCES

Songs of Buddhist nuns

Ubbiri’s song Ubbiri was a woman of Shravasti, who attained nibbana (enlightenment) as an upasika, i.e., laywoman. The turning point in her life was an encounter with the Buddha, which took place while she was lamenting the death of her daughter Jiva. The following song is in the form of a dialogue between the Buddha and Ubbiri. [Buddha:] Mother, you cry out ‘O Jiva’ in the woods. Come to yourself, Ubbiri. Eighty-four thousand daughters all with the name ‘Jiva’ have burned in the funeral fire. For which one do you grieve? [Ubbiri:] I had an arrow hidden in my heart

and he took it out — that grief for my daughter. The arrow is out, the heart healed of hunger. I take refuge in the Buddha-sage, the Dharma, the Sangha. Mitta's song Mitta was a Sakya woman of Kapilavastu. The first verse of her song speaks of the observances she followed as a laywoman, the second of her life after she became a nun. To be reborn among the gods I fasted and fasted every two weeks, day eight, fourteen, fifteen and a special day. Now with a shaved head and Buddhist robes I eat one meal a day. I don’t long to be a god. There is no fear in my heart. Source Murcott, 1991: 81, 21

www.pearsoned.co.in/upindersingh EXTRACTS FROM BUDDHIST TEXTS

SEE CHAPTER 8, PP . 443–44 FOR DETAILS ON THE SHVETAMBARA AND DIGAMBARA SCHOOLS

Sanskrit Buddhist texts include Ashvaghosha’s Buddhacharita (1st/2nd century) and the Avadana texts. The latter contain stories of noteworthy deeds with a moral; they include the Avadanashataka (2nd century) and the Divyavadana (4th century) which have stories connected with the Buddha and the Maurya emperor Ashoka. The 1st century Ashtasahasrika-prajnaparamita and Saddharma-

pundarika offer accounts of the various Buddhas, bodhisattvas (future Buddhas), and Mahayana doctrines. Later works of Mahayana thinkers such as Nagarjuna, Vasubandhu, Asanga, Aryadeva, Buddhapalita, and Dignaga are all in Sanskrit. Buddhist texts are important sources for the history of Buddhism, its doctrines, monastic order, and royal patrons such as Ashoka, revealing many other facets of the polity, society, and economy of their times as well. They offer a non-Brahmanical window into ancient India; however, the Brahmanical perspective is replaced by a Buddhist one. JAINA LITERATURE

The sacred books of the Jainas are collectively known as the Siddhanta or Agama. The language of the earliest texts is an eastern dialect of Prakrit known as Ardha-Magadhi. The Jaina monastic order came to be divided into the Shvetambara and Digambara schools, perhaps in about the 3rd century ce. The Shvetambara canon includes the 12 Angas, 12 Uvamgas (Upangas), 10 Painnas (Prakirnas), 6 Cheya Suttas (Cheda Sutras), 4 Mula Suttas (Mula Sutras), and a number of individual texts such as the Nandi Sutta (Nandi Sutra) and Anugodara (Anuyogadvara). There is some overlap in the content of the canonical literature of the two schools. For instance, the Digambaras accept and give prime importance to the Angas, and some of the texts they club together as the Angabahyas have corresponding Shvetambara texts. According to Shvetambara tradition, the Angas were compiled at a council held at Pataliputra. The compilation of the entire canon is supposed to have taken place in the 5th or 6th century at a council held in Valabhi in Gujarat, presided over by Devarddhi Kshamashramana. Some of the material in the canon may go back to the 5th or 4th century BCE, but changes and additions continued to be made till the 5th–6th centuries CE. In order to use such texts as historical sources, a clearer identification of their internal chronology is required. The non-canonical Jaina works are partly in Prakrit dialects, especially Maharashtri, and partly in Sanskrit, which started being used in the early centuries CE. Commentaries on the canonical works include the Nijjuttis (Niryuktis), Bhashyas, and Churnis in Maharashtri and Prakrit; the early medieval Tikas, Vrittis, and Avachurnis are in Sanskrit. The genealogical lists in the Jaina Pattavalis and the Theravalis contain very precise chronological details about the Jaina saints, but they sometimes contradict each other. The Jaina Puranas (the Shvetambaras call them Charitas) are hagiographies of the Jaina saints known as tirthankaras (literally ‘ford makers’), but they contain other material as well. The Adi Purana (9th century) narrates the life of the first tirthankara Rishabha, also known as Adinatha. The 8th century Harivamsha Purana gives a Jaina version of the stories of the Kauravas, Pandavas, Krishna, Balarama, and others. The Trishashtilakshana Mahapurana by Jinasena and Gunabhadra (9th century) has life stories of various Jaina saints, kings, and heroes. It also has sections on topics such as life-cycle rituals, the interpretation of dreams, town planning, the duties of a warrior, and how a king should rule. The Parishishtaparvan (12th century) by Hemachandra gives a history of the earliest Jaina teachers and also mentions certain details of political history. A number of Prabandhas (12th century onwards) from Gujarat offer semi-historical accounts of saints and historical characters. Jaina texts also include hymn literature and lyrical poetry. The vast Jaina didactic story (katha) literature in Sanskrit, Prakrit, and Apabhramsha can offer historians clues on the everyday life of their time. The Jaina texts in the Kannada language are discussed further on in this chapter.

www.pearsoned.co.in/upindersingh EXTRACTS FROM JAINA TEXTS

Jaina literature offers information regarding the history and doctrines of Jainism, the doctrines of rival schools, the life stories of the saints, and the life of monks and nuns in the sangha. The texts can also be used for information on other aspects of the cultural history of their times. Jaina texts have not, however, been studied or used as extensively by historians as Buddhist sources. SANGAM LITERATURE AND LATER TAMIL WORKS

The earliest literature of South India is represented by a group of texts in old Tamil, often collectively referred to as Sangam literature. A tradition recorded in post-7th century texts speaks of three Sangams or literary gatherings in ancient times. The first is supposed to have been held in Madurai for 4,440 years, the second at Kapatapuram for 3,700 years, and the third in Madurai for 1,850 years. Although the details of this legend obviously cannot be considered historical, the similarity of language and style within the Sangam corpus suggests the possibility that they were the product of some sort of literary gathering. The case for the historicity of at least the third Sangam is that some of the kings and poets associated with it are historical figures. On the other hand, there is a possibility that the legend of the Sangams may have been based on a very different event—the establishment of the Jaina sangha in Madurai in about the 5th century. In view of the controversy surrounding the tradition of the three Sangams, some scholars prefer to use the term ‘early classical Tamil literature’ rather than ‘Sangam literature’. The Sangam corpus includes six of the eight anthologies of poems included in the Ettutokai (The Eight Collections), and nine of the ten pattus (songs) of the Pattuppattu (The Ten Songs). The style and certain historical references in the poems suggest that they were composed between the 3rd century BCE and the 3rd century CE. They were compiled into anthologies in about the mid-8th century. A few centuries later, these anthologies were collected into the super-anthologies (i.e., anthologies of anthologies) called the Ettutokai and the Pattuppattu. The earliest parts of the first two books of the Tolkappiyam can also be included in Sangam literature. The Tolkappiyam is essentially a work on grammar, but it also includes a discussion of phonology, semantics, syntax, and literary conventions. There are two kinds of Sangam poems—akam and puram. Akam poems had love as their theme, while puram poems were mostly about war. A. K. Ramanujan (1999) describes puram poetry as ‘public poetry’ which dealt with all kinds of themes other than love, such as good and evil, community and kingdom. The poems were modelled on the bardic songs of older times and were orally transmitted for an indefinite period before they were written down. The anthologies include a total of 2,381 poems ascribed to 473 poets, 30 of whom were women. The poets came from cities and villages and had varied social and professional backgrounds. They included teachers, merchants, carpenters, astrologers, goldsmiths, blacksmiths, soldiers, ministers, and kings. Due to

their varied themes and authorship, Sangam poems offer a good idea of everyday life in the time when they were composed. A number of Tamil didactic works were written in the post-5th century period. The most famous of these is Tiruvalluvar’s Tirukkural, a work on ethics, polity, and love (5th– 6th centuries). Of the several Tamil epics, two of the best known are the Silappadikaram and Manimekalai. The former is a little earlier that the latter, but both were composed in about the 5th–6th centuries CE.

www.pearsoned.co.in/upindersingh POEMS FROM THE SANGAM CORPUS

Early medieval Tamil literature includes the inspired and intense devotional poetry of the Vaishnava saints (Alvars) and Shaiva saints (Nayanars or Nayanmars) and their hagiographies. Vaishnava poetry took off with the compositions of Peyalvar, Puttalvar, and Poikaialvar. In the 10th century, Nathamuni collected the Alvar hymns into the canon known as the Nalayira Divya Prabandham. The Alvarvaipavam is a sacred biography of the Vaishnava saints. Shaiva devotional literature began with the compositions of Tirumular and Karaikal Ammaiyar. The hymns of the Nayanmar saints were compiled in the 10th century by Nambi Andar Nambi and this compilation formed the core of the Shaiva canon, the Tirumurai. Nambi also wrote a work called the Tiruttondar Tiruvantati about the saints. In the 12th century, the accounts of the Shaiva saints were collected in a text called the Periyapuranam. All these texts provide valuable insights into the religious and social history of early medieval South India. New genres of Tamil poetry emerged in early medieval times, many in praise of kings and gods. The Kalampakams were poetic compositions in which the last line, word, foot, or syllable of the preceding poem formed the beginning of the succeeding one. Kovai were poems in which the verses are arranged in a thematic sequence. Compositions in this genre included: the Pantikkovai, a 6th/7th century work written in honour of the Pandya king Netumaran; Manikkavachakar’s Tirukkovaiyar (9th century) in praise of the god Shiva; and Poyyamolip Pulavar’s Tanchaivanan Kovai (13th century) about Tanchaivanan, a minister and general of a Pandya king. Ula literature comprised songs in praise of gods, sung when the image of the deity was taken out in procession. Tutu poetry consisted of poems in which a message is delivered to a god, lover, or someone else. The moral aphorisms and sayings of Avvaiyar (9th/10th century), the second of three poetesses by this name, are still popular among Tamil-speaking people today. PRIMARY SOURCES

The stories of the two Tamil epics

Although the northern epics were certainly known in early historical South India, the origins of Tamil epic narratives seem to lie in late Sangam compositions such as the Kalittokai and Paripatal rather than in northern influence. The Silappadikaram (The Song of the Anklet) by Ilankovatikal (prince ascetic) consists of 30 cantos arranged in three books. The outline of the story is as follows: Kovalan (the son of a wealthy merchant) and Kannaki are a young, happily married couple living in Puhar. Kovalan falls in love with a beautiful courtesan named Madhavi and abandons his wife. He eventually returns home after quarrelling with Madhavi. Kannaki welcomes him back and offers him her golden anklet to raise some money. They travel to Madurai, capital of the Pandya king, accompanied by a Jaina nun named Kavundi. Kovalan goes off to sell his wife’s anklet. He is accused of stealing the queen’s anklet, which looks just like Kannaki’s, and is executed. Kannaki is devastated. She proves her husband’s innocence by bursting open her other anklet—it contains a ruby, whereas the queen’s was filled with pearls. The king, who had executed a man unjustly, dies of remorse; his wife dies of grief. Kannaki tears off her left breast and hurls it onto the city in fury. Madurai is engulfed in flames. Kannaki joins her husband in heaven; on earth she comes to be worshipped as the ideal wife. Zvelebil points out that the epic’s complex treatment of guilt and evil is one of its strengths. So are its multi-layered characters with human flaws and frailties, which evolve as the story progresses. The anklet has an important symbolism— Kannaki wears her anklets in the beginning of the story, when she is happy; she removes them after she is abandoned by Kovalan. The anklet is the cause of Kovalan’s tragic end and the symbol of truth which ultimately proves his innocence. When Kannaki is united with her husband in heaven, she again wears both her anklets. Although the epic no doubt catered to an elite, educated audience, it tells us a great deal about the lives of ordinary people of the time. The Manimekalai (The Jewel Belt) of Sattanar consists of 30 cantos and a preamble. The outline of the story is as follows: Prince Udayakumara is in love with Manimekalai, who is not interested in him because she wants to renounce the world and become a Buddhist nun. In order to escape the attentions of the prince, Manimekalai assumes the form of a woman named KayaChandikai. She distributes food to the needy people of Madurai, using a magic alms-bowl. The husband of the real Kaya-Chandikai sees Manimekalai with the prince and kills him in a fit of jealousy. Manimekalai is put in prison, where she survives many ordeals to which she is subjected. Realizing that she is a saintly person, the queen begs forgiveness and sets her free. Manimekalai eventually reaches Kanchi, where a famine is raging and feeds the poor with her magic alms-bowl. She ultimately fulfils her heart’s desire by joining the Buddhist sangha. The Manimekalai is often considered somewhat inferior to the Silappadikaram in terms of its formal literary features. While the Silappadikaram has a Jaina flavour, the Manimekalai has a strong, strident Buddhist tone. Its characters are either good or bad, with few shades of grey, and the narrative is marked by many more miracles and supernatural interventions. SOURCE Zvelebil, 1974: 131–35, 140–42

Of the many Tamil renderings of the Rama legend, the most famous is Kamban’s Iramavataram. Tamil versions of the Mahabharata story were also written, of which some fragments survive. Several Tamil lexicons and grammatical works belong to the early medieval period. EARLY KANNADA AND TELUGU LITERATURE

The earliest Kannada inscriptions date from the 5th/6th century onwards, but the oldest surviving piece of literature in this language is the Kavirajamarga (The Royal Road of the Poets), a 9th century work on poetics. A well-developed tradition of prose and poetry must have existed for some time, as this work mentions many earlier writers and their works which have not survived. Karnataka was a stronghold of Jainism and a significant part of early medieval Kannada literature had Jaina themes. The best known poets of the 10th century were Pampa, Ponna, and Ranna, all of whom wrote Jaina Puranas. Pampa, author of the Adi Purana (an account of the life of the first tirthankara Rishabha or Adinatha), also wrote the Vikramarjunavijaya, based on the Mahabharata story. Ponna wrote both in Sanskrit and in Kannada, and was given the title of Ubhaya-kavichakravarti (imperial poet in both languages). Chavunda Raya, a general and minister under the Ganga kings, wrote the Trishashtilakshana Mahapurana, an account of the 24 Jaina saints, in continuous prose. In the 12th century, Nagachandra or Abhinava Pampa wrote the Ramachandracharitra Purana, one of many Jaina versions of the Rama story. The interesting Kannada works of the 12th century include Neminatha’s Lilavati, in mixed verse and prose, which tells the love story of a Kadamba prince and a beautiful princess. Place names in inscriptions from the 2nd century CE suggest the antiquity of Telugu, while epigraphs of the 5th–6th centuries CE reflect the shaping of the classical form of the language. Early medieval inscriptions used verse and are marked by a literary flavour and style. Although there may have been older works, the earliest surviving work of Telugu literature is Nannaya’s 11th century rendering of the first two-and-a-half books of the Mahabharata in mixed verse and prose. This work was written at the request of the eastern Chalukya king Rajarajanarendra. Nannaya laid the foundations of Telugu poetic style, and Telugu tradition gave him the epithet Vaganushasanundu (maker of speech). His style is marked by the use of a variety of Sanskrit and regional metres, and a combination of lengthy Sanskrit compounds with Telugu words. Tikkana, a minister associated with the court of Manumasiddhi, a ruler based in the Nellore area, added 15 Parvas to Nannaya’s Mahabharata and set new trends in narrative style. He also composed a work called the Uttararamayanamu. Another writer who seems to have lived in about this period was Nanne Choda—author of the Kumara-sambha-vamu—who describes himself as a ruler of a small principality called Orayuru. Telugu literature reached a level of maturity in the 14th century during the Kakatiya period and its highest point of achievement during the reign of the Vijayanagara king Krishnadevaraya (1509–29 CE). OTHER ANCIENT TEXTS, BIOGRAPHIES, AND HISTORIES

Early Indian literature includes a number of masterpieces of poetry and drama which can be read and appreciated for their sheer beauty and fine literary qualities. Such texts are used by historians as sources of information about the times in which they were composed. The earliest Sanskrit poets and

playwrights include Ashvaghosha and Bhasa. Ashvaghosha was the author of the Buddhacharita (which he describes as a mahakavya), Sariputraprakarana, and Saundarananda. Bhasa wrote several dramas including the Pancharatra, Dutavakya, Balacharita, and Svapna-Vasavadatta. One of the most celebrated names among Sanskrit writers of the 1st millennium is that of Kalidasa (4th– 5th centuries), author of the dramas Abhijnana-Shakuntala, Malavikagnimitra, Vikramorvashiya, and poetic works such as the Raghuvamsha, Kumarasambhava, and Meghaduta. The major early medieval poets and writers include Bharavi, Rajashekhara, and the poetess Vijayanka. Ancient dramas on historical themes are of special interest to historians, although it is necessary to remember that they were plays and not historical accounts. Vishakhadatta’s Mudrarakshasa (7th/8th century) revolves around the manoeuvres of Chanakya to win over Rakshasa, a minister of the Nandas, to Chandragupta’s side. His Devichandragupta centres on an incident set in the reign of the Gupta king Ramagupta. Narrative literature such as the Panchatantra (5th–6th centuries) and the Kathasaritsagara (Ocean of Streams of Stories, 11th century) are collections of popular folk tales that ordinary people may have known, listened to, and enjoyed. There is a vast body of ancient and early medieval technical literature on varied subjects such as grammar, mathematics, statecraft, astronomy, medicine, architecture, poetics, dramaturgy, and philosophy. Reference has already been made to grammatical texts such as Panini’s Ashtadhyayi and Patanjali’s Mahabhashya. Kautilya’s Arthashastra is a major work on statecraft. Aryabhata’s Aryabhatiya and Varahamihira’s Brihatsamhita are important astronomical texts. Other technical treatises include the Kamasutra (on sensual pleasure), the Charaka Samhita and Sushruta Samhita (on medicine), the Natyashastra (on theatre and the performing arts), and the Shilpashastras (on architecture and sculpture). Apart from indicating the level of expertise and knowledge in their respective fields, such treatises also provide various kinds of useful historical information. Philosophical texts and commentaries reflect the ideas and intellectual debates of their times. Apart from the Buddhist and Jaina texts which have already been mentioned, there is a voluminous darshana (literally, ‘a way of looking at things’) literature belonging to the Samkhya, Yoga, Nyaya, Vaisheshika, Purva Mimamsa, and Uttara Mimamsa schools. These also mention the philosophical ideas of schools whose texts have not survived, such as the materialist Charvaka or Lokayata school.

SEE CHAPTER 8, PP . 426–28 FOR A DISCUSSION OF THESE SCHOOLS

Summaries of ancient literary sources tend to miss out on unusual texts that do not fall within any of the main categories. These include a Sanskrit work on agriculture called the Krishi-Parashara, composed in Bengal some time between the 6th and 11th centuries CE. The early medieval literature of this region also includes the Dakar Bachan and the Khanar Bachan in old Bengali. These contain aphorisms and wise sayings, mostly concerning agriculture, but also other issues such as family life, illness, and astrology.

The courts of early medieval kings attracted writers and poets, some of whom wrote biographical compositions in praise of their royal patrons. The famous Sanskrit biographies include Banabhatta’s Harshacharita (7th century) about king Harshavardhana. Vakpati wrote the Prakrit Gaudavaha (8th century) about Yashovarman of Kanauj. Bilhana’s Vikramankadevacharita (12th century) is woven around the Chalukya kings, especially Vikramaditya VI. Royal biographies in Tamil include the anonymous Nandikkalambakkam (9th century), a long poem about the events of the reign of the Pallava king Nandivarman III. An 11th century work, the Kalinkattupparani by Cheyankontar, is based on the war between the Chola king Kulottunga and Anantavarman Chodaganga, the ruler of Kalinga. The poet describes and praises the heroism of the Chola king and his army commander, presenting the war as a divine conflict between the principles of good and evil. The Prithvirajaraso by Chand Bardai is an epic poem in the early Braj-bhasha dialect, woven around the Rajput king Prithviraja Chauhan. Sandhyakara Nandi’s Ramacharita is a Sanskrit work with double meaning, simultaneously narrating the story of the Ramayana and of Ramapala, an 11th/12th century king of Bengal. The 12th century Kumarapalacharita by Hemachandra is a long poem in Sanskrit and Prakrit, which tells the story of the Chaulukya kings of Gujarat and simultaneously illustrates the rules of Sanskrit and Prakrit grammar. The establishment of the Delhi Sultanate in the 11th century gave rise to a series of Persian chronicles narrating the history of various dynasties. The aim of ancient and early medieval biographers and chroniclers was as much to display their literary skills as to produce a work that would flatter their royal patrons. This has to be kept in mind when using their works as sources of history. PRIMARY SOURCES

Banabhatta and his royal biography

Banabhatta’s Harshacharita is the oldest surviving biography in India. Apart from painting a glowing picture of his patron Harsha of the Pushyabhuti dynasty, the writer also speaks about himself. The early part of Bana’s pedigree is mythical and narrates the origins of the Vatsyayana branch of the Bhargava Brahmanas, to which he belonged. The later part is historical. Bana was born in Pritikuta, a Brahmana village in the Kanyakubja area, famed for the learning and stature of its residents. His mother Rajadevi died when he was a small child, and he was brought up by his father who died when he was 14. Bana was taught by an illustrious teacher named Bharchu. In his youth, he set out on a series of travels, accompanied by his half-brothers and a colourful entourage including poets, philosophers, artists, actors, monks, ascetics, a gambler, singer, snake-doctor, goldsmith, and dancing girl. It is no wonder that he acquired a bit of a reputation. The story goes that one day Bana received a letter summoning him to present himself in Harsha’s court. The audience started off badly. The king had apparently believed the gossip about Bana’s wayward ways and treated him with scant regard. Bana was quick to defend himself, arguing that although he may have been a bit wild in his youth, he came from a respectable Brahmana family

and was currently living a blameless married life. Within a few days, he became a court favourite and many lavish presents and honours were showered on him. Bana went on to write the Harshacharita, a eulogistic biography of his patron, as well as a prose romance called the Kadambari. Bana describes the Harshacharita as an akhyayika, a genre of texts related to the itihasa tradition. The episodes in the biography are selected and narrated from a literary and aesthetic perspective. Its descriptions are vivid and literary, and sometimes show a touch of humour. The work displays Bana’s skills as a master of Sanskrit prose. Typical of the genre of royal biographies are long, elegant passages eulogizing the king. Consider, for example, the following sentence: He (i.e., Harsha) was embraced by the goddess of Royal Prosperity, who took him in her arms, and, seizing him by all the royal marks on all his limbs, forced him, however reluctant, to mount the throne—and this though he had taken a vow of austerity and did not swerve from his vow, hard like grasping the edge of a sword; clinging closely to duty through fear of stumbling in the uneven path of kings, and attended with all her heart by Truth who had been abandoned by all other kings, but had obtained his promise of protection, and waited on reverentially by the reflected images of a fair handmaid standing near, which fell on his toe-nails, as if they were the ten directions of space impersonate. According to some scholars, the Harshacharita is incomplete because it ends after Harsha’s rescue of his sister Rajyashri from the flames of the pyre on which she sought to end her life, and his accession to the thrones of Thanesar and Kanauj. However, V. S. Pathak argues that the work is complete as it has all the five well-defined thematic stages of a beginning, effort, the hope of achieving the end, certainty of success, and a conclusion. Rajyashri was Harsha’s sister, but her name also means royal glory, and Harsha’s rescuing her symbolically represents his successful acquisition of royal glory. Although Bana paints Harsha as an ideal, exemplary ruler, traces of a less perfect picture can be found in the nuances of the narrative. For instance, there are hints of a fratricidal struggle for the throne behind the portrayal of the deep brotherly love between Harsha and Rajyavardhana. SOURCE Cowell and Thomas, 1993: 57; Pathak, 1966: 30–32

This chapter opened with mention of the Rajatarangini, the 12th century historical chronicle of Kashmir by Kalhana. Kalhana refers to earlier historians and chronicles. Apart from the Nilamata Purana, he mentions 11 works of earlier scholars, none of which have survived. THE NATURE OF ANCIENT INDIAN HISTORICAL TRADITIONS

As we have seen, the literary sources for ancient and early medieval India include a large volume and variety of texts. Is there any evidence of an interest in preserving the memory of the past, of a historical tradition, in these texts? Romila Thapar (2000) has made a useful distinction between ‘embedded’ and ‘externalized’ forms of history. Embedded history is where the historical

consciousness has to be prised out, as in myth, epic, and genealogy. Externalized history reflects a more evident and self-conscious historical consciousness, reflected for instance in chronicles and biographies. Thapar points out that the embedded forms of historical consciousness tended to be connected with lineage-based societies and the externalized ones to state societies. Apart from lists of teachers, later Vedic texts contain certain types of compositions that reflect a historical consciousness. These include the dana-stutis, gathas, narashamsis, and akhyanas. The dana-stutis are hymns praising the generosity and exploits of kings. The gathas are songs in praise of kings, sung on the occasion of certain sacrifices. Narashamsis were used in rituals and are preserved in texts such as the Brahmanas and Grihyasutras. Akhyanas are narrative hymns in dialogue form, referring to mythical and possibly historical events. It is interesting to note that all these types of compositions were directly connected with the performance of sacrifices (yajnas). The king-lists in the Puranas and epics represent more substantial evidence of an ancient Indian historical tradition. As mentioned earlier, the epics are known as itihasa, and are supposed to record things that actually happened (whether they did happen in the way in which they are described is another issue). Bards known as sutas and magadhas played an important role in maintaining these historical traditions. The poets and bards of the ancient Tamil land who eulogized their royal patrons can also be seen as creators and transmitters of a historical tradition. The Buddhist Dipavamsa and Mahavamsa, which offer a mythico-historical account of how Buddhism travelled to Sri Lanka, represent a historical tradition as well. Mention may also be made of sacred biographies in the Buddhist, Jaina, and Hindu traditions. Notwithstanding their eulogistic nature, royal biographies too reflect a historical tradition. Mention can also be made of royal inscriptions, many of which have a prashasti (panegyric) containing the king’s genealogy and references to his exploits, usually with a view to shower praise on him. The Arthashastra and the Chinese pilgrim Xuanzang mention royal archives preserving official records in every Indian city, while Al-Biruni’s 11th century Tahqiq-i-Hind refers to the archives of the Shahi kings of Kabul. Unfortunately, no such ancient archives survive. While there is evidence of different kinds of historical traditions in ancient and early medieval India, these traditions were very different from our modern notions of history. The intellectuals of every age and society select the aspects of the past they consider important, and interpret and present them in their own way. Since ancient and modern societies differ from each other in so many respects, it is not surprising to find major differences in their ways of looking at the past. Modern historians distinguish between myth and history, ancient texts do not. The historical traditions of ancient India were connected with religious, ritualistic, and court contexts. History in our times is an academic discipline based on research, linked to modern institutions such as universities and research institutes. The ways in which the past was understood and represented in ancient texts are very different from the methods, techniques, and goals of historical research today. THE ACCOUNTS OF FOREIGN WRITERS

As mentioned earlier, the subcontinent was never an isolated geographical area. Since early times, traders, travellers, pilgrims, settlers, soldiers, goods, and ideas moved to and fro across its frontiers, covering vast distances over land and water. It is therefore not surprising that there are many references to India in foreign texts. Such texts reveal how people from other lands viewed India and its people, what they noticed and found worthy of description. Historians have to distinguish

between statements based on hearsay and those grounded in personal experience, between perceptive observations and cases where the writer got things completely wrong. An example of a very unreliable account is the Indica of Ktesias (4th century BCE), which is full of bizarre stories about India and Indians, collected by the author while living in Persia as a royal physician. The earliest references to India in Greek texts date from the 5th century BCE and their frequency increases thereafter. One of the most famous works is the Indica of Megasthenes, ambassador of Seleucus Nikator to the court of Chandragupta Maurya. The book is lost, but later Greek works preserve paraphrases of some of its sections. The many Greek and Latin texts of the 2nd century BCE to the 2nd century CE referring to India include the works of Arrian, Strabo, and Pliny the Elder, and the anonymous Periplus Maris Erythraei (Periplus of the Erythraean Sea). These texts are especially important for the history of Indian Ocean trade. PRIMARY SOURCES

Al-Biruni on the writing of the Hindus

The tongue communicates the thought of the speaker to the hearer. Its action has therefore, as it were, a momentary life only, and it would have been impossible to deliver by oral tradition the accounts of the events of the past to later generations, more particularly if they are separated from them by long periods of time. This has become possible only by a new discovery of the human mind, by the art of writing, which spreads news over space as the winds spread, and over time as the spirits of the deceased spread. Praise therefore be unto Him who has arranged creation and created everything for the best! The Hindus are not in the habit of writing on hides, like the Greeks in ancient times. Socrates, on being asked why he did not compose books, gave this reply: ‘I do not transfer knowledge from the living hearts of men to the dead hides of sheep.’ Muslims, too, used in the early times of Islam to write on hides, e.g., the treaty between the Prophet and the Jews of Khaibar and his letter to Kisra. The copies of the Koran were written on the hides of gazelles, as are still nowadays the copies of the Torah…. The kirtas (or charta) is made in Egypt, being cut out of the papyrus stalk…. It was in China that paper was first manufactured. Chinese prisoners introduced the fabrication of paper into Samarkand and thereupon it was made in various places, so as to meet the existing want. The Hindus have in the south of their country a slender tree like the date and coconut palms, bearing edible fruits and leaves of the length of one yard, and as broad as three fingers one put beside the other. They call these leaves tari and write on them. They bind a book of these leaves together by a cord on which they are arranged, the cord going through all the leaves by a whole in the middle of each. In central and northern India people use the bark of the tuz tree, one kind of which is used as a cover for bows….

As for the writing or alphabet of the Hindus, we have already mentioned that it once had been lost and forgotten; that nobody cared for it, and that in consequence people became illiterate, sunken into gross ignorance, and entirely estranged from science. But then Vyasa, the son of Parashara, rediscovered their alphabet of fifty letters by an inspiration of God. A letter is called an akshara. Some people say that originally the number of their letters was less, and that it increased only by degrees. This is possible, or I should even say necessary…. The great number of the letters of the Hindu alphabet is explained, firstly, by the fact that they express every letter by a separate sign if it is followed by vowel or a diphthong or a hamza (visarga), or a small extension of the sound beyond the measure of the vowel; and, secondly, by the fact that they have consonants which are not found together in any other language, though they may be found scattered through different languages—sounds of such a nature that our tongues, not being familiar with them, can scarcely pronounce them, and that our ears are frequently not able to distinguish between many a cognate pair of them. The Hindus write from the left to the right like the Greeks. They do not write on the basis of a line, above which the heads of the letters rise whilst their tails go down below, as in Arabic writing. On the contrary, their ground-line is above, a straight line above every single character, and from this line the letter hangs down and is written under it. Any sign above the line is nothing but a grammatical mark to denote the pronunciation of the character above which it stands…. After describing these characteristics of ‘Hindu’ writing, Al-Biruni goes on to acknowledge the existence of many different scripts in the land of Hind—Siddhamatrika, the most widely known and used in Kashmir, Varanasi, and the country around Kanauj; Nagara in Malwa; Ardhanagari in Bhatiya and some parts of Sindh; Malwari in Sindh; Karnata in Karnatadesha; Andhri in Andhradesha; Dirwari in Dravidadesha; Lari in Latadesha (in Gujarat); Gauri (i.e., Gaudi) in Purvadesha, i.e., the eastern country; and the Bhaikshuki, used in Udunpur in Purvadesha, described as the writing of the Buddha. SOURCE Sachau, 1964: 170–73

Many Chinese monks made long and arduous overland journeys to India, crossing mountains, plateaux, and deserts, in order to collect authentic manuscripts of Buddhist texts, meet Indian monks, and visit places of Buddhist learning and pilgrimage. The best known among those who wrote accounts of their Indian travels are Faxian (Fa Hien) and Xuanzang (Hiuen Tsang). Faxian’s travels extended from 399 to 414 CE and were confined to northern India. Xuanzang left his home in 629 CE and spent over 10 years travelling the length and breadth of the country. Yijing, another 7th century Chinese traveller, lived for 10 years in the great monastery of Nalanda. The accounts written by these pilgrims throw light on the history of Buddhism and various other aspects of their time. The rapid political expansion of the Arabs, the unity given to them by Islam, the spread of urban centres, and the patronage of the Caliphs had important and far-reaching impact on intellectual ideas

and technology in Asia and Europe. Al-Mamun, the 9th century Abbasid Caliph, established an academy called the Beyt-al-Hikma (House of Wisdom) in Baghdad. Scholars of this academy busied themselves with an ambitious project of translating Greek, Persian, and Sanskrit texts on philosophy and science into Arabic. The flexibility of Arabic lent itself to the creation of a very precise scientific and technical vocabulary. Moreover, since this was a spoken language, the knowledge of ancient texts became theoretically available to anybody in the swiftly expanding Arab-speaking world. Within the span of a few centuries, the learning and accomplishments of different cultures spread far beyond their original geographical frontiers. There was also a dissemination of elements of popular culture. For instance, the Arabic Kalila-wa-Dimma collected fables from various places, including India. Arab scholars initially relied heavily on Greek works, but men such as Jaihani, Gardizi, and AlBiruni developed their own independent critical points of view. Abu Ri-han or Al-Biruni, a native of Khwarizm or Khiva (in modern Turkmenistan), was one of the greatest intellectuals of early medieval times. Only 40 of the 180 books he wrote have survived. Al-Biruni travelled to India to satisfy his curiosity about the land and its people, and to study their ancient texts in their original language. His Tahqiq-i-Hind covers a large number of topics including Indian scripts, sciences, geography, astronomy, astrology, philosophy, literature, beliefs, customs, religions, festivals, rituals, social organization, and laws. Apart from the historical value of his descriptions of 11th century India, Al-Biruni helped modern historians identify the initial year of the Gupta era. The Tahqiq-iHind states that the Gupta era began 241 years after the beginning of the Shaka era. Since the Shaka era began in 78 CE, this places the beginning of the Gupta era in 319–20 CE. Several Arabic geographical and travel accounts were written in the early medieval period. Some of these, such as the account of the traveller Sulaiman, refer to India. This is not surprising considering that both Arabs and Indians were actively involved in Indian Ocean trade. Such works throw light on trade and aspects of Indian political history. Persian was the language of royal courts and high culture in central and West Asia in early medieval times, and a number of Persian texts refer to India. The anonymous Chachnama describes how a Brahmana named Chach usurped the throne of Sindh in the mid-7th century and narrates the Arab conquest of that region by Muhammad bin Qasim. The Shahnama of Firdausi, a classic of Persian poetry, and the Gulistan by the famous poet Saadi, refer incidentally to aspects of Indian trade. Archaeology and the Early Indian Past

We turn now from texts to archaeology. Archaeology—the study of the human past though material remains—is closely connected with history. Material remains range from vestiges of grand palaces and temples to the small, discarded products of everyday human activity such as pieces of broken pottery. They include different things such as structures, artefacts, bones, seeds, pollen, seals, coins, sculptures, and inscriptions. Historians, anthropologists, and archaeologists understand ‘culture’ as something that includes all patterns of people’s learnt behaviour, the ways of thinking and doing things that they learn from the social group of which they are a part. Archaeologists also use the word culture in a more specific, technical sort of way connected with certain other important terms—artefact, industry, and

assemblage. An artefact is any portable object made or altered by human hands (e.g., pottery, tools). Similar artefacts made of the same material found at a site comprise an industry (e.g., a microlith industry, blade and burin industry). All the industries found at a site form its assemblage. If similar assemblages are found at several sites, these sites are said to belong to the same archaeological culture. Material evidence is a key to understanding human behaviour and experience. It is not enough to describe a stone tool or pot; the challenge is to get the stone tool or pot to tell their stories about the people who made and used them. As the products of craft traditions and part of the lifestyles of people, artefacts are rooted in specific cultural contexts. So, the narrow technical meaning of ‘culture’ in archaeology can be stretched to correspond to the wider meaning mentioned earlier. The rhythms and patterns of time based on material culture are generally slower and longer than those of historical events, and archaeological cultures do not coincide with the rise and fall of dynasties or kingdoms. Field archaeology deals with the exploration and excavation of sites. Sites are places where material remains of past human activity can be identified. In the plains, in areas where mud and brick were used for making houses, archaeological sites occupied by people for a very long time are often visible as mounds. Mounds get formed over the centuries due to the rebuilding of structures and the accumulation of rubbish, windblown sand, and other sediments. Sites are often discovered by sheer accident. They can also be discovered by using clues in literature, by regional or village surveys, or with the help of aerial photography. Sites buried underground can be detected by simple methods like inserting metal probes or rods into the ground. There are also the more sophisticated remote-sensing techniques such as LANDSAT imagery. Scanners of LANDSAT satellites create digital images of the earth’s surface and can help identify features such as ancient river courses, canals, embankments, and buried settlements. Archaeological evidence does not necessarily provide a complete picture of the material culture of ancient people. Artefacts found in the archaeological record generally consist of things that have been thrown away, lost, forgotten, hidden, or left behind (intentionally or unintentionally) by people when they moved elsewhere. Furthermore, not all material traits survive. Archaeological reconstruction depends on the amount and kind of material that is preserved, and this in turn depends on the objects themselves and on environmental factors, particularly soil and climate. Inorganic materials like stone, clay, and metal objects are most likely to survive in the archaeological record. Stone age people must have used tools of wood and bone as well, but it is the stone tools that have survived in large numbers. Tropical regions, with heavy rains, acidic soils, warm climates, and dense vegetation are not favourable for preservation. These things have to be kept in mind when assessing archaeological evidence. Sites can get destroyed by the forces of nature (e.g., floods, tectonic movements, volcanic eruptions), but they are more often destroyed by people when they clear land for farming or build houses, factories, roads, and dams. Sites can be explored by carefully examining what lies on the surface or they can be excavated, i.e., dug. Sites are not excavated just to see what they contain, but rather to uncover their stratigraphic sequence. The basic principle of stratigraphy is that if there are different layers, strata, or levels at a site, the lower ones are older. Of course, if a site gets disturbed, this principle does not apply. It is very important to know the stratigraphic context of artefacts, i.e., the precise level at which they were found, and what other kinds of things were found along with them.

Excavations can be horizontal (where a large surface area is exposed) or vertical (where the digging involves a small surface area), and are accompanied by careful recording, mapping, photographing, labelling, and preserving of artefacts. Recording is very important because excavation is destructive—some features of the upper layers have to be destroyed as archaeologists move from one layer to the next. Equally important is the publication of results, otherwise no one except the excavators will know what was discovered at the site. These days, an important trend within field archaeology is to try to understand sites within their larger landscape and context. Archaeologists are also increasingly moving towards non-destructive methods of investigation, such as remote-sensing and regional surveys. Regional surveys are conducted by walking over carefully selected sections of an area, observing the distribution and nature of surface features and finds. These are recorded and the surface finds collected. A great deal of valuable archaeological information can be gathered in this way.

THE HASTINAPURA MOUND

TABLE 1.1 THE CULTURAL SEQUENCE AT HASTINAPURA

P ERIOD DATE

V

CULTURAL TRAITS

Pottery—very different from earlier periods; coarse to medium-grained red ware; glazed Late wares with floral designs. Structures made of broken bricks from remains of earlier periods; 11th– four structural sub-periods identified. Many types of iron objects including nails, arrowheads, 15th spearheads, hoes, knife blades, etc. A stone image of Parvati and Rishabhadeva. Terracottas of centuries poor workmanship. Bangles of glass, ivory, shell, bone, etc. A coin of Balban (1266–87) from the middle level.

Site deserted

IV

Early 2nd century BCE– late 3rd century CE

Site deserted

III

Evidence of a massive fire Pottery—Northern Black Polished Ware (NBPW), coarse grey ware, unslipped red ware. Early 6th Houses of mud-bricks and kiln-burnt bricks (17.5 × 10 × 2.7 inches). Brick-lined drains. century– Terracotta ring wells. Copper objects. Iron arrowhead, chisel and sickle. Punch-marked and early 3rd uninscribed cast coins. Human and animal figurines (many of elephants) made of terracotta. century Beads of etched carnelian and crystalline quartz. Rings made of copper, chalcedony, gold, and BCE horn.

Site deserted

II

Evidence of a flood in the Ganga Pottery—Painted Grey Ware (PGW), black-slipped ware, and ordinary red and red-slipped ware. House walls of mud, mud-brick, reed, and mud plaster; one fragmentary burnt brick. c. 1100– Copper artefacts. Iron slag in the uppermost levels. Chert and jasper weights. Glass bangles. 800 BCE Terracotta objects including animal figurines. Bone needles. Charred grains of rice. Bones of horse, pig, cattle, etc.

Site deserted I

—— Pre1200 BCE

Natural Soil

Pottery—red ware, some with stamped designs; black-on-red painted pottery found in the upper levels. Houses mostly made of burnt bricks (14½ × 9 × 2½ inches); squarish bricks (11 × 11 × 4 inches) used for floors. Several house plans were reconstructed and seven structural sub-phases identified. Copper objects. Iron objects including nails, an axe/adze, sickle, and pan. A fine and varied range of moulded terracotta figurines (including many of the humped bull), wheels, carts, and votive tanks and a fine headless figure of the Bodhisattva Maitreya. Well-made rings and beads. Inscribed potsherds and a seal. Coins of the rulers of Mathura, the Yaudheyas, and imitation coins of the Kushana king Vasudeva.

Pottery—fragments of Ochre Coloured Pottery (OCP). No structures found, maybe because a very limited area was excavated. Habitation seems to have been sporadic ——

NOTE The mound of Hastinapura in Meerut district, Uttar Pradesh, was excavated in 1950–52 (see Lal, 1954–55). Its cultural sequence extended over an enormously long stretch of time, with four breaks in occupation. The earliest settlement belonged to the period before c. 1200 BCE and the latest level to the early 15th century CE. This table gives a brief synopsis of some of the main features of the various levels known as Periods I–V. Read the table from bottom to top, starting from

the lowest and earliest level, Period I. Note the range of evidence and the remarkable snapshot it gives of the life of people who lived at this site over the centuries. The cultural sequence at Hastinapura is a very important reference point for other sites in the upper Ganga valley.

BANGARAN I S LAND, LAKS HADWEEP: MARINE ARCHAEOLOGIS T AT WORK

While archaeologists generally work on land, marine or underwater archaeology is a rapidly growing area of study. In most other countries, marine archaeology deals mainly with shipwrecks. But in India, there are instances of entire cities that have been submerged by the sea. Marine archaeology involves many specialists such as oceanographers, geologists, geophysicists, and diverphotographers. It also requires the use of special equipment and scientific instruments. For instance, an echo-sounding system registers a rise when a boat passes over an underwater object. A side scan electronic system gives a view of the sea floor. Underwater metal detectors held by divers give a signal if they sense any kind of metal object between 3 and 4 m away. In recent times, exciting underwater discoveries have been made off the coast of Dwarka and Bet Dwarka in Gujarat. At Dwarka, there are remains of a submerged port-city, including fortification walls and stone anchors, perhaps going back to c. 1500 BCE.

ANCIENT SHIP ANCHOR

SCIENTIFIC TECHNIQUES IN ARCHAEOLOGY

Archaeologists increasingly rely on various scientific techniques in order to obtain precise information about the lives of past communities. These are especially useful in dating archaeological material. Many dating methods are based directly or indirectly on the principle of radioactive decay. Carbon-14 or radiocarbon dating is the best known of these, but others include thermoluminescence, potassium-argon, electron spin resonance, uranium series, and fission-track dating. The word archaeometry refers to a range of scientific techniques and analyses involving the use of measurement to analyse ancient objects or materials. The chemical analysis of pottery and metal artefacts can give clues about how they were produced. A comparison of the chemical composition of metal artefacts and ores can help identify the source of ores. Chemical analysis of soil can be used to determine the degree of human presence and activity at a site. For instance, the decomposition of animal excreta increases the nitrogen content of the soil. At the chalcolithic site of Inamgaon in Maharashtra, the soil in the courtyards had higher nitrogen content than that inside the house. This shows that people tied their animals in their courtyards. Palaeontology is the study of the remains of dead organisms over enormous spans of time. Within this discipline, molecular biology and DNA studies have been used to understand hominid evolution, to answer questions about what ancient people looked like, and to plot patterns of migration. Bones provide a great deal of information. The distribution of faunal remains (animal bones) at a site can indicate which areas were used for butchering, cooking, eating, bone tool making, and refuse dumping. Faunal analysis gives information about the animals people hunted and domesticated, the age of animals at death, and the diseases that afflicted them. The bones of wild and domesticated species can usually be differentiated. The joints of animals used for agriculture or draught purposes get fused and can be identified. Faunal remains can lead to inferences about aspects of environment such as climate, vegetation, and the season during which a site was occupied. Sometimes, bones reveal contacts between communities. For instance, the identification of marine fish bones and shells at Inamgaon—at least 200 km from the sea—shows that its inhabitants had contacts with coastal communities. The dental structure of humans is connected to subsistence patterns and methods of food preparation. Trace element analysis of human bones and scanning electron microscopic (SEM) analysis of tooth enamel can help identify the kind of food people ate and whether they suffered from nutritional deficiencies. Diseases such as arthritis and tuberculosis leave their mark on bones. Palaeo-pathology is the study of diseases ancient people suffered from by analysing their bones. Human bones are also examined to make inferences about population size, density, mortality, fertility, and life expectancy. Since food and nutrition are related to social standing, assessing the nutritional inputs in the bones of men and women at a site can indicate whether there were marked status differences between groups of people or between men and women. Of course, all the scientific techniques mentioned here require specialized laboratories, expensive equipment, and skilled specialists. FURTHER DISCUSSION

Radiocarbon dating

Discovered by an American chemist named Willard Libby in 1949, radiocarbon dating is today a very widely used dating method in archaeology. The atmosphere contains a fixed ratio of Carbon-12 (C-12, ordinary carbon) and Carbon-14 (C-14, a radioactive isotope of carbon). The latter is formed due to the influence of cosmic radiation on nitrogen in the atmosphere. Plants absorb C-14 in the atmosphere through their intake of carbon dioxide during the process of photosynthesis. C-14 passes into animals as they feed off plants or other animals. The intake of C-14 stops when the plant or animal dies, after which the C-14 in the physical structure of the organism begins to disintegrate at the rate of one half every 5,730 years (this is known as the ‘half-life’ of C-14). By measuring the amount of C-14 remaining in the organism, scientists can figure out when it died, i.e., how old it is. The radiocarbon method can be used to date various organic materials such as wood, charcoal, bone, and shell. Like all other scientific dating methods, the C-14 method provides approximate, not exact dates, and a standard error margin (known as the standard deviation) is recognized. Radiocarbon dates are accompanied by a plus/minus factor. Take the following date: 2500 ± 100 BP. This means a date range between 2600 and 2400 BP. ‘BP’ stands for ‘Before Present’, and the year 1950, which was about the time the radiocarbon method of dating was introduced, is taken as the base line, i.e., year one. Archaeologists sometimes resort to multiple dates from the same sample in order to arrive at mean dates with a smaller standard deviation. Sometimes radiocarbon dates can be way off the mark. This could be because the sample has got contaminated, or due to some procedural error. Scientists have known for some time that the amount of radiocarbon produced in the atmosphere has not been constant over time. They have also noticed a discrepancy between the more accurate dates produced by tree-ring dating and those arrived at by the radiocarbon method. Therefore, it is clear that some calibrations, i.e., corrections, have to be made while converting radiocarbon dates to calendar dates, i.e., BCE and CE dates. In view of the fact that there is still some debate regarding calibration procedures, some archaeologists prefer to publish uncalibrated dates. However, certain calibration tables have been more or less accepted by many scholars. Radiocarbon dates have certainly made a dramatic difference to our understanding of the chronology of ancient cultures. But why is it that radiocarbon dates for cultures given in different books are not always the same? This could be because some dates are calibrated, while others are uncalibrated. Another reason is that there is an element of interpretation and judgement involved even in the use of radiocarbon dates. When there is a string of radiocarbon dates for a site, which one is to be highlighted? Since radiocarbon dates give us a date bracket, which end of the bracket should be emphasized? Sometimes, instead of giving a whole string of radiocarbon dates with the standard deviation, archaeologists calculate the mean date and give that as a single radiocarbon date. There are thus choices to be made in the use of radiocarbon dates. How an archaeologist interprets and presents them depends on his/her larger understanding of the relative chronology of cultures.

Environments are not just backdrops to human activity; they are an important part of human experience. The relationship between people and their environmental landscape not only forms an important part of what people do but also of how they think about the world and their place in it. An understanding of the natural environment in which people lived is therefore an important aspect of prehistory, protohistory, and history. Archaeologists are increasingly becoming aware of the importance of the interactive relationship between environment and people. Environmental archaeology, which aims at understanding how societies adapted to their environment and how they used environmental resources, involves the collaboration of scientists and archaeologists. Palaeobotanical studies include the analysis of pollen and other minute plant remains, seeds, charcoal, sediments, and geological strata.

TABLE 1.2 SOME DATING METHODS USED IN ARCHAEOLOGY DATING METHOD

USED ON

TIME RANGE/LOWER TIME LIMIT (YA=YEARS AGO)

Carbon-14

Organic material, e.g., charcoal, wood, seeds, plant remains, bones

From 50,000 to 80,000 ya

Inorganic material that has been Thermoluminescence heated rapidly to 500ºC or above, e.g., pottery, terracotta, burnt flint

Even objects older than 50,000-80,000s ya

Potassium-Argon

Volcanic rocks older than about 100,000 years

Hundreds of millions of ya

Electron spin resonance

Bone, shell

Hundreds of thousands ya

Uranium series

Rocks rich in calcium carbonate

50,000–500,000 ya

Fission track

Certain kinds of rocks and minerals, obsidian, glass, mica, etc.

About 300,000 ya to millions of ya

Palaeomagnetic dating

Magnetized sediments, volcanic lava, Can only be used to date very old deposits from clay baked to 650–700ºC. hundreds of thousands ya to millions of ya

Amino acid analysis

Bone

Up to 100,000 ya

Dendrochronology (tree-ring dating)

Timber in areas outside the tropics

Up to about 8,000 ya

Optically stimulated or infrared stimulated luminescence

Any sediment which is believed to have been undisturbed after its burial under other sediments

Still undefined as refinements in the process continue; extends up to at least 17,000 ya; more accurate than C-14 calibrations for CE dates.

INTERPRETING ARCHAEOLOGICAL EVIDENCE

Interpretation is as crucial in archaeology as in using literary sources. It is involved at all levels, from the seemingly simple stage of classifying artefacts to the framing of historical hypotheses. Just as it is possible to identify trends in history writing, similarly, there have been several changes in approach and method within the disciple of archaeology. For example, in the 1960s, the traditional cultural history perspectives were challenged by the emergence of what came to be known as New

Archaeology and a school known as ‘processualism’. Closely allied with anthropology, this school tried to understand cultures and cultural processes holistically, especially in relation to ecology, human adaptation, and the interaction of different kinds of variables. It advocated a problem-oriented approach, emphasizing the importance of explanation, generalization, and theory building. The postprocessual school of archaeology, which emerged subsequently, challenged many of the assumptions, methods, and goals of processualism. Post-processualists question the possibility of objective knowledge about the past. Their understanding of material culture is also more complex. They point out that material culture can be used by social groups not only to reflect but also to disguise existing social relations. Archaeology usually provides an anonymous history, one that sheds light on cultural processes rather than events. It is the only source for prehistory, the longest part of the human past, during which many major discoveries and developments took place. It is also the only source for those parts of the past covered by non-deciphered written records, and continues to provide valuable information even after the beginning of the historical period. Unfortunately, once literary sources become available, historians tend to use archaeology as a secondary, corroborative source. One of the current challenges for early Indian history is to adequately incorporate archaeological evidence into the larger historical narratives. Archaeology often tells us about aspects of everyday life that are not revealed or emphasized in texts. It provides information on the history of human settlements and can give very specific details about modes of subsistence—the food people procured in order to live, and how they obtained it. It offers details about the crops people grew, the agricultural implements they used, and the animals they hunted and tamed. It is an excellent source of information on various aspects of the history of technology—raw materials, their sources, the methods used to make artefacts of various kinds. Archaeology also helps reconstruct routes and networks of exchange, trade, and interaction between communities. Cognitive archaeology, which deals with ways of thinking, beliefs, and religion, is a fastdeveloping area within archaeology. Although a large number of religious texts are available for ancient and early medieval India, an exclusively text-based view of religion will not tell us everything we want to know about religious practice. The material evidence of ancient religions can make a major contribution in this area. There are many problems involved in translating archaeological cultures into history. An archaeological culture need not necessarily correspond to a linguistic group, political unit, or a social group such as a lineage, clan, or tribe. One of the most important questions is how to explain changes in material culture, especially pottery traditions. This is an issue that has not yet been adequately addressed or understood in the context of ancient India. ETHNO-ARCHAEOLOGY

Ethnography is the study of living cultures and communities. Ethno-archaeology studies the behaviour and practices of living communities in order to interpret the archaeological evidence related to communities of the past. The Indian subcontinent is an area where many traditional features and methods survive—for instance in agriculture, animal husbandry, house building, the clothes people wear and the food they eat. Modern craftspersons are an important guide for understanding the ways in which ancient

craftspersons made things. Technology involves much more than the techniques used for making artefacts. It is necessary to explore the social organization of craftspersons, the customs and beliefs that material objects were part of, how goods were marketed, the relationship between craftspersons and traders, and between craftspersons and customers. Ethno-archaeology helps answer these sorts of questions as well. For instance, a tradition of carnelian bead manufacturing exists in Khambhat, in Gujarat, today. Studying modern bead making in this region gives valuable clues about the way in which the Harappan beads may have been made and the possible social organization of the bead makers. Ethno-archaeology can contribute towards filling the silences and gaps in history. For instance, it has helped archaeologists make inferences about women’s role in subsistence and craft-related activities in early times. Studies of modern communities of hunter-gatherers and shifting cultivators can help understand the life-ways of people who followed similar subsistence strategies in the past. Such studies have pointed out that tribal communities were never completely isolated, and they have also highlighted the important link between the ways in which people obtain their food and their identity as a community. Of course, ethno-archaeological evidence must be used cautiously, and it should be seen as suggesting possible and not necessarily conclusive ways of interpreting the archaeological data, always keeping in mind the differences between the present and past contexts.

HARAPPAN CARNELIAN BEADS

NEW DIRECTIONS IN RESEARCH

The social and cultural aspects of technology

Gundiyali and Lodai are two pottery manufacturing villages in Kutch, Gujarat. Archana Choksi’s case study explores the social and cultural aspects of technology and raises several important points that archaeologists and historians need to keep in mind when interpreting ancient pottery traditions: Pots 1. of different shapes, sizes, and forms are found in both villages. The form of vessels is connected to their specific function. For example, the mouth of a vessel used for storing dry material like grain and flour is wide so that it is easy to put a hand into it. Vessels used to carry water into fields have small mouths to minimize spillage. Cooking vessels have wide mouths to allow stirring and enlarged, thick

rims so that they can be handled when hot. Vessels used for eating are open and shallow, with rim bases that give them stability. The connection between the form and function of pots can help archaeologists interpret the function of the pots they find at sites. The 2. potters of Gundiyali and Lodai produce rather different vessels. This is because Gundiyali is dominated by farmers, labourers, and the service class, while Lodai is dominated by farmers and herders. These groups have different life-styles and needs and they use different kinds of pots. It is clear that potters make the sorts of pots their clients want, and consumer demand for pottery is shaped by occupation, family and community identity, food habits, and ritual practices. Inferences about patterns of social and economic organization can be made on the basis of the range of pots found at a site. The 3. potters of Gundiyali and Lodai are reluctant to experiment or change the forms and designs of the vessels they make. Pots change when there are significant socio-economic changes. For instance, the shapes of some of the traditional vessels have been modified to suit urban kitchens, although the decoration remains the same. This is relevant to understanding general patterns of continuity and change in ancient ceramic traditions.

SOURCE Choksi, 1995

F IGURE 1.3 P OTS FROM G UNDIYALI AND LODAI

PROTECTING SITES

The processes of rural and urban expansion pose constant threats to archaeological sites and their protection is crucial to the protection of the cultural heritage. Salvage archaeology aims at identifying endangered sites and saving them from destruction. Many decades ago, the site of Nagarjunakonda in the Guntur district of Andhra Pradesh was submerged in water when the Nagarjunasagar dam was built across the Krishna. Before this happened, between 1954 and 1960, officers of the Archaeological Survey of India thoroughly explored, excavated, and documented the valley. The next step was a massive salvage operation. Nine of the most important structures were transplanted and re-built on top of the Nagarjunakonda hill and on the banks of the reservoir. Replicas of 14 other structures were made. Apart from such spectacular sites and huge salvage projects, there are thousands of smaller sites all over the subcontinent that need to be noticed, documented, and cared for. Protecting the archaeological heritage is not just the responsibility of the Archaeological Survey or the Government. It is essential for ordinary people to realize the importance of protecting and cherishing these fragile links to the past.

NAGARJUNAKONDA SALVAGE OPERATIONS IN PROGRESS

Epigraphy: The Study of Inscriptions ANCIENT AND EARLY MEDIEVAL SCRIPTS

Inscriptions and coins come under the general umbrella of archaeology and archaeological sources, but they are subjects of specialized study in their own right. The study of inscriptions is known as epigraphy. An inscription is any writing that is engraved on something—stone, wood, metal, ivory plaques, bronze statues, bricks, clay, shells, pottery, etc. Epigraphy includes deciphering the text of inscriptions and analysing the information they contain. It also includes palaeography, the study of ancient writing. As mentioned earlier, the oldest inscriptions in the Indian subcontinent are in the yet undeciphered Harappan script. The oldest deciphered inscriptions belong to the late 4th century BCE, and are in Brahmi and Kharoshthi (sometimes spelt Kharoshti). These include those of the Maurya emperor Ashoka, which are in a number of different languages and scripts, but mostly in the Prakrit language and Brahmi script. As there are no obvious links between the Harappan script and Brahmi or Kharoshthi, what happened to writing in between remains a mystery. There is no direct mention of writing in Vedic literature, but references to poetic metres, grammatical and phonetic terms, very

large numbers, and complex arithmetical calculations in later Vedic texts are taken by some historians to indicate the possibility that writing may have been known at the time. The first definite literary references to writing and written documents occur in the Buddhist Pali texts, especially the Jatakas and the Vinaya Pitaka. Panini’s Ashtadhyayi refers to the word lipi (script). The Brahmi of Ashoka’s inscriptions seems a fairly developed script, and it must have had a prior history of at least a few centuries. Recently, important direct evidence that Brahmi existed in pre-Maurya times has come from Anuradhapura in Sri Lanka, where excavations unearthed potsherds with short inscriptions (probably names of people) that can be dated to at least the early 4th century BCE. There are three main types of scripts. In a logographic script, written symbols stand for a word, in a syllabic script for a syllable, and in an alphabetic script for a single phonetic sound. In the strict sense of the term, in an alphabet, the vowels should have a separate and fully independent status equal to that of consonants. Both the Brahmi and Kharoshthi scripts stand midway between alphabetic and syllabic scripts, and can be described as semi-syllabic or semi-alphabetic.

J. F. F LEET (1847–1917), ONE OF THE LEADING B RITISH EPIGRAPHISTS IN COLONIAL INDIA

Kharoshthi’s core area lay in the north-west—in and around the Indus, Swat, and Kabul river valleys, the land known as Gandhara in ancient times. Ashoka’s Shahbazgarhi and Mansehra inscriptions are in this script. Kharoshthi was later used in north India under the Indo-Greek, IndoParthian, and Kushana kings, and was also used in certain records outside the Gandhara area, including in parts of central Asia. Written from right to left, Kharoshthi seems to have been derived from the north Semitic Aramaic script.

D. C. SIRCAR (1907–85), A DISTINGUISHED EPIGRAPHIST AND SCHOLAR

The origins of Brahmi, a script written from left to right, are not as clear. Some scholars have suggested an indigenous origin, others an Aramaic origin. A problem in accepting the latter theory is that the direction of writing and the forms of the letters in Brahmi and Kharoshthi are different, so it is unlikely that they were derived from the same script. Kharoshthi declined and died out in about the 3rd century CE. Brahmi, on the other hand, became the parent of all the indigenous scripts of South Asia, and also of those used in parts of central and Southeast Asia. The different stages of the Brahmi script are often labelled on the basis of dynasties, e.g., Ashokan Brahmi, Kushana Brahmi, and Gupta Brahmi. The epigraphist D. C. Sircar identified three stages of development in the history of this script in northern India: early Brahmi (3rd–1st centuries BCE); middle Brahmi (1st century BCE–3rd century CE); and late Brahmi (4th–6th centuries CE). In the late 6th century, Gupta Brahmi evolved into a script known as Siddhamatrika or Kutila, which had sharp angles at the lower right hand corner of each letter. Regional differences became sharper after this point of time. The modern north Indian scripts gradually emerged out of Siddhamatrika. Nagari or Devanagari was standardized by about 1000 CE and an eastern script (known as proto-Bengali or Gaudi) took shape between the 10th and 14th centuries. From here, it was a short step to the emergence of the Bengali, Assamese, Oriya, and Maithili scripts in the 14th–15th centuries. This is also the time when the Sharada script emerged in Kashmir and adjoining areas. The earliest inscriptions in the Tamil language (with some Prakrit elements) are engraved in rock shelters and caves, mostly in Tamil Nadu, especially in the area near Madurai. They are in a script known as Tamil–Brahmi, an adaptation of Brahmi for writing the Tamil language. Iravatham Mahadevan (2003) has identified two phases in the evolution of the Tamil–Brahmi script—early Tamil-Brahmi (c. 2nd century BCE–1st century CE) and late Tamil–Brahmi (2nd–4th centuries CE). Three southern scripts emerged in the early medieval period—Grantha, Tamil, and Vatteluttu. The first of these was used for writing Sanskrit, the second and third for writing Tamil. These three scripts may have emerged out of southern varieties of Brahmi; or they may have emerged from some other earlier southern scripts. The Tamil script first appeared in the Pallava territory in the 7th

century CE. Something similar to the modern Telugu and Kannada scripts took shape in the 14th–15th centuries, while the Malayalam script developed out of Grantha at about the same time. Ancient Indian inscriptions include a few bi-script documents, in which the text is given in the same language written in two different scripts. Most of the instances come from the north-west and consist of short bi-script Brahmi–Kharoshthi inscriptions. The longer records include an 8th century Pattadakal pillar inscription of the Chalukya king Kirttivarman II. The language is Sanskrit; the text is written both in the north Indian Siddhamatrika script and in the local southern proto-Telugu–Kannada script. LANGUAGES OF ANCIENT AND EARLY MEDIEVAL INSCRIPTIONS

The earliest Brahmi inscriptions, including those of Ashoka, are in dialects of Prakrit (also known as Middle Indo-Aryan). Between the 1st and 4th centuries CE, many inscriptions were written in a mixture of Sanskrit and Prakrit. The first pure Sanskrit inscriptions appeared in the 1st century BCE. The first long Sanskrit inscription is the Junagadh rock inscription of the western Kshatrapa king Rudradaman. By about the end of the 3rd century CE, Sanskrit had gradually replaced Prakrit as the language of inscriptions in northern India. PRIMARY SOURCES

Deciphered and undeciphered scripts

The story of the decipherment of ancient scripts is an exciting one. Ashokan Brahmi was deciphered as a result of the slow, painstaking efforts of a number of administrator-scholars working in India as employees of the East India Company. They included Charles Wilkins, Captain A. Troyer, W. H. Mill, J. Stevenson, and James Prinsep. These scholars first tried to read early medieval Brahmi inscriptions and then worked at deciphering the older Brahmi letters. The final step in the decipherment of the 3rd century BCE Maurya Brahmi was made by Prinsep in 1837. Even though Prinsep managed to read these inscriptions, he had no idea about the identity of the king Piyadassi mentioned therein. The answer came soon enough, when George Turnour, an officer of the Ceylon civil service, identified the king as Ashoka on the basis of references in the Pali chronicle, the Dipavamsa. Prinsep also played a role in the decipherment of Kharoshthi, along with other scholars such as Christian Lassen, Charles Masson, Alexander Cunningham, and E. Norris. The decipherment of Kharoshthi was easier because of the availability of bi-script coins in Greek and Kharoshthi issued by the Indo-Greek kings. Apart from the Harappan script, there are some other scripts that are still undeciphered or difficult to read. These include an elaborate, calligraphic variation of Brahmi known as ornate or ornamental Brahmi, found on short inscriptions in various parts of the country. Another stylized, ornate form of the Brahmi script, referred to by scholars as Shankhalipi (because its characters

look like shankhas, i.e., conch shells) is found in inscriptions of the 4th–8th centuries CE in various parts of India except the far south. Both ornate Brahmi and Shankhalipi seem to have been used mainly for names and signatures. There is a script similar to Brahmi on terracotta seals at sites such as Chandraketugarh and Tamluk in eastern India. An undeciphered script similar in some ways to Kharoshthi has been found in Afghanistan. SOURCE Salomon, 1998 Kharoshthi Script Vowels

Consonants

Brahmi Script Vowels

Consonants

The development of some Brahmi letters

Development of the letter ṇa in Brahmi and its derivative scripts

F IGURE 1.4 KHAROSHTHI AND B RAHMI SCRIPTS

In the Deccan and South India, Sanskrit inscriptions appeared along with Prakrit ones in the late 3rd/early 4th century CE, for instance at Nagarjunakonda in Andhra Pradesh. The Sanskrit element gradually increased. In the transitional phase of the 4th and 5th centuries, there were bilingual Sanskrit–Prakrit inscriptions, as well as those in a mixture of the two languages. Thereafter, Prakrit fell into disuse. Between the 4th and 6th centuries, Sanskrit emerged as the premier language of royal inscriptions all over India. Thereafter, it attained the status of a language associated with high culture, religious authority, and political power not only in the subcontinent but also in certain other areas such as Southeast Asia. However, in the post-Gupta period, there was also an important parallel trend towards the evolution of regional languages and scripts. Even Sanskrit inscriptions show the influence of local dialects in spellings and words of non-Sanskrit origin. In South India, inscriptions in the old Tamil language (and the Tamil–Brahmi script) appeared in the 2nd century BCE and the early centuries CE. Tamil became an important language of South Indian inscriptions under the Pallava dynasty. There are examples of bilingual Tamil–Sanskrit Pallava inscriptions from the 7th century onwards. In these, the invocation, genealogical portion, and concluding verses are often in Sanskrit and the details of the grants in Tamil. Kings of the Chola and Pandya dynasties also issued Tamil and bilingual Sanskrit–Tamil inscriptions. Hundreds of donative Tamil inscriptions were inscribed on temple walls in various parts of South India in early medieval times. The earliest Kannada inscriptions belong to the late 6th/early 7th century CE. From this period onwards, there were many private donative records in Kannada, and this language was also used in

some royal grants. There are some bilingual Sanskrit– Kannada inscriptions and a 12th century inscription found at Kurgod (in Bellary district, Karnataka) is in three languages—Sanskrit, Prakrit, and Kannada. The late 6th century epigraphs of the early Telugu Chola kings mark the beginnings of Telugu as a language of inscriptions. Thereafter, there are many private donative records in this language. Malayalam inscriptions appeared in about the 15th century. There are also a few late inscriptions in Tulu, a Dravidian language which is similar in some ways to Kannada and is spoken in parts of Karnataka. As for inscriptions in the modern north Indian (New Indo-Aryan) languages, Marathi and Oriya inscriptions can be identified from the 11th century. Inscriptions in dialects similar to what is referred to today as Hindi appear in Madhya Pradesh from the 13th century onwards, and Gujarati can be identified in epigraphs from the 15th century. DATING THE INSCRIPTIONS

Inscriptions are usually dated in regnal years or eras. The dates of eras are given in words, numerals, or both. The ancient Indian calendar system often had a combination of lunar as well as solar units. Inscriptions sometimes specify the month, lunar fortnight (paksha), lunar day (tithi), weekday (the civil day or solar day), and may give additional astronomical details. The specification of the year and day began in the 2nd century BCE. Some later inscriptions give the date in the form of chronograms. Instead of numbers, words standing for these numbers are used—e.g., bhumi (the earth) = 1; kara (hand) = 2; loka (the worlds) = 3; veda = 4, etc. These words are given in the reverse sequence of the numbers in the date, and are to be read backwards. For example, ‘kara-veda-bhumi’ means the year 142. If an inscription is not dated, it can be assigned a rough date on palaeographic grounds. Many different eras were used in ancient and early medieval India. To cite a few examples—the Vikrama era of 58 BCE, the Shaka era of 78 CE, the Kalachuri-Chedi era of 248 CE, and the Gupta era of 319–20 CE. The Kollam era of 824 CE was used in inscriptions of Kerala and adjoining parts of Tamil Nadu, while the Chalukya-Vikrama era of 1076 was used in some inscriptions of Karnataka and adjoining areas. The eras marked important events, usually the accession of a king. Subordinate kings used the era of their overlord, and some eras continued to be used long after their founding dynasty had disappeared. While the initial year of most ancient and early medieval eras is known, uncertainty still surrounds a few. For instance, the suggested dates for the beginning of the Harsha era include 612, 619, and 648 CE. Similarly, the dates for the era of the Ganga kings of Orissa range from the 4th to the 9th century CE. FURTHER DISCUSSION

How to convert ancient era dates into modern ones

How do you convert a date in an ancient era into BCE/CE dates of the Common Era, which is based on the Christian calendar? All that is required is a bit of simple arithmetic. For a date in an era that began in a BCE year, subtract the initial BCE year of that era from it. If the era began in a CE year, add the initial CE date.

For example, year 179 of the Vikrama era (which began in 58 BCE) = 179 - 58 = 121 CE; year 179 of the Shaka era (which began in 78 CE) = 179 + 78 = 257 CE. There can be a bit of variation in the conversion of ancient dates, depending on whether the months mentioned are solar or lunar months. The month is also relevant because the traditional Indian year did not begin in the same month as the Western year, which begins in January. Another point that can create some confusion is whether the year mentioned in the inscription is to be understood as expired or current; this is sometimes, but not always, indicated in the text. To give an example, when we celebrate a child’s first birthday, going according to expired years, he has completed one year, but going by current years, he has begun his second year of life. In spite of these kinds of issues, if an inscription is dated in a known era, it is possible to pin it down within a very narrow margin. THE CLASSIFICATION OF INSCRIPTIONS

Inscriptions can be classified in several different ways, for instance according to the surface they are engraved on, language, age, and geographical region. They can also be classified into official and private records, depending on whose behalf they were inscribed. Ashoka’s edicts and royal land grants are examples of official records. Inscriptions recording grants made by private individuals or guilds to temples, or to Buddhist or Jaina establishments are examples of private records. Inscriptions can also be classified according to their content and purpose into types such as donative, dedicative, and commemorative inscriptions. For instance, the Lumbini pillar inscription of Ashoka is a royal commemorative inscription, recording a specific event—the visit of the king to the Buddha’s birth-place. In many parts of India, there is evidence of an ancient practice of erecting memorials to dead people. Thousands of memorial stones are found all over the country, not always connected with burials. Some only have sculpted scenes (realistic or symbolic), others also have inscriptions. The most common memorial stones were erected in memory of dead heroes or women who committed sati. But there are other kinds as well. Stones were set up in honour of Jaina men and women who gave up their lives in the exemplary Jaina fashion of death by starvation. On the Konkan coast, many stones were erected in memory of sailors who lost their lives in sea battles. Some memorial stones were worshipped. Donative inscriptions in favour of religious establishments were inscribed on shrine walls, railings, and gateways. The excavation and donation of caves to ascetics was recorded in inscriptions in the caves. Donative inscriptions include records of the installation of religious images, often inscribed on the images themselves. Others record investments of money made by people, out of the interest of which lamps, flowers, incense, etc. were to be provided for the worship of the deity.

A P ALA PERIOD IMAGE WITH A DONATIVE INSCRIPTION ON THE BASE; COPPER PLATE INSCRIPTIONS

Royal land grants are an important category of donative records. There are thousands of such inscriptions, some on stone, but mostly inscribed on one or more copper plates. Most of them record grants made by kings to Brahmanas and religious establishments. The earliest stone inscriptions recording land grants with tax exemptions are Satavahana and Kshatrapa epigraphs found at Nashik. The mid-4th century Pallava and Shalankayana grants are the earliest surviving copper plate grants. One of the oldest copper plate grants from north India is the late 4th century CE Kalachala grant of

king Ishvararata. Copper plate grants increased in number and frequency in the early medieval period. PRIMARY SOURCES

Memorializing death in stone

Memorial stones and their inscriptions reflect the values and ideals that ancient communities associated with life and death. In the Andhra region, such stones are known as chhaya stambhas. At Nagarjunakonda, there are memorial stones in memory of kings, queens, soldiers, chieftains, generals, holy people, and an artisan. At the base of a 12 ft high limestone pillar is an inscription recording the names of 29 royal women—sisters, mothers, and queens of the Ikshvaku family, collectively mourning king Chantamula I. Above the inscription are five panels of relief carving, one on top of the other, depicting the dead king in different poses. In the first (lowest) scene, he appears as a plump figure distributing gifts during the performance of a religious ceremony. In the next one, he is riding an elephant. In the panel above this, he is surrounded by women, three seated on the floor (perhaps musicians), the fourth dancing. In the next scene, he is sitting on a throne, flanked by women, two of whom may represent his queens. The topmost panel depicts a building, possibly a palace or heaven. The Nagarjunakonda memorial pillar in honour of an artisan is naturally much simpler. It just gives the name of the artisan Mulabhuta and states that he came from a place called Pavayata. Above the inscription is a narrow-necked vase, which may have been the emblem of the guild to which Mulabhuta belonged. The largest concentration of memorial stones is in Karnataka. About 2,650 hero stones dated between the 5th and 13th centuries have been found here. Inscriptions on some of these give only a name, others offer details of the circumstances in which the person died. Hero stones usually commemorated male heroes, but two inscriptions from Siddhenahalli and Kembalu refer to the heroic death of a woman and of a queen who launched a cattle raid. An inscription from Shikaripur refers to a woman laying down her life to defend others. There are also some interesting memorials for pets. An inscription from Gollarahatti is in memory of a hunting dog named Punisha who died after killing a wild boar, while another one from Atkur commemorates the death of a dog named Kali who died fighting a wild boar during a hunt. A 12th century inscription at Tambur mourns the death of the pet parrot of a king of the Kadamba dynasty of Goa. The parrot was eaten by a cat in the palace and the inscription tells us that the king was so filled with grief at this event that he killed himself. The tradition of memorial pillars lives on in certain parts of the country today, e.g., in Karnataka and among tribal communities of Gujarat and Madhya Pradesh. The Maria and Muria Gond tribes of the Bastar region of Madhya Pradesh still erect memorials of stone and wood. Some are plain, others are beautifully carved or painted. These memorials are linked with beliefs and

rituals related to death and afterlife, and are a very important part of the cultural life and identity of the people. SOURCE Settar and Sontheimer, n.d.; Postel and Cooper, 1999

HERO STONE FROM KHANAPUR, KARNATAKA

Royal inscriptions include prashastis (panegyric). Most royal inscriptions (and some private ones too) usually begin with a prashasti, but some inscriptions are entirely devoted to eulogizing their subject. Well-known examples are the Hathigumpha inscription of Kharavela, a 1st century BCE/1st century CE king of Kalinga in Orissa, and the Allahabad prashasti of the 4th century Gupta emperor Samudragupta. Certain inscriptions record the building of waterworks, wells, and charitable feeding houses by private individuals. A series of unique records of royal initiatives of this kind are inscribed on a granite rock at Junagadh (Girnar) in Gujarat. Apart from a set of Ashokan edicts, this rock bears two other important inscriptions. A 150 CE inscription of the Shaka ruler Rudradaman records the beginning of the construction of a water reservoir known as Sudarshana lake in the 4th century BCE during the time of the Maurya emperor Chandragupta, its completion during the reign of Ashoka, and its repair in the 2nd century CE. A 5th century inscription on the same rock, of the time of the Gupta king Skandagupta, describes how the lake burst its banks due to excessive rains and was repaired after two years’ work. What we have here is an amazing history of the building and repair of an ancient water reservoir over a period of about 1,000 years! There are other miscellaneous types of inscriptions—labels, graffiti left by pilgrims and travellers, religious formulae, and writing on seals. Certain inscriptions from Madhya Pradesh give

a condensed summary of the basics of Sanskrit grammar. ‘Footprint inscriptions’ are found in many parts of the country, accompanying a pair of engraved footprints of a holy man, king or other noteworthy person. INSCRIPTIONS AS A SOURCE OF HISTORY

Compared with manuscripts of texts, inscriptions have the advantage of durability. They are usually contemporaneous to the events they speak of and their information can be connected to a time and place. Changes and additions made to them can usually be detected without great difficulty. The text of inscriptions may be brief, but a large number of short inscriptions can often provide important historical information. Compared to literary sources, which tend to give a theoretical perspective, inscriptions often reflect what people were actually doing. And although epigraphs of different categories usually follow a standard format, some of them do have the ability to surprise. Inscriptions are a valuable source of information on political history. The geographical spread of a king’s inscriptions is often taken as indicating the area under his political control. But the discovery of inscriptions depends on chance and not all the inscriptions inscribed during a king’s reign need necessarily be found. Furthermore, moveable inscriptions are not always found in situ, i.e., in their original place. The earliest royal inscriptions do not contain much genealogical material, but later ones generally do. Their prashastis give details about the history of dynasties and the reigns of kings. Of course, there are problems. Royal inscriptions naturally tend to exaggerate the achievements of the ruling king. Sometimes, confusion is created when a genealogy mentions kings with the same name, or when different inscriptions contradict each other on particular details. Sometimes genealogies skip names. This kind of skipping occurs, for instance, in the case of Skandagupta and Ramagupta, who are ignored in Gupta genealogies because they did not come within the direct line of succession of later rulers. There are cases where inscriptions of different dynasties make conflicting claims. For instance, a Gurjara-Pratihara inscription states that king Vatsaraja conquered all of Karnataka. However, the contemporary Rashtrakuta king claims in his inscriptions to have defeated Vatsaraja and to have ruled over the Karnataka area. Wherever possible, details of political events given in inscriptions have to be cross-checked. Inscriptions, especially those of the early medieval period, have been used as a major source of information on political structures and administrative and revenue systems. They can also shed light on the history of settlement patterns, agrarian relations, forms of labour, and class and caste structures. Analysing epigraphic evidence involves unravelling the technical vocabulary of inscriptions—for instance, the designations of officials, fiscal terms, and land measures—the meanings of which are not always clear. There are very few ancient records of secular land transactions and records of land disputes, but these take us straight to the heart of social and economic issues. For instance, an inscription of the time of the Chola king Rajaraja III (1231 CE) states that farmers of a certain village found the burden of arbitrary levies in money and paddy and the demand of compulsory labour made on various pretexts by several agencies so unbearable that they could no longer carry on cultivation. A meeting of the Brahmana assembly and the leading men of the locality was held in the village temple.

Decisions were taken, fixing the dues that farmers were to give to the Brahmanas and royal tax collectors, and the labour services that they were expected to perform.

An ancient theatre, an ancient love story

The Sitabenga and Jogimara caves on Ramgarh hill (in Chhattisgarh) can be reached through a natural tunnel known as Hathipol, 180 ft long and so high that an elephant can pass through it. Both caves have inscriptions in a Prakrit dialect, engraved in Brahmi letters of the 3rd century BCE. In front of the entrance of the Sitabenga cave is a row of rock-cut benches arranged in terraces in the shape of a crescent, with aisles. The two-line inscription in the cave cannot be read clearly or fully. It seems to talk of venerable poets who kindled the heart of others with their poetry, and people tying garlands of jasmine flowers around their necks at the swing festival of the full moon, when there was much fun, frolic, and music. The inscriptions and the layout of the cave and the area around it suggest that this may have been an ancient theatre, a place where poets recited their poems and where plays were performed long ago.

The Jogimara cave lies to the south of Sitabenga. Here, there is a five-line inscription which can be translated thus: ‘Sutanuka by name, a devadasi. The excellent among young men, Devadinna by name, the rupadaksha, loved her.’ In later times, the word devadasi referred to a temple woman, but its meaning in this early context is uncertain. Rupadaksha can be interpreted in different ways—it could mean someone skilled in sculpture, or a scribe, or an officer connected with coinage. But there are paintings on the roof of the cave, so maybe the word means painter or artist.

However, there is another possible translation of the Jogimara inscription: ‘Sutanuka by name, a devadasi, made this resting place for girls [perhaps actresses who performed in the dramas enacted here]. Devadinna by name, skilled in painting, made the paintings in this cave.’ The Jogimara cave inscription can thus be interpreted in two very different ways. The first interpretation conveys raw emotion in its startling brevity, while the second one is more matter of fact. SOURCE Annual Report of the Archaeological Survey of India, 1903–04: 123–31

Inscriptions provide dateable information on the history of religious sects, institutions, and practices. Donative records help identify the sources of patronage enjoyed by ancient religious establishments. They also give glimpses into sects and cults that were once important but did not leave any literature of their own— e.g., the Ajivika sect and the yaksha and naga cults. Inscriptions can help identify and date sculptures and structures, and thus throw light on the history of iconography, art, and architecture. They are also a rich source of information on historical geography. In fact, the location of several ancient Buddhist monastic sites such as Kapilavastu (identified with Piprahwa in Basti district, UP) has been fixed on the basis of inscribed monastic seals. Inscriptions reflect the history of languages and literature and a few refer to the performing arts. For instance, the 7th century Kudumiyamalai inscription gives the musical notes used in seven classical ragas. Inscriptions from Tamil Nadu refer to the performance of various kinds of dances. The pillars of the eastern and western gateways of the Nataraja temple at Chidambaram have label inscriptions describing the dance poses of 108 sculpted figures carved on them, quoting verses from the Natyashastra of Bharata. Inscriptions are material remains and have to be understood in relation to the larger contexts in which they are found. They are also texts, connected with prevailing structures of power, authority, and social status. Whether fragmentary or complete, whether consisting of one word or hundreds of lines, an inscription has to be read and analysed carefully. Its contents can then be compared with those of other inscriptions and with information from other kinds of sources. Numismatics: The Study of Coins

In modern times, money functions as a medium of exchange, a store of value, a unit of accounts, and a medium of deferred payment. In its most general sense, money is any item that is accepted by a community for the exchange of goods or services or for the discharge of debt. Currency and coinage are more specific terms. Currency is a medium of exchange backed by an issuing authority, one that can be used to immediately discharge any kind of financial obligation. Coinage is metal currency. It has a definite size, shape, and weight standard, and bears the stamp of an issuing authority. The main message-bearing side of a coin is known as the obverse and the other side the reverse. In the world context, the earliest coins appear in Lydia in West Asia in c. 700 BCE and were made of electrum, a natural alloy of gold and silver. Numismatics or the study of coins includes the analysis of the material out of which coins were made; the identification of the sources of the metals; the classification and study of the form of coins on the basis of their fabric (size, shape, thickness, design, workmanship), metrology (weight), design, metallic composition, techniques of manufacture, and message content. Ancient coins are usually discovered by accident. A very small proportion finds its way into the hands of coin collectors or governments; the majority end up getting lost, melted down, or destroyed. Coins occur as stray individual finds or as part of coin hoards. Hoards are especially valuable for monetary history and consist of coins withdrawn from human custody (due to being buried underground for safety, or fire, floods, loss, etc.) and found subsequently. Metrology—the measurement and arrangement of coins by weight—is an important aspect of numismatics. In the course of circulation, coins are subjected to wear and tear and their weight gradually decreases. This fact enables numismatists to arrange them in a chronological sequence and to distinguish between coins of a hoard that have been in circulation for greater and less periods of time. Various techniques are used for ascertaining the metal content of coins. One method is to carefully inspect their colour and lustre. There are other informal physical procedures such as testing for resonance by dropping the coin on a hard surface to produce a sound or testing its ductility by biting it. A water displacement test can be conducted to measure a coin’s specific gravity. There are also several chemical testing procedures for ascertaining metal composition. These are more accurate but generally damage the coin. Non-destructive scientific techniques such as X-ray fluorescence (XRF) spectrometry, which are now being used to analyse the elemental composition of coins, provide quick and accurate results. Mint towns can be identified by noting sites where large numbers of coin moulds have been found. An analysis of coin dies can help identify the number and sequence of issues and estimates of the volume of coins produced by these dies can be made by extrapolation. A BRIEF HISTORY OF INDIAN COINAGE

Stone age people had neither currency nor coinage and conducted exchange via barter. Chalcolithic cultures too conducted trade without the use of coins. The Harappans, for instance, had a very extensive trade network based on barter. The Rig Veda mentions words such as nishka and nishkagriva (gold ornaments), and hiranya-pinda (gold globules), but these cannot be understood as coins. Later Vedic texts use terms such as nishka, suvarna, shatamana, and pada. These may have been metal pieces of definite weight, not necessarily full-fledged coins. The earliest definite literary and archaeological evidence of coinage in the Indian subcontinent dates from the 6th–5th centuries BCE in a context of the emergence of states, urbanization, and expanding

trade. Buddhist texts and the Ashtadhyayi refer to words such as kahapana/karshapana, nikkha/nishka, shatamana, pada, vimshatika, trinshatika, and suvanna/suvarna. The basic unit of Indian coin weight systems was a red-and-black seed of the gunja berry (Abrus precatorius) known as the raktika, ratti, or rati. In South India, the standard weight of coins was theoretically calculated on the basis of the relationship between two kinds of beans—the manjadi (Odenathera pavonina) and the kalanju (Caesalpinia bonduc). The advent of coinage did not mean the disappearance of barter—both co-existed for a very long time.

RATTI SEEDS

The oldest coins found in the subcontinent are punch-marked coins, made mostly of silver, some of copper. They are usually rectangular, sometimes square or round. The blanks for making these coins were generally cut from a metal sheet or made from flattened metal globules. The symbol or symbols were then hammered on separately, using dies or punches. These coins are often irregular in shape, their corners sometimes snipped off to adjust their weight. Most of the silver punch-marked coins weighed 32 rattis or about 56 grains (grain is a weight measure used for metals; 1 grain = 64.79 mg). Punch-marked coins are found all over the subcontinent, and continued to circulate in many places till the early centuries CE, with a longer period of circulation in peninsular India.

SILVER PUNCH-MARKED COIN OF MAGADHA

The punch-marked coins of northern India can be divided into four main series on the basis of their weight, the number and nature of punch marks, and their area of circulation—the TaxilaGandhara type of the north-west with a heavy weight standard and a single punch type; the Kosala type of the middle Ganga valley, with a heavy weight standard and multiple punch marks; the Avanti type of western India, with a light weight standard and single punch mark; and the Magadhan type with a light weight standard and multiple punches (Mitchiner, 1973). Changes in coinage patterns mirrored political changes. With the expansion of the Magadhan empire, the Magadhan type of punch-marked coins came to gradually replace those of other states. Although these coins do not have any legends (i.e., anything written on them), it is likely that most of them were issued by states. In later times, there is evidence of city issues and guild issues, and it is possible that this practice also prevailed in the period of the punch-marked coins. Symbols on these coins include geometric designs, plants, animals, the sun, wheel, mountain, tree (including treein-railing), branches, and human figures. Some symbols may have had a religious or political importance, but their precise significance is not always certain. The coins often have primary and secondary punch marks. The latter are ‘counterstamps’ or ‘countermarks’ which were added on later, without heating the coins.

UNINSCRIBED CAST COPPER COIN OF KAUSHAMBI

Uninscribed cast coins made of copper or alloys of copper appeared soon after the punch-marked coins. They have been found in most parts of the subcontinent except the far south. Some types have a fairly wide distribution, while others (such as those found at Ayodhya and Kaushambi, which seem to have been issued in the late 3rd or early 2nd century BCE) have a more restricted range of circulation. These coins were made by melting metal and pouring it into clay or metal moulds. Clay moulds have in fact been found at many sites and a bronze mould was found at Eran in central India. The discovery of punch-marked and uninscribed cast coins in the same archaeological level at some early historical sites indicates that they overlapped in time. Other early Indian coin types include uninscribed die-struck coins, mostly in copper, rarely in silver. The symbols, some similar to those on the punch-marked coins, were struck onto coin blanks with metal dies that were carefully carved with the required designs. The minting of such coins may have begun in about the 4th century BCE and they have been found in large numbers at sites such as Taxila and Ujjain. The next stage in the history of Indian coinage is marked by the die-struck Indo-Greek coins of the 2nd/1st century BCE. These are very well-executed, usually round (a few are square or rectangular) and mostly in silver (a few are in copper, billon [a silver–copper alloy], nickel, and lead). They bear the name and portrait of the issuing ruler on the obverse. Coins of Menander and Strato I show them slowly aging from teenagers to old men, indicating their long reigns. Coins issued jointly by kings reflect the practice of conjoint rule. The reverse of the coins usually had religious symbols. The Indo-Greeks issued bilingual and bi-script coins, the name of the issuer appearing on the obverse in Greek and on the reverse in the Prakrit language and usually in the Kharoshthi script (rarely in Brahmi). The coins of these kings also have certain symbols referred to as monograms by numismatists, the precise significance of which is not certain. Coins of the Shakas, Parthians, and Kshatrapas follow the basic features of Indo-Greek coinage, and include bilingual and bi-script issues.

SILVER COIN OF INDO-G REEK KING DEMETRIUS

The Kushanas (1st–4th centuries CE) were the first dynasty of the subcontinent to mint large quantities of gold coins; their silver coins are rare. They also issued many copper coins of low denominational value, which indicates the increasing spread of the money economy. Kushana coins have the figure, name, and title of the king on the obverse. On the reverse are deities belonging to the Brahmanical, Buddhist, Greek, Roman, and other pantheons. The legends are either entirely in Greek, or in some cases in Kharoshthi on the reverse. A number of coin types ranging from the 3rd century BCE to the 4th century CE, referred to by numismatists as indigenous, tribal, janapada, or local coins form an important source of information on the history of the dynasties of northern and central India. These coins are mostly cast or die-struck in copper or bronze, but there are some silver coins and a few rare examples of ones in lead and potin (an alloy of copper, lead, tin, and dross). They include those issued by chieftains, kings, and non-monarchical states such as the Arjunayanas, Uddehikas, Malavas, and Yaudheyas. There are also coins bearing the name of cities such as Tripuri, Ujjayini, Kaushambi, Vidisha, Airikina, Mahishmati, Madhyamika, Varanasi, and Taxila, presumably issued by the administration of these cities. Some coins with the word negama seem to represent coins issued by merchant guilds. Certain Taxila coins with the legend pancha-nekame may have been issued jointly by five guilds.

G OLD COIN OF KUSHANA KING VIMA KADPHISES

In the Deccan, the pre-Satavahana coinage was followed by the copper and silver coins of the Satavahana kings. Rulers of this dynasty also issued coins of small denominational value made of lead and potin. Most Satavahana coins were die-struck, but there are some cast coins, and a combination of techniques was also used. The legends were generally in the Prakrit language and Brahmi script. However, the portrait coins (mostly in silver, but also in lead) use a Dravidian language and Brahmi script. Punch-marked coins continued to circulate alongside the Satavahana issues. There was a greater demand for silver currency in the western Deccan, perhaps due to commercial reasons. The Kshatarapa ruler Nahapana introduced a silver currency in the Nashik area. Roman gold coins also flowed into peninsular India in large quantities in the early centuries CE and may have been used as a medium of exchange for large-scale transactions or as currency reserves and capital deposits. Locally made imitations of Roman gold coins have also been found. So, in the early centuries CE in the western Deccan, there was a co-existence of Satavahana, Kshatrapa, punch-marked, and Roman coins. Currencies of the western Deccan also flowed into the eastern Deccan. Some of the punch-marked coins found in various parts of South India have been identified as dynastic issues on the basis of their symbols. For example, coins found in a hoard at Bodinaikkanur near Madurai had a double carp fish—the symbol of the Pandya kings. In recent years, there has been increasing evidence of dynastic issues (some with portraits) with legends of the Cholas, Cheras, and Pandyas. This evidence has come from private collections and as surface and stray finds, but not so far in stratified archaeological contexts. Coins with the legend Valuti have been assigned to the Pandyas. Silver coins with the portrait of a Chera king and the legend Makkotai have been found in

the Krishna riverbed near Karur. There are also coins with the legends Kuttuvan Kotai and Kollippurai along with the Chera symbols of the bow and arrow and the double fish and tiger. The imperial Gupta kings issued well-executed die-struck gold coins with metrical legends in Sanskrit. Known as dinaras, these coins have been mostly found in north India. The obverse depicts the reigning king in various poses, usually martial ones, but there are interesting instances of coins of Samudragupta and Kumaragupta I showing them playing the vina (a stringed instrument). The reverse of the Gupta coins have religious symbols indicating the kings’ religious affiliations. There was a decline in the metallic purity of gold coins in the later part of Skandagupta’s reign. The Guptas also issued silver coins, but their copper coins are rare.

G OLD COIN OF G UPTA KING KUMARAGUPTA I

In the post-Satavahana period in the eastern Deccan, the Ikshvakus of the lower Krishna valley (3rd–4th centuries) issued lead coins similar in fabric to the Satavahana ones. Some copper issues have been attributed to the Shalankayana dynasty (early 4th– mid-5th centuries) and the Vishnukundins (mid-5th–mid-7th centuries). Coins of the Traikutakas (3rd–4th centuries) circulated in the western Deccan and silver issues of the early Kalachuris (6th century) in the Maharashtra area. The numismatic history of the early medieval period is a subject of continuing debate. Historians who describe this period as marked by a feudal order talk of a decline in coinage along with a decline in trade and urban centres, followed by a revival in the 11th century. This hypothesis can be questioned. There was certainly a decline in the aesthetic quality of coins, in the number of coin types, and in their message content. Many are devoid of names or titles, and are therefore difficult to associate with a particular king. However, as demonstrated by John S. Deyell (1990), there does not seem to have been a decline in the volume of coins in circulation.

SILVER G URJARA-P RATIHARA COIN

A number of base metal alloy coin series were issued by dynasties in early medieval times. In the Ganga valley, billon coins circulated in the Gurjara-Pratihara kingdom, while other coin types circulated in Rajputana and Gujarat. Copper coins were minted by the Arab governors of Sindh between the mid-8th to mid-9th centuries. In Kashmir, copper coins were supplemented by bills of exchange (hundikas) denominated in terms of coins or grain, and the use of cowries. During the 6th– 7th centuries, kings of Bengal such as Shashanka issued gold coins. No coin issues of the Pala and Sena dynasties have so far been identified. It has been suggested that the references to currency units in their inscriptions do not represent actual coins but theoretical units of value made up by a fixed number of objects such as cowries. However, a number of silver coins known as Harikela coins were circulating in Bengal between the 7th and 13th centuries and these had corresponding local eastern series, issued in the name of various localities. In the western Deccan, some early medieval coin types have been tentatively identified with the Chalukyas of Badami. Although gold and silver coins found in the Andhra region have been attributed to the early eastern Chalukyas, there seems to be a subsequent gap of about three centuries till the end of the 10th century, when there was a revival of gold and copper coinage under the later kings of this dynasty. The attribution of certain gold and silver coins to the Chalukyas of Kalyana (8th–12th centuries) and to the Kalachuri Rajputs remains uncertain. Coins issued by the Kadambas of Goa (11th–12th centuries) have been identified, and a few gold coins have been attributed to the Shilaharas of the western Deccan (11th century). In the far south, coins with lion and bull motifs, some inscribed with titles, have been associated with the Pallavas. The tiger crest is the emblem on Chola coins. The seals of several Chola copper plate inscriptions show the tiger, fish (the Pandya emblem), and bow (the Chera emblem), indicating that the Cholas had achieved political supremacy over these two dynasties. The appearance of these

three emblems on many gold, silver, and copper coins suggests that these were Chola issues. Gold coins found at Kavilayadavalli in the Nellore district of Andhra Pradesh have the motifs of the tiger, bow, and some indistinct marks. The obverse has the Tamil legend sung which seems to be a short form of sungandavirttarulina (abolisher of tolls), one of the titles of the Chola king Kulottunga I. The legends on the reverse—either Kanchi or Ne (maybe short for Nellur)—may indicate the names of mint towns. The last phase of Chola rule is only represented by copper coins. Coins—mostly copper ones—of the early medieval Pandyas have been found largely in Sri Lanka. A few bear names like Vira Pandya or Sundara Pandya; the problem is one of figuring out which of the several kings of these names they refer to. In many parts of early medieval India, cowries continued to be used as money along with coins. At Sohepur in Orissa, 25,000 cowries were found along with 27 Kalachuri coins. At Bhaundri village in Lucknow, 54 Pratihara coins were found along with 9,834 cowries. Cowries were probably used by people either for small-scale transactions or where coins of small denominational value were in short supply. The market value of cowries fluctuated, depending on demand and supply.

COPPER P ALLAVA COIN

COINS AS A SOURCE OF HISTORY

At first glance, coins may appear to carry little historical information, but they provide clues to several important historical processes. They are linked to monetary history, which includes an analysis of the production and circulation of coinage, the monetary values attached to coins, and the frequency and volume of issues. Monetary history is in turn an important aspect of the history of exchange and trade. At another level, legends on coins give information on the history of languages and scripts.

COWRIE SHELLS

The wide distribution of Kushana coins indicates the flourishing trade of the period. The ship on certain Satavahana coins reflects the importance of maritime trade in the Deccan during this period. Roman coins found in various parts of India provide information on Indo-Roman trade. The few coin series issued by guilds indicate the importance of these institutions. Coins are often taken to indicate levels of economic prosperity (or the lack of it) or the financial condition of ancient states. Historians frequently interpret the debasement of coins as an indication of a financial crisis in the state or more general economic decline, for instance, in the time of the later Guptas. However, in a situation where the supply of precious metals is restricted or reduced, alloying or debasement can be a response to an increase in the demand for coins created by an increase in the volume of economic transactions (Deyell, 1990). As already indicated, the numismatic record of early medieval India is closely tied up with broader debates about the nature of political, social, and economic structures of the time. Dates appear rarely on early Indian coins. Exceptions are western Kshatrapa coins which give dates in the Shaka era and some Gupta silver coins which give the regnal years of kings. Whether dated or undated, coins discovered in archaeological excavations often help date the layers. An example is the site of Sonkh near Mathura, where the excavated levels were divided into eight periods on the basis of coin finds. As important royal message-bearing media, coins form a vital source of political history. The area of circulation of dynastic issues is often used to estimate the extent and frontiers of empires. However, caution has to be exercised, because coins made of precious metals had an intrinsic value and often circulated beyond the borders of the state issuing them. They also sometimes continued to circulate for some time after a dynasty faded from power. Several different currency systems could prevail in an area, and it is necessary to visualize multiple overlapping and intersecting spheres of coin circulation. Numismatic evidence is an especially important source for the political history of India between c. 200 BCE and 300 CE. Most of the Indo-Greek kings are known almost entirely from their coins. Coins also offer information on the Parthians, Shakas, Kshatrapas, Kushanas, and Satavahanas. The coins of over 25 kings with names ending in the suffix ‘mitra’ have been found in the area from east Punjab to the borders of Bihar. Coins found in various parts of north and central India (Vidisha, Eran, Pawaya, Mathura, etc.) mention kings whose names end in the suffix ‘naga’, about whom little is known from other sources. Coins also offer information on ancient political systems. The term

gana on coins of the Yaudheyas and Malavas points to their non-monarchical polity. City coins are suggestive of the importance and possible autonomy of certain city administrations. Sometimes, numismatic evidence offers more than just the names of kings and provides biographical details. For instance, the only specific detail we know about the life of the Gupta king Chandragupta I is that he married a Lichchhavi princess, and this detail comes from coins commemorating the marriage. Coins have helped prove that a Gupta king named Ramagupta ruled between Samudragupta and Chandragupta II. The performance of the ashvamedha sacrifice by Samudragupta and Kumaragupta I is recorded on coins. The archer and battleaxe coin types of Samudragupta predictably advertise his physical prowess, while the lyrist type, which shows him playing the vina, represents a completely different aspect of his personality. PRIMARY SOURCES

Counter-struck coins of the Kshatrapas and Satavahanas

In 1906, a spectacular discovery was made in Jogalthembi, a small village on the outskirts of Nashik in Maharashtra. It was a hoard of 13,250 silver coins of Nahapana, a king belonging to the Kshaharata house of the Kshatrapa rulers, who established his base in the Gujarat area in about the 2nd century CE. As many as 9,270 of these coins had marks of counter-striking by Gautamiputra Satakarni, a king of the Satavahana dynasty, which was a major political force in the Deccan in the early centuries CE. Counter-striking is the phenomenon of coins issued by one authority being re-struck by another authority. Numismatists refer to the original strike of counter-struck coins as the ‘undertype’ and the new one as the ‘overtype’. When properly done, re-striking can completely erase the original under-type. However, in many cases, if the re-striking is not forceful enough, some of the motifs of the undertype can be seen along with the overtype. The authority that originally struck the coin must in all cases have been earlier than or contemporary to the one responsible for the overtype. Shailendra Bhandare describes how counter-striking can provide important historical information about the relative chronology and political history of the Kshatrapa and Satavahana rulers. The design of Nahapana’s silver coins were based on the Indo-Greek silver drachms. The obverse bore his portrait along with a legend in a corrupt form of the Greek script. On the reverse was his dynastic emblem—a thunderbolt and arrow—along with inscriptions in the Brahmi and Kharoshthi scripts. All the coin legends were in the Prakrit language and proclaimed Nahapana as the Kshatrapa of the Kshaharata house. Gautamiputra Satakarni counter-struck Nahapana’s coins with his own symbols. These included an arched hill surrounded by a Prakrit legend giving his name on the obverse. On the reverse was his dynastic emblem—four circles joined by a cross, with a small crescent on top of one of the circles. (Numismatists call this the ‘Ujjain’ symbol.) Another interesting example of counter-striking comes from certain coins issued by Nahapana with counter-strikes by an otherwise unknown Satavahana king named Shiva Satakarni. There are

also coins issued by Shiva Satakarni, counter-struck by Nahapana. The fact that these two rulers were counter-striking each other’s coins indicates that they must have been contemporaries. Counter-striking is generally interpreted as a graphic indication of political rivalry and contest, showing which king had the upper hand over the other at a particular point of time. The rivalry between the Kshaharata and Satavahana rulers is well known from other sources, including inscriptions. However, Bhandare points out that counter-striking was a way of efficiently and swiftly providing an acceptable exchange medium when the political authority in an area had changed, announcing the change to money users. Continuity was an important factor in ensuring that people had faith in the authenticity and value of money, and a sudden change in coin types could create a situation of ‘circulatory shock’— uncertainty and mistrust among coin users. Therefore, when a new political authority took over, it often tried to ensure that its coins did not look too different from those of its predecessors. This is why when Nahapana took over the Nashik area, his coins retained the elephant and a modified form of the tree-in-railing motifs of the earlier Satavahana coins that were in circulation here. Similarly, his Junnar coins retained the lion emblem of earlier coins. At the same time, while trying to maintain continuity, Nahapana made his point on the reverse of the new coins issued from Nashik and Junnar, where the Satavahana ‘Ujjain symbol’ was replaced by his own thunderbolt and arrow emblem. SOURCE Bhandare, 2006

SILVER COIN OF NAHAPANA, RE-STRUCK BYG AUTAMIPUTRA SATAKARNI

The depiction of deities on coins provides information about the personal religious preferences of kings, royal religious policy, and the history of religious cults. For instance, representations of Balarama and Krishna appear on 2nd century BCE coins of the Indo-Greek king Agathocles at AïKhanoum (in Afghanistan), indicating the popularity and importance of the cults of these gods in this region. The depiction of a great variety of figures from Indian, Iranian, and Graeco-Roman religious traditions on the coins of the Kushana kings is generally interpreted as a reflection of their eclectic religious views. But it can equally be read as evidence of the many religious cults prevailing in their

empire and the wide range of religious symbols through which the Kushanas chose to legitimize their political power. CONCLUSIONS

A meticulous and skilful analysis of the sources is the foundation of history. The various literary and archaeological sources for ancient and early medieval India have their own specific potential as well as limitations, which have to be taken into account by the historian. Interpretation is integral to analysing the evidence from ancient texts, archaeological sites, inscriptions, and coins. Wherever several sources are available, their evidence has to be co-related. The co-relation of evidence from texts and archaeology is especially important for a more comprehensive and inclusive history of ancient and early medieval India. However—as will become evident in later chapters—given the inherent differences in the nature of literary and archaeological data, it is not always easy to integrate them into a smooth and seamless narrative.

www.pearsoned.co.in/upindersingh Further resources

Chapter Two Hunter-Gatherers of the Palaeolithic and Mesolithic Ages Chapter outline THE GEOLOGICAL AGES AND HOMINID EVOLUTION HOMINID REMAINS IN THE INDIAN SUBCONTINENT PALAEO-ENVIRONMENTS CLASSIFYING THE INDIAN STONE AGE THE PALAEOLITHIC AGE THE MESOLITHIC AGE CONCLUSIONS

MESOLITHIC PAINTINGS: KATHOTIA, RAMCHAJA, B HIMBETKA (AFTER NEUMAYER, 1983)

In the summer of 1863, Robert Bruce Foote, an officer of the Geological Survey of India, was busy with his routine survey duties at Pallavaram near Madras (modern Chennai). A stone embedded in a

gravel pit caught his eye and he picked it up. It was a seemingly unremarkable piece of brownish quartzite, with one end chipped off, but Foote recognized the unmistakable signs of human workmanship in its form. He had found a handaxe, the first palaeolithic tool discovered in India. Foote went on to find and study many more stone tools and made major contributions to research on Indian prehistory. The Pallavaram handaxe was not the first prehistoric tool discovered in India. In 1856, Le Mesurier, a railway engineer, had found a small chert arrowhead near Nyagurhee village in central India. Prehistoric tools were subsequently reported from many areas including the eastern Vindhyas, the Jabalpur area, Sindh, the Andaman islands, and Bengal. The geologists who played a major role in these discoveries shared their evidence and ideas with European geologists such as Charles Lyell and archaeologists such as J. D. Evans. In 1868, Foote travelled to England to inform the scholarly community about his work, and in 1873, some of the prehistoric tools he had discovered in India were displayed at the International Exhibition held at Vienna. Within two decades, the foundations of Indian prehistory had been laid and had received international recognition.

Since the 19th century, hundreds of prehistoric sites have been identified in the Indian subcontinent and new methodologies and perspectives have enhanced our understanding of the stone age—the longest part of the human past. The sources of information include structural remains, burials, plant remains, bones of humans and animals, and rock art. However, the most prolific and important sources are the tools, mostly of stone, made and used by prehistoric humans. The craft skills represented by these tools must have been developed through experimentation over centuries and carefully transmitted from one generation to the next. It took time, strength, labour, skill, and patience to make stone tools. Some of them are so aesthetically fashioned that they look like works of art. Stone tools are found in different contexts. They may occur on the surface of the ground as surface finds, embedded in river deposits at habitation sites, or at factory sites (places where tools were made). It is important to know whether the artefacts were found in a primary context (in the place where they were made or used), semi-primary (slightly removed from their original place), or secondary context (far removed from their original position). There are various ways in which prehistorians try to ascertain how stone tools may have been made and what they were used for. They can experiment and try to make similar tools, or they can study communities who make and use stone tools today. Another method of understanding the functions of stone tools is microwear analysis. In the course of its frequent use, as a tool comes into repeated, regular contact with certain kinds of materials, its surface and edges develop wear marks and a polish or gloss. Different kinds of activities—cutting plants, chopping meat, cutting hides, etc. —leave different kinds of wear marks and polish. By carefully examining these under a powerful microscope, it is possible to make inferences about what the tool was used for. The question of who made the tools is more difficult to answer. However, considering the active involvement of men and women in subsistence activities, it is very likely that both sexes participated in making stone tools. Stone tools were a very important part of the lives of stone age humans and are therefore an important key to understanding their world. But prehistory is not only about describing and classifying stone tools. It is about using these and other remains to try to discover the life-ways of prehistoric people.

The Geological Ages and Hominid Evolution

Humans like to think that they have always been the centre of the universe, but science has proved that this is not so. This planet and its innumerable species are part of an amazingly long, complex, and continuing drama of evolution, in which human beings made a very late entry, and have so far played a very minor role. The earth is about 4.5 billion years old and humans appeared on it only some 200,000 years ago. The many advances in the physical sciences in the 20th century have greatly amplified our understanding of the earth’s history, while genetic science has unveiled the complex mechanisms that underlay the biological evolution of species. In recent years, advances in DNA analysis have provided important evidence regarding the process of human evolution. The foundations of geological and biological evolutionary theories were laid in the 19th century. Charles Robert Darwin’s path-breaking book, The Origin of Species (1859) explained how new species arose due to adaptation and how the process of natural selection led to the survival of the fittest. Darwin had been deeply influenced by Charles Lyell’s Principles of Geology (1830–33) which explained the past changes in the earth’s surface as results of still-continuing processes such as wind action, erosion, earthquakes, and volcanic eruptions. Thomas Henry Huxley’s Evidence as to Man's Place in Nature (1863) extended Darwin’s idea of evolution to human beings. The authoritative writings of such scholars ultimately revolutionized prevailing ideas about how and when human beings appeared on the earth. Evolutionary theory had enormous and unsettling implications, and it is not surprising that many 19th century Europeans found it difficult to accept. It ran counter to the biblical theory of creation according to which nature and humans were created in all their perfection by a divine agency according to a divine plan. It was not easy to accept the idea that reptiles and insects had appeared on the earth long before human beings, or to recognize certain similarities between humans and chimpanzees, or to think of the world as millions of years old. Just as disconcerting was the fact that evolutionary theory suggested that change in nature was continuing, unpredictable, and unstoppable. The breakthroughs in the natural sciences had an immediate and major impact on prehistoric archaeology. Stone tools had been found and reported in earlier decades, but a theoretical perspective within which such finds could be understood was absent. For instance, in 1836, a French customs officer named Jacques Boucher de Perthes had discovered flint tools in the Somme valley. He had argued that such tools, in some instances found along with bones of extinct animals, were remains of humans who had lived long before the biblical flood. De Perthes’ work was greeted by general scepticism until his finds were authenticated many years later by the geologists Hugh Falconer and Joseph Prestwich, and the archaeologist John Evans. Today, geologists divide the history of the earth into four eras or ages related to the evolution of life forms: Primary (Palaeozoic), Secondary (Mesozoic), Tertiary, and Quaternary. The Tertiary and Quaternary together form the Cenozoic or the age of the mammals, which began about 100 million years ago (mya). The Cenozoic is divided into seven epochs, of which the last two—the Pleistocene and Holocene—are especially important for the story of hominid evolution. The Pleistocene began about 1.6 mya, and the Holocene (or Recent period, in which we live) about 10,000 years ago. In biology, evolution refers to the gradual changes in the heritable features of a species population over successive generations due to changes in gene frequencies and the process of natural selection,

which favours traits that help the species in adapting to the environment. Over time, this process can give rise to a new species. The terms species (or specie) and genus are central to discussions of evolution. A species includes organisms that are similar in physical structure and behaviour and which interbreed with each other, or which could do so if they had access to each other. A genus is an assemblage of related species. Take the following example: Canis familiaris (the domesticated dog), Canis lupus (wolf), and Canis aureus (jackal) all belong to the same genus—Canis—which is mentioned first. The second word is the name of the species they represent. There are many differences in skin colour, facial features, hair colour, body build, height, etc. among modern human beings living in different parts of the world, but we all belong to the same species of anatomically modern humans—Homo sapiens sapiens (the second sapiens refers to our sub-species). Homo sapiens is a Latin term, meaning ‘thinking man’.

TABLE 2.1 GEOLOGICAL AGES AND CORRESPONDING LIFE FORMS

Palaeo-anthropologists have used fossil evidence to piece together the fascinating story of the biological and cultural evolution of early humans. This is not an easy task. It is sometimes difficult to identify a species on the basis of incomplete skeletal material and it is not always clear whether these remains are representative of the entire population of an area. Nevertheless, different stages in the process of human evolution can be identified, as can the implications of crucial biological markers such as increase in cranial capacity (brain size), changes in pelvic structure and the beginnings of bipedalism (walking erect on two legs), and the modification of dental structure due to changing food habits. Some important aspects of the cultural evolution of early humans include the

making of stone tools, the emergence of some kind of social organization, the beginnings of language, and the capacity for symbolic thought. The earliest known hominids (man-like species) were members of the Australopithecus genus, who lived roughly between 4.4 and 1.8 mya, and their remains have so far only been identified in Africa. The earliest of these, Ardipithecus (or Australopithecus ramidus) seems to have evolved from some common ancestor of the hominid and pongid ape lines in sub-Saharan Africa about 4.4 mya. While the Australopithecines may have used naturally available material as tools, there is no conclusive evidence that they were tool makers. Fossil evidence of the earliest representatives of the genus Homo—Homo habilis (hand-using man)—was found at sites such as Koobi Fora in Kenya and the Olduvai gorge in Tanzania, and is dated about 2 mya. The earliest stone tools have been found at Hadar in Ethiopia and have been dated about 2.5 mya.

MAP 2.1 EARLY HOMINID REMAINS

Homo erectus (named for his/her fully erect posture) appeared in East Africa around 1.7 mya. From here, this species seems to have spread to various parts of Africa, Asia, and Europe. The first Homo sapiens appeared a little less than 500,000 years ago. From about 130,000 years ago, there is

evidence of Homo sapiens neanderthalis (Neanderthals) in various parts of western and central Asia and in Europe. Whether the Neanderthals evolved into Homo sapiens or whether they became extinct remains a mystery. Apart from Africa and Europe, hominid remains have also been found in various parts of Asia. Remains of Homo erectus in Java have been dated between 1 to 2 mya and were associated with animal bones of many species but no stone tools. Remains of Homo erectus discovered in the Zhoukoudian caves 50 km south-west of Beijing are dated between 0.58 to 0.25 mya. This site also yielded over 20,000 stone tools and bones of 96 mammalian species. KEY CONCEPTS

What does it mean to be human?

Homo sapiens are one of 180 species of primates (the highest order of mammals). They share some characteristics along with certain other mammals, but they also have their unique features. They are bipedal, that is, they walk upright on two, not four legs. As an adaptation to bipedalism, their legs are longer than their arms, and their back-bone has an S-shape. Their hands are prehensile, i.e., are well suited to grasping. The fingers and large thumb (which can rotate through a 45 degree angle) can be used together to grip a stone tool or a pencil. Compared to other animals, their jaw is small and they do not have protruding canine teeth. Females of most animal species are sexually active only during limited periods known as estrus; such a cycle is absent in human females. Human infants are born with undeveloped brains (only 25 per cent of the full adult size) and remain helpless and dependent on maternal care for a very long time compared to other mammalian species. The story of hominid evolution is, among other things, a story of an increase in brain size, and increased brain size can be connected to greater memory storage, learning abilities, and more complex behaviour. The average brain size of modern humans is large (1450 cc, i.e. cubic centimetres), compared to that of chimpanzees (393.8 cc), Australopithecines (507.9 cc) and Homo erectus (973.7 cc). However, the issue is not just one of absolute brain size or weight, but brain size and weight in proportion to the total body size. The brain of an elephant is more than three times as heavy as that of a human; this doesn’t make the elephant smarter than us. Similarly, the brain size of men tends on average to be larger than that of women. This does not mean that men are necessarily more intelligent than women. Human-ness includes cultural as well as biological characteristics and these have always been interdependent. ‘Modern human behaviour’ includes several traits, not all of which are easy to deduce from archaeological evidence. All animals adapt to and interact with their environment, but human communities have a greater ability to manipulate and transform their environment through the creation of specialized technology. It has been argued on the basis of experiments that chimpanzees and orangutans can make and use simple tools. But humans have a unique ability to make specialized tools, both varied as well as standardized, and travel considerable distances to obtain the desired raw materials.

It is possible that orangutans can learn to use symbols for communication. But there is no doubt that the human thinking capacity is far superior to that of members of the ape family and that human social behaviour and cultural systems are far more diverse and complex than those of the apes. Other traits of human behaviour include the organization and delimitation of living space (camp floors, structures, etc.), symbolic thought and expression reflected in art, ceremonial or ritualistic activity (e.g., burials), and ideas of individual and group identity. Some palaeo-anthropologists argue that while anatomically modern humans appeared on the earth almost 200,000 years ago, fully modern humans—i.e., those whose behaviour can be described as human in the senses mentioned above— appeared only about 50,000 years ago. Others argue that the earliest traces of some of these ‘human’ traits can in fact be found in species other than Homo sapiens sapiens, for instance, among the Neanderthals as well as among some of the archaic hominids.

F IGURE 2.1 SKULL STRUCTURE OF GORILLA, HOMO ERECTUS, HOMO SAPIENS SAPIENS

Anatomically modern humans, known as Homo sapiens, seem to have appeared in Africa between 195,000 and 150,000 years ago, and eventually replaced all other Homo species. Important fossil remains have come from the site of Herto in Ethiopia, where hominid remains were found along with stone tools and animal bones in levels dated between 160,000 to 154,000 years ago. There are many questions to which there are yet no definite answers and which remain matters of debate. It is possible that Homo sapiens evolved in Africa and then migrated to various parts of Asia and Europe. Or, the migration out of Africa could have happened at an earlier stage, and modern Homo sapiens may have evolved from Homo erectus and archaic Homo sapiens more or less simultaneously on different continents. Evolution was not a neat unilinear process, one species making way for another. There is evidence from various parts of the world of the overlap and co-existence of species. For instance, the remains in Olduvai gorge in east Africa show the co-existence of Homo habilis and Australopithecus, and there is similar evidence of the co-existence of the Neanderthals and anatomically modern humans in the eastern Mediterranean. Hominid Remains in the Indian Subcontinent

In sharp contrast to the widespread occurrence of animal fossils and stone tools all over the subcontinent, the evidence of hominid fossils is at present very meagre (Kennedy, 2000; Chakrabarti,

2006: 10–16). This is no doubt due to inadequate investigations. From the 19th century onwards, several remains of fossil apes were discovered in the Siwalik hills, the outermost range of the Himalayas. Given rather dramatic names such as Ramapithecus, Sivapithecus, and Brahmapithecus, they came to be collectively known as the ‘God-Apes of the Siwaliks’. Remains of Ramapithecus were subsequently found in other parts of Asia, Africa, and Europe as well, and were dated between 10–14 mya. Ramapithecus, who lived in the Miocene– Pliocene transition, was once thought to represent the oldest direct ancestor of modern humans. However, this has been questioned on the basis of new dating methods and a reassessment of the fossil evidence. Authenticated early human remains in South Asia are relatively recent. In 1966, Louis Dupree discovered a fragment of a right temporal bone at the cave site of Darra-i-Kur in north-eastern Afghanistan. The deposit in which it was found gave a radiocarbon date of 30,000 ± 1900–1200 BP i.e., 28,950 ± 1960–1235 BCE. The fragment was considered consistent with Neanderthals as well as anatomically modern humans. The associated stone tools seem to belong to a middle palaeolithic context. Several cave sites in Sri Lanka—Fa Hien Lena, Batadomba Lena, Beli Lena, and Alu Lena —also yielded remains of anatomically modern humans in contexts ranging between 37,000–10,500 BP. More recently, hominid fossils have been found in central India. In 1982, Arun Sonakia of the Geological Survey of India made an important discovery near Hathnora village on the northern bank of the Narmada, about 40 km north-east of Hoshangabad. Here, embedded in thick, closely packed sandy, pebbly gravel he found a fossilized fragment of a cranium (skull cap) along with some fossils of vertebrates (proboscideans and bovids) and a few late Acheulian tools. The skull fragment seems to have belonged to a woman about 30 years old. Sonakia suggested that she represented an advanced variety of Homo erectus —‘advanced’ because of her larger cranial capacity range of 1155 to 1421 cc— and named her Homo erectus narmadensis. However, according to other scholars, the cranium belongs to an early (archaic) variety of Homo sapiens. Its date too is uncertain. One view is that it belongs to the early part of the middle Pleistocene, beginning about 500,000 BP. Between 1983 and 1992, the Anthropological Survey of India launched an intensive search for human fossils and tools in the central Narmada valley. This led to the discovery of hundreds of palaeolithic tools and some animal fossils. In 1997, A. R. Sankhyan announced important discoveries in the same boulder conglomerate deposit at Hathnora where the cranial fragment had been found some years earlier. These included a hominid clavicle (collar bone) along with animal fossils and several late or middle palaeolithic tools. Estimated dates of these finds range between 0.5 to 0.2 mya. Sankhyan suggested that the two sets of human fossils found at Hathnora may well belong to the same woman. In 2001, P. Rajendran, a teacher in the Department of History of Kerala University, found a complete fossilized human baby skull in Odai in the Villupuram district of Tamil Nadu. Rajendran was excavating a trench which had microliths in the upper levels and upper palaeolithic tools at the lower ones. At a depth of 6 m, just under the upper palaeolithic deposit, there was a ferricrete deposit (a mineral conglomerate consisting of sand and gravel, cemented into a hard mass by iron oxide). The skull was found close to this trench, embedded in a similar ferricrete deposit which was later dated 166,000 BP, placing it in the middle or upper Pleistocene.

The antiquity of certain other reported hominid finds is uncertain. This is the case with the two human mandibles of an adult male and female Homo sapiens found by H. D. Sankalia and S. N. Rajaguru on the bank of the Mula-Mutha river in Pune district, Maharashtra. The age of the mandible of an adult male found by V. S. Wakankar in a cave at Bhimbetka in Madhya Pradesh is similarly uncertain.

SEE P . 69 FOR AN EXPLANATION OF ACHEULIAN TOOLS

ARUN SONAKIA HOLDING THE HATHNORA SKULL CAP

MAP 2.2 HOMINID DISCOVERIES IN THE SUBCONTINENT

Only a very miniscule proportion of the hominid record of the Indian subcontinent has so far been discovered. More concerted efforts are likely to add to the data and may transform the larger story of human evolution, which has so far concentrated more on Africa and Europe than on South Asia. Palaeo-environments

The environments in which prehistoric people lived were very different from ours. Some of the major changes that gave the subcontinent its present form took place millions of years ago, in some instances long before hominids appeared on the planet. Hundreds of millions of years ago, the peninsula was part of a huge land mass that geologists call Gondwanaland, which included Australia, Africa, South America, and Antarctica. At some point of time, this land mass broke up and the Indian landmass started drifting northwards at the rate of 20 cm a year, eventually joining up with the Asian landmass, between 50 and 35 mya. All this was the result of the movement of massive

tectonic plates embedded within the earth. The collision and intermittent pressure of the Indian and Asian plates led to uplifts that resulted in the creation of the Tibetan plateau and the Himalayas. Rivers brought down immense volumes of eroded sediments from the mountains, and this resulted in the creation of the fertile northern alluvial plain. The process of plate tectonics (the word ‘tectonic’ means movements in the earth’s crust) is not over. The Indian plate continues to press into Asia at the rate of 5 cm a year. The Himalayas and the Tibetan plateau are still rising at an average of 5–10 mm per year. Occasional movements in the tectonic plates lead to intermittent earthquakes and changes in the course of rivers in the northern parts of the subcontinent. All over the world, the Pleistocene era, which began about 1.6 mya, was marked by dramatic climatic changes. The earlier idea of a sequence of four ice ages and four interglacial periods for the higher latitudes has been questioned. There seem to have been more than four ice ages and interglacials, corresponding to alternating periods of cold and warmer climate. During the cold phases, when ice sheets covered one-third of the earth’s landmass, sea levels fell dramatically. When the climate became warmer, the ice melted and sea levels rose. It is believed that the tropical and semi-tropical regions went through alternating dry and wet phases (interpluvial and pluvial phases), but the rhythm of Pleistocene climatic changes in these parts of the world is not fully understood. The Pleistocene environments of the subcontinent were influenced by larger global patterns of climate, but sometimes also by distant seismic events. For instance, about 75,000 years ago, a gigantic volcanic super-eruption occurred in Sumatra at a place today represented by lake Toba. This seems to have led to a complex series of palaeo-environmental changes in late Pleistocene times, which had a significant impact on hominid populations in the region. Tephra ash deposits arising from this eruption have been found embedded in river valleys in peninsular India, and the impact of the Toba eruption on hominid populations in this area is being studied. About 10,000 years ago, the Pleistocene era made way for the Holocene era (which continues into our own time) and the basic climatic patterns that prevail in the world today were established. This does not mean that there have been no significant climatic changes in the last 10,000 years. It is just that these changes have not been as enormous as those that occurred within the Pleistocene. The beginning of the Holocene was marked by wetter climatic conditions than those of the late Pleistocene. The study of the specific features of palaeo-environments is a very important part of prehistory. Detailed palaeo-environmental studies are so far available for very few parts of the subcontinent. One of the earliest such studies was conducted in 1935 by H. de Terra and T. T. Paterson on the Soan (Sohan) river in the Potwar plateau, between the Pir Panjal and Salt ranges in Pakistan. Their team found a large number of tools, mostly of the middle and upper palaeolithic, some of the lower palaeolithic as well. De Terra and Paterson identified five tool-bearing terraces (a terrace is an old bed of a river) of the Soan and tried to correlate these terraces with the theory of a four-fold glacial cycle in Kashmir, and further, with a four-fold European glacial cycle. This framework was extended, through comparisons, to the Narmada and the area around Chennai. Although most of the correlations, sequences, and conclusions of the de Terra–Paterson study are no longer accepted, it marked an important stage in the history of prehistoric research in India. In 1930, L. A. Cammiade and M. C. Burkitt carried out a similar study, correlating the stratigraphy of prehistoric stone tools and their environment in the Eastern Ghats of Andhra Pradesh.

Studies of the Son valley (in northern MP) and Belan valley (in southern UP) have thrown light on the connections between the changes in river systems, climate, and stone age sites in the valleys of these southern tributaries of the Ganga (Clark and Williams, 1986). During the late Pleistocene, the climate in this area was much cooler and drier than it is today. At the same time, hippopotamus and crocodile bones show that some permanent water was available in rivers and streams. In the early Holocene, the climate seems to have become warmer and wetter, probably leading to an expansion of forests and shrinking of grasslands. The Thar desert today has very little naturally occurring surface water, except for short periods in the rainy season, and people have to rely on rain water stored in tanks, wells, tube wells, and canals. A study of the western Rajasthan section of the Thar desert (Misra and Rajguru, 1985), especially around Didwana in Nagaur district, indicates that the present environment of the Thar is very different from what it was like in the Pleistocene era. Except for a phase in the upper Pleistocene (25,000–13,000 BP), during most of that era, surface water in some quantity was always available; the flora and fauna was as a result much more abundant than it is today. The sediments of the salt lakes indicate a significant increase in rainfall in the mid-Holocene (6,000–4,000 BP). It is not a coincidence that the most widespread prehistoric occupation in this area belongs to that period. Classifying the Indian Stone Age

The three-age system—the idea that there was an age of stone tools, followed by one dominated by those of bronze and then of iron—was first put forward in the late 18th and early 19th centuries by the Danish scholars P. F. Suhm and Christian Thomsen. The accuracy of this theory was proved by excavations by another Danish scholar, Jacob Worsaae. The next important step was to identify changes within the stone age. In 1863, John Lubbock divided the stone age into two parts, the palaeolithic and neolithic. A few years later, Edouard Lartet suggested the division of the palaeolithic into the lower, middle, and upper palaeolithic, largely on the basis of changes in fauna associated with the different tool types. Archaeologists gradually identified distinct tool-making traditions within the palaeolithic and also recognized the significance of changes in subsistence patterns within the stone age. The use of the term mesolithic is relatively recent. The Indian stone age is divided into the palaeolithic, mesolithic, and neolithic on the basis of geological age, the type and technology of stone tools, and subsistence base. The palaeolithic is further divided into the lower, middle, and upper palaeolithic. A general time range for the lower palaeolithic is from about 2 mya to 100,000 years ago, the middle palaeolithic from about 100,000 to 40,000 years ago, and the upper palaeolithic from about 40,000 to 10,000 years ago. However, there is a great deal of variation in the dates for different sites. The palaeolithic cultures belong to the Pleistocene geological era, while the mesolithic and neolithic cultures belong to the Holocene era. While Table 2.2 explains the basic features of the different phases of the stone age, it also tends to over-simplify matters. It must be remembered that this classification is an analytical tool used by scholars to identify patterns across a very long and complex span of the human past. Except for the dividing line of the Holocene, stone age cultures did not evolve uniformly in a neat unilinear fashion all over the subcontinent. There are regional variations in some of their features and their dates also vary considerably. The ‘typical Indian tool types’ column in the table indicates

the tools that are considered characteristic of that particular phase. However, it does not mean that there is complete uniformity in tools found at different sites, or that tools typical of one phase were absent in another. For example, celts are associated with the neolithic, but are known to occur as late as the historical period in certain parts of eastern India. Similarly, with regard to the subsistence base, it should be noted that hunting and gathering did not come to an end with the beginnings of animal and plant domestication. Many agricultural communities continued to hunt and forage for food. In fact, these subsistence activities continue to be prevalent in certain niches of the subcontinent even today.

TABLE 2.2 IMPORTANT FEATURES OF THE STONE AGE

It is easier to identify and describe stone tools than to know whether, or to what extent, a community was producing its food through plant or animal domestication. Sometimes, there is insufficient data to reach a conclusion. Finally, there is the issue of overlap. Although there are some ‘pure’ neolithic sites in India, early agricultural sites frequently show an intermixture of neoliths with copper and copper-alloyed objects.

SEE CHAPTER 3, PP . 95–101 FOR A DISCUSSION OF PLANT AND ANIMAL DOMESTICATION AND FOODPRODUCTION

The Palaeolithic Age LOWER PALAEOLITHIC SITES

Palaeolithic tools have been found in almost all parts of the subcontinent (Chakrabarti, 1999: 54–75; Allchin and Allchin, 1997: 47–85). Although hardly any sites have so far been discovered in the alluvial stretches of the Indus or Ganga valleys (Kalpi in UP is an exception), they have been identified on rocky areas within or on the margins of these valleys, e.g., in the Rohri hills in Sindh and the northern fringes of the Vindhyas. Sites are prolific in other parts of the subcontinent,

especially in peninsular India, leaving aside the coastal plains. Comparatively few palaeolithic habitation sites have been identified, but it can be assumed that people lived close to sources of food, water, and stone in different kinds of habitats—for instance, along the banks of rivers or streams and in caves and rock shelters. Excavated sites are comparatively few and most of the evidence comes from surface finds of stone tools. Because of insufficient data from most sites, it is necessary to focus on the published results of stone tools found in clearly defined stratigraphic contexts. Some sites were inhabited over many stages of the stone age. Even in the absence of detailed studies, some broad inferences about Pleistocene climate can be made on the basis of the deposits in which palaeolithic tools are found. For instance, tools often get embedded in river terraces. Although a number of other factors are also involved, the erosion and deposition activity of rivers can be related to rainfall. Cemented gravel (a deposit in which small pebbles are packed tightly together in soil) is generally taken to represent a wet climatic phase. A boulder conglomerate (a deposit where larger boulders are packed together) is interpreted as representing a drier phase, while clay or silt deposits represent still drier conditions. Early palaeolithic tools were fairly large core tools made of quartzite or other hard rocks. They include chopping tools, handaxes, and cleavers. Apart from directly breaking off pieces of stone from large boulders, which would have required considerable strength, it is possible that people lit fires against rocks and threw water over them so that large fragments broke off more easily. Within the palaeolithic, there is a gradual increase in the range and variety of stone tools and a shift in preference from coarse-grained to fine-grained stone. In recent years, important evidence of dates for lower palaeolithic contexts has come from the Potwar plateau and the Siwaliks. At Dina and Jalalpur in the Jhelum basin, members of a British archaeological team discovered 15 artefacts including three handaxes in a boulder conglomerate deposit dated c. 700,000–500,000 years ago by the palaeo-magnetic method. There are much earlier dates from Riwat near Rawalpindi in Punjab province of Pakistan. Here, in 1983, members of the British Archaeological Mission to Pakistan’s Potwar Project, working with the Department of Archaeology and the Geological Survey of Pakistan, discovered stone artefacts embedded in a stone conglomerate deposit dated 2.01 mya by the palaeo-magnetic method. At the sites of Gurha Sahan and PS-57, stone tools were found embedded in the Pinjor bed of the Siwaliks, dated between 2.4 and 2 mya. Stone tools reported in the Jammu and Himachal sections of the Siwalik hills seem to belong to about the same age. For instance, at Uttarbaini in the Jammu area, early palaeolithic tools were found in a deposit dated 2.8 ± 0.5 mya.

F IGURE 2.2 THE PERCUSSION TECHNIQUE OF MAKING FLAKES

PRIMARY SOURCES

Typical lower palaeolithic tools

Stone tools are an important key to understanding the lives of prehistoric humans. It is therefore very important to understand the meaning of terms used by prehistorians for different stone tool types, especially since some of them can be rather misleading. If you take a piece of stone and break it into two or more pieces, the largest piece is called a core and the smaller piece or pieces are called flakes. A stone tool made out of the largest piece (core) is called a core tool, while tools made out of the smaller pieces (flakes) are called flake tools. Removing slivers or pieces from a rock is called flaking. The depressions or marks formed on the surface of a stone when flakes are removed are known as flake scars. A handaxe is generally a core tool. It is also known as a biface, because it is usually worked on both sides. Generally made on a core, it is roughly triangular in shape, broad at one end and

pointed at the other. Not all handaxes were handheld tools; some of them could have been hafted onto handles. Pebble tools are tools of different types made on pebbles, in which only the working edge is flaked, the rest of the tool remaining untouched. A chopping tool is a tool made on a core or a pebble and is flaked alternately on both sides to produce a wavy cutting edge. A chopper is a large, unifacial tool, i.e. worked on one side only. A cleaver is a flattish tool made on a broad rectangular or triangular flake, on one end of which is a broad and straight cutting edge. The term Acheulian is often used to refer to an assemblage of stone tools marked by advanced and increasingly symmetrical handaxes and cleavers. These are associated with the lower palaeolithic, but continue well afterwards as well. SOURCE Sankalia [1964], 1982: 45–58

F IGURE 2.3 LOWER PALAEOLITHIC TOOLS

MAP 2.3 MAJOR PALAEOLITHIC SITES

Some absolute dates are now available for lower palaeolithic contexts in other areas as well. Didwana in Rajasthan has been dated 390,000 BP (by the uranium/thorium series dating method). In the Hiran valley in Gujarat, the lower palaeolithic context is dated 190,000–69,000 BP (via the uranium/thorium series dating method). For the Son valley (MP), there is a thermoluminescence date of 103,800 ± 19,800 BP. Nevasa (in Maharashtra) has given a date of 350,000 BP (via uranium/thorium series dating). In Karnataka, the site of Yedurwadi has been dated 350,000 BP. Factory sites are generally located close to the sources of raw materials and are marked by a profusion of stone tools in various stages of preparation. In many instances, they were visited and used during several phases of the stone age, sometimes even later. In Sindh, there are a number of such sites in the limestone hills capped by flint nodules. In lower Sindh, stone tools belonging to the lower, middle, and upper palaeolithic were found at sites such as Jerruk and Milestone 101. In upper Sindh, there are factory sites in the Sukkur and Rohri hills.

Many people tend to think of stone age sites as distant, isolated places. As a matter of fact, stone age tools are often found in places that are today bustling with activity. A good example are the many sites found in and around the modern city of Delhi. Four lower palaeolithic stone tools were found in 1956 on the Delhi Ridge, near the main gate of the University of Delhi, and more were subsequently discovered on the northern Ridge. In 1983, a late Acheulian handaxe was found on the campus of Jawaharlal Nehru University. A systematic study of stone age sites in south Delhi and adjoining areas (Chakrabarti and Lahiri, 1986) identified 43 sites ranging from the lower palaeolithic to the microlithic. Excavations at Anangpur in the Badarpur hills to the south of the city revealed thousands of early and late Acheulian tools along with traces of several palaeo-channels of the Yamuna river. The evidence indicates that this was a large lower palaeolithic habitation and factory site. In Rajasthan, lower, middle, and upper palaeolithic tools have been found around Ajmer and stray finds of lower palaeolithic tools occur in the Luni valley. There is a detailed profile of the Didwana area of the Nagaur district in western Rajasthan, with a sequence extending from the early to the middle palaeolithic. The Mogara hill near Jodhpur seems to have been a factory site where lower, middle, and upper palaeolithic as well as mesolithic tools were made.

THE B HIMBETKA ROCK SHELTERS

In Gujarat, lower palaeolithic tools have been found in the valleys of the Sabarmati, its Orsang and Karjan tributaries, and in the Bhadar valley in Saurashtra. Lower palaeolithic and later artefacts have been found all along the Konkan coast up to Goa. In Maharashtra, palaeolithic tools have been found in many places along the coast and in the Wardha–Wainganga valleys. Stratigraphic profiles of sections of the Mula-Mutha, Godavari, Pravara, and Tapi rivers are available. Lower and middle palaeolithic tools have been found in stratigraphic contexts in the Dattawadi area of the Mutha river in Pune. Lower palaeolithic tools have been found in a stratigraphic context in the Gangawadi area on the Godavari at Nasik. Prehistoric remains occur in various parts of central India in Damoh, Raisen, and the Narmada, upper Son, and Mahanadi valleys. The Narmada valley is an especially rich and well-researched area. Excavations at Adamgarh hill, not far from Hoshangabad, revealed a sequence of lower and middle palaeolithic tools. However, the most spectacular finds come from hundreds of rock shelters

at Bhimbetka (in Raisen district, MP), 30 km north of Hoshangabad, which have given evidence of an enormously long sequence of occupation stretching from the lower palaeolithic to the historic period. The Bhimbetka hillside is composed of sandstone and quartzite. There are three perennial freshwater springs in the area, and several creeks filled with water. A study of the present-day flora and fauna indicates the presence of at least 30 plant types which yield edible fruits, tubers, and roots. There are fish in the streams, and the hillside is home to many animals such as the deer, boar, nilgai, leopard, wolf, hare, and fox. Of course, in prehistoric times, conditions wouldn’t have been exactly like this. Nevertheless, it is clear that this site must have been attractive for stone age people from the points of view of shelter, food, and raw material for tools. Most of the stone tools at Bhimbetka were made of a yellowish quartzite available in plenty in the area, but a grey quartzite was also obtained from further away. Five floors paved with flat stone slabs belonging to the lower palaeolithic were identified. No bones have been found so far, perhaps because of the acidic soil. In the Belan valley in Uttar Pradesh, detailed studies have revealed a sequence of stone age industries from the lower palaeolithic to neolithic to protohistoric. In Bihar in eastern India, a lower palaeolithic living and working floor was excavated at Paisra in the Kharagpur forests near Munger (Pant and Jayaswal, 1991). The whole area was rich in finished and unfinished artefacts, broken pieces of stone, and anvils. Eight post-holes were found, marking places where wooden posts had been dug into the ground to support thatched huts. The river valleys and foothills of the Chhotanagpur plateau in Jharkhand and the adjoining areas of West Bengal have yielded lower palaeolithic tools. In Orissa, tools of all three phases of the palaeolithic have been found in many places. A large number of lower and middle palaeolithic tools were found in explorations at Dari-dungri in Sambalpur district, and lower palaeolithic tools have also been found along the valleys of the Budhabalan and Brahmani rivers.

QUARTZITE HANDAXE FROM THE NARMADA VALLEY

At one time, it was believed that the lower palaeolithic industry of the south (which was given the name ‘Madrasian’) was different from that of other parts of the country because of a supposed absence of pebble tools. The research of the past few decades has proved that this is incorrect, and that pebble tools such as choppers and chopping tools are found along with handaxes at several sites. A stratigraphic sequence of lower and upper palaeolithic tools was identified in the Malaprabha– Ghataprabha valleys in Karnataka. Lower palaeolithic tools have also been found in the Hunsgi– Baichbal and Krishna valleys. Lower palaeolithic tools occur at many places at Hunsgi (in the Gulbarga district of Karnataka), on the banks of the Hunsgi, a tributary of the Krishna river (Paddayya, 1982). Here, sites with very few types of artefacts may represent places where certain specific activities such as making tools or killing game were carried out. Sites where tools occur in larger number and variety, may have been temporary camp sites. Still larger sites, where stone tools have been found in great profusion and variety, may have been places where groups of people lived for longer periods of time. The Hunsgi tools were mostly made of various kinds of stone including limestone, sandstone, quartzite, dolerite, and chert, some of which were not locally available. In one of the excavated areas, huge granite blocks were arranged around a 63 sq m area, perhaps used as a support for temporary shelters made of branches, grass, and leaves. Today, the area around Hunsgi supports about 40 types of wild edible plants as well as plenty of small game. RECENT DISCOVERIES

Isampur: a centre of stone tool manufacture

Isampur (Gulbarga district, Karnataka) is a village located in the north-western part of the Hunsgi valley, drained by a small seasonal stream known as the Kamta Halla. The palaeolithic site lies about 2 km north-west of the village, close to the bank of the stream, covering an area of about 7,200 sq m. It was discovered in 1983, when the silt deposits overlying the limestone floor of the valley were exposed due to quarrying activity carried out as part of a major irrigation project. This site offered some obvious advantages to prehistoric humans. Water and a variety of wild animal and plant food were available. Another advantage was that siliceous limestone blocks and slabs occur plentifully in the area at the intersection of flat and steep surfaces. There is evidence of Acheulian as well as middle palaeolithic occupation at the site. The Acheulian material mostly consisted of cores in different shapes, large flakes, and debitage (waste material). The main tool types were chopping tools, knives, handaxes, cleavers, and scrapers. While unfinished tools occurred in large numbers, there were relatively few finished ones. Hammer stones of different sizes, made of hard rocks such as quartzite, basalt, and chert were found in very large numbers on the surface and in the excavated levels. There is evidence of quarrying and of different stages in tool manufacture. The middle palaeolithic assemblage consisted of flake tools, mostly made out of locally available chert nodules. These included finished tools, cores, hammer stones, flakes, and debitage. There were also tools made of quartzite and limestone. Scrapers of various types were the most numerous. Tools were made both by simple flaking and through the use of a prepared core technique. The site consisted of four sub-localities, each measuring 300–400 sq m, within which there were many limestone slabs and blocks suitable for making tools. These rocky patches must have been centres of tool-making activity. Given the large extent of the site and the huge number of tools found here, it seems that Isampur was one of several hubs of stone tool manufacture in the Hunsgi–Baichal valleys, from where hominids must have ranged out to the valley floor and the uplands for foraging. Some of the tools found here are weathered and have use-marks, showing that the site was also a habitation site where people lived and carried out subsistence activities such as food processing. The prehistoric occupation at Isampur seems to go back to between 500,000 and 600,000 years ago. The continuing investigations at this site are likely to provide further valuable data about the lower palaeolithic. SOURCE Paddayya et al., 1999–2000 F IGURE 2.4 ISAMPUR TOOLS

In Andhra Pradesh, lower palaeolithic tools have been found in inland areas as well as the coastal Visakhapatnam area, where they have been connected to a sea level over 7 m above the present one. Nagarjunakonda, one of the sites that have been studied extensively, has given palaeo-climatic

evidence of three alternating wet and dry cycles. Choppers and scrapers made of quartz have been found in the Palghat district of Kerala. In Tamil Nadu, there is a stratigraphic sequence from the early palaeolithic to the mesolithic from near Chennai. Gudiyam cave, not far from Chennai, has yielded a sequence of lower, middle, and upper palaeolithic tools. The fewness of the tools and the absence of other remains suggest that the site was occupied for short periods of time.

H. D. SANKALIA (1908–89), A PIONEER OF INDIAN ARCHAEOLOGY

Attirampakkam, in the Kortallayar river basin, is one of the richest palaeolithic sites in Tamil Nadu (Pappu et al., 2003). The site was discovered in 1863, and has been excavated, off and on, since then. The most recent excavations revealed a sequence of lower, middle, and upper palaeolithic cultures, with a break in occupation after the middle palaeolithic. Acheulian tools were found in a 4 m thick deposit of clay. The artefacts, mostly handaxes, were made of quartzite stones that were not available locally. Very little debitage was discovered at the site, suggesting that the tools were made somewhere else and then brought here. One of the most interesting discoveries was a set of animal foot-prints found along with Acheulian tools. The 17 round impressions (15–20 cm) of animal feet and a set of hoofprints are still being studied by experts. This is the first discovery of its kind in South Asia. Another interesting discovery was of three animal fossil teeth, possibly those of some kind of horse, water buffalo, and nilgai, suggesting an open and wet landscape in early palaeolithic times.

LOWER PALAEOLITHIC TOOLS FROM ATTIRAMPAKKAM

MIDDLE PALAEOLITHIC SITES

Within the palaeolithic, there were gradual changes in stone tools. Handaxes, chopping tools, and cleavers did not altogether disappear, but the balance shifted towards smaller, lighter flake tools, some of them made by prepared core techniques, including the Levallois technique. Middle palaeolithic tools have been found in many parts of the subcontinent, often in river gravels and deposits, which give clues about prevailing climatic conditions. There are some dates for middle palaeolithic contexts. Didwana (Rajasthan) has given two thermoluminescence dates of 150,000 BP and 144,000 BP. The Hiran valley (Gujarat) has yielded a uranium–thorium series date of 56,800 BP. In the north-west, lots of stone tools, mostly of the middle palaeolithic, have been found in the Potwar plateau between the Indus and Jhelum rivers. The over 3 m thick deposit in the Sanghao cave in the North-West Frontier Province of Pakistan revealed a sequence of middle and palaeolithic occupation. Thousands of stone tools were found, along with bones (of animals, some perhaps of humans) and hearths. All the tools are made of quartz, which is easily available around the site.

Many of the tools of Period I were made from flakes stuck from prepared cores, and there were lots of burins. In the Thar region, middle palaeolithic artefacts occur in reddish brown soil, which indicates more abundant vegetation, more surface water, and a cooler, wetter, and more humid climate compared to lower palaeolithic contexts. Small factory sites and camp sites have been found in various parts of the Thar, especially near rivers and lakes. A large number of stone age sites belonging to the middle palaeolithic phase onwards are located around Budha Pushkar lake, an area which offers advantages of the easy availability of water and stone. Middle and upper palaeolithic tools are also found around Ajmer. There is evidence of middle palaeolithic working floors at Hokra and Baridhani, close to the now dried-up lakes. In the Jaisalmer area, upper palaeolithic material is not as abundant as are artefacts of the middle palaeolithic. Middle palaeolithic sites have also been located along the now virtually extinct Luni river system. The term Luni industry is used for middle palaeolithic assemblages west of the Aravallis, and can be contrasted with the industry of the regions lying east of the Aravallis. Although certain forms are common to both areas, sites to the west of the Aravallis display more variety in stone tool types and larger numbers of reworked flakes. Middle and upper palaeolithic tools have also been found along the eastern margin of the Gujarat plain.

B ORER FROM NELLOR DISTRICT (AP)

PRIMARY SOURCES

The Levallois technique

The Levallois technique is an advanced way of making flake tools. It is named after a place called Levallois Perret near Paris, where this technique was first noticed on prehistoric stone tools. Instead of breaking off a flake and working on it to produce the desired shape, the core was carefully prepared. Its sides were trimmed, and flakes were then systematically removed from its surface, from the centre outwards in all directions. Then, a striking platform was created

by flattening the top of the prepared core, and perpendicular blows were struck at that point, either directly or through an intermediary tool. The flake detached in this way was thin, roughly triangular or oval in shape, with a clean undersurface, and shallow, centrally directed flake scars on the upper side. It would need very little further working, because its edges were already sharp. Because the core of a Levallois flake looks like the shell of a tortoise, it is sometimes referred to as a tortoise core. There are other prepared core techniques as well. For instance, in the discoid core technique, flakes are scalloped from the circumference of a large core or flake with at least one flat side. The remaining core has a bevelled rim and is flat in the centre. The Levallois technique can be used to produce only one flake at a time, while the discoid core technique can produce several flakes. Flakes produced by the latter method tend to be small. SOURCE Sankalia [1964], 1982: 29–30 F IGURE 2.5 P REPARATION OF A LEVALLOIS FLAKE

F IGURE 2.6 MIDDLE PALAEOLITHIC TOOLS

The middle palaeolithic industry of central and peninsular India is sometimes referred to as the Nevasan industry after the site of Nevasa, where the pioneering archaeologist H. D. Sankalia first discovered middle palaeolithic artefacts in a stratified context. The tools, which include a wide variety of scrapers, are made of smooth, fine-grained stone such as agate, jasper, and chalcedony. Patne in the Tapi valley revealed a stratigraphic sequence of middle and upper palaeolithic and mesolithic tools. There is evidence of a middle palaeolithic living and factory site at Chirki near Nevasa. The earliest trace of human occupation in the Ganga plain is found embedded in a 20 m thick cliff section at Kalpi (in Jalaun district, UP), on the southern bank of the Yamuna. A number of vertebrate fossils—elephant tusk, shoulder blade of elephant, molars of Equus and bovids—were found here. Middle palaeolithic stone tools (including pebble tools, points, and side scrapers) and bone tools (such as end scrapers, points, and burins) were found along with them. The tool-bearing level at Kalpi has been dated about 45,000 years ago. There are several middle and upper palaeolithic sites further east, especially in the western part of West Bengal.

In South India, the middle palaeolithic culture is marked by a flake tool industry. On the Visakhapatnam coast, quartzite, chert, and quartz were frequently used to make stone tools. There is evidence of tools made by the Levallois technique at many places. In addition to smaller handaxes, cleavers, and choppers, the middle palaeolithic tool kit included new tool types such as scrapers of different shapes. A C-14 date for the middle palaeolithic context at the coastal site of Nandipalli in Cuddapah district indicates that it is older than 23,000 years ago.

MIDDLE PALAEOLITHIC SCRAPER FROM ATTIRAMPAKKAM

UPPER PALAEOLITHIC SITES

The important technical advance of the upper palaeolithic was the making of parallel-sided blades. There was also an increase in the number of burins. The trend was towards smaller tools, and this must have been due to adaptations to environmental changes. It is known, for instance, that the climate of northern and western India seems to have become increasingly arid during the upper palaeolithic. Older tool types continued to be made for activities that required heavier tools. There are some dates for upper palaeolithic contexts. Site 55 at Riwat gives the earliest date for the upper palaeolithic—c. 45,000 years ago. C-14 dates from the Sanghao cave range from 41,825 ± 4,120 BCE to 20,660 ± 360 BCE. In central India, the Son valley has given radiocarbon dates within the range of 12,000–10,000 BP, and a piece of ostrich eggshell at Mehtakheri has been dated to over 41,900 BP. Two dates from the Kurnool caves (in Andhra Pradesh) are 19,224 BP and 16,686 BP (based on the electron spin resonance method). In the north-west, the Sanghao cave has given evidence of middle and upper palaeolithic tools, hearths, animal bones, and what appear to be burials. Upper palaeolithic tools have also been found in the Rohri hills in upper Sindh and Milestone 101 in lower Sindh. In north India, the Kashmir upper palaeolithic has been dated to about 18,000 BP and coincides with the onset of a milder climate.

In the Thar, the number of upper palaeolithic sites is fewer than those of the preceding phase, due to increasing aridity. However, there was continuing human occupation around the Budha Pushkar lake. In central India, upper palaeolithic habitation sites have been found in caves and rock shelters of the Vindhyas. The upper palaeolithic context in the Belan valley has been dated between 25,000 and 19,000 years ago, and that of the Son valley about 10,000 years ago. Chopani Mando in the Belan valley seems to be a habitation site with a cultural sequence from the upper palaeolithic to neolithic. The upper palaeolithic assemblage consisted of tools made from chert, a stone available in the nearby Vindhyas. The animal bones discovered in the Belan valley included those of wild cattle, sheep, and goats. Since sheep and goats do not seem to be indigenous to this area, they may have been brought here from the north-west. If this was indeed the case, it could represent an early stage of animal domestication. In Siddhi district of Madhya Pradesh, in the valley of the Son river, an archaeological team led by G. R. Sharma and J. D. Clark excavated the upper palaeolithic site of Baghor I. A subsequent microwear study of the site Baghor III (not far from Baghor I) (Sinha, 1989) has thrown light on the subsistence activities of this phase. The study identified the different kinds of activities that the stone tools found at the site were used for. Some of these activities, such as boring, scraping, and whittling, were probably related to craft work. Others, such as cutting, slicing, piercing, and chopping, could have been associated with food processing, hunting, or craft work. Microwear analysis identified the proportion of tools used on vegetal materials, those used for processing nonvegetal material, and those used to work on wood or bamboo to make hunting and gathering gear. Some tools showed a kind of wear and polish that indicated they had been hafted onto handles.

UPPER PALAEOLITHIC CHERT BLADES FROM THE NARMADA VALLEY

PRIMARY SOURCES

Upper palaeolithic tools

A blade is a flake tool, the length of which is more than twice its width. A blade with more or less even, parallel sides is known as a parallel-sided blade. A burin is a small tool made on a blade. It has a sharp but thickset working border, similar to that of a modern screwdriver. Burins may have been used as engraving tools or for making grooves in wood or bone for hafting stone tools. SOURCE Sankalia [1964], 1982: 66–68

F IGURE 2.7 UPPER PALAEOLITHIC TOOLS

There are many upper palaeolithic sites in the Chhotanagpur region and the Damin area of the Rajmahal hills. These include Paisra in Munger district. Upper palaeolithic tools have been found in the various districts of West Bengal. There is not enough evidence of the palaeolithic phase in Assam and other parts of the north-east. But in the Lalmai hills of Bangladesh and in the Haora and Khowai river valleys in western Tripura, a number of tools, including typical upper palaeolithic types such as blades, burins, points, etc. made out of fossil wood have been found. Similar tools have been found in the upper Irawaddy valley in Myanmar. The upper palaeolithic cave sites of Kurnool and Muchchatla Chintamanu Gavi in Andhra Pradesh are the only places in the subcontinent where tools made of animal bones have been found in an upper palaeolithic context. In one of the caves, as many as 90 per cent of the excavated tools were made of this material. The faunal remains at the site included those of the bat, nilgai, four-horned antelope, gazelle, chital, sambar deer, barking deer, mouse deer, wild boar, tiger, leopard, jungle cat, rusty-spotted cat, spotted hyena, civet, fresh-water fish, mongoose, sloth bear, porcupine, bandicoot rat, gerbil (a rodent), mouse, bush rat, black-naped hare, grey langur, baboon, horse, ass, rhinoceros, shrew, and giant pangolin. Apart from giving valuable information about the animals that upper palaeolithic people shared their landscape with, this list also suggests that thick forests and more humid conditions prevailed in this area. Upper palaeolithic artefacts were also found in a cave at Renigunta in Chittor district of southern Andhra Pradesh. Stone tools of this phase occur at many places along the east coast of peninsular India, and their antiquity ranges between 25,000 and 10,000 years ago.

B URIN FROM MUKAT MANIPUR (WEST B ENGAL)

PALAEOLITHIC ART AND CULTS

Prehistoric art marks the beginning of the history of art. It is also an important window into the world of prehistoric people. Apart from paintings on rocks, rock art includes petroglyphs, a word used when some substance of a rock surface is removed through engraving, bruising, hammering, chiselling, or scooping. Prehistoric art can occur in permanent places (e.g., cave paintings) or can be portable (e.g., figurines). Such remains were clearly an integral and important part of community life and some of them seem to have had some sort of cultic or religious significance. In Europe, Australia, and southern Africa, there is clear and considerable evidence of upper palaeolithic rock paintings and engravings. Animals are the predominant motif, and some of the representations may have been part of hunting rituals. Female figurines known as ‘Venus figurines’ may represent fertility beliefs and rituals. In India, however, there is very little evidence of palaeolithic art. This is partly because most of the evidence must have perished over time. However, much still remains to be discovered. We may in fact have to redefine what we consider as ‘art’ in order to recognize the remains of artistic activity of prehistoric people. It has been suggested that some of the paintings at sites such as Bhimbetka go back to the upper palaeolithic period, but this is far from certain. There are problems in dating and interpreting prehistoric art, and of ascertaining if an object was simply utilitarian or whether it had some other sort of function and significance. For instance, a very damaged upper palaeolithic carved bone

object found at Lohanda Nala in the Belan valley (UP) has been identified as a mother goddess figurine by some and as a harpoon by others. Animal teeth found in a cave at Kurnool have grooves which suggest that they may have been attached to a string and worn as ornaments. A circular disc made of chalcedony at Bhimbetka and a soft sandstone disc at Maihar (south-west of Allahabad) were found in Acheulian contexts; neither seem to be tools. A piece of ostrich eggshell engraved with two panels of criss-cross designs was discovered at Patne. Four perforated beads and one incomplete bead made of ostrich eggshell came from Patne and one from the Bhimbetka rock shelters, all from upper palaeolithic contexts. Dramatic evidence of artistic-cum-cultic activity comes from Cave III F-24 at Bhimbetka, known as the ‘auditorium cave’. This seems to belong to the borderline between the lower and middle palaeolithic. A roomy tunnel, about 25 m long, leads into a hall which has three other entrances. In the middle of the cave is a large rock. The part of the rock facing the tunnel is flat and vertical. On it are seven cupules (cup-like depressions), up to 16.8 mm deep. A few metres away from this rock, at the bottom of a pit, is another huge rock. This has one single large cup mark, along with a meandering line carved on its surface. One interpretation is that the rock with multiple cupules was used as a rock gong and that the marks were made when it was hit repeatedly. It is more likely that they were deliberately made as part of some important prehistoric community ritual. The site of Baghor I in Madhya Pradesh has given fascinating evidence of an upper palaeolithic shrine dated c. 9000–8000 BCE. Here, there was a roughly circular platform made of sandstone rubble, about 85 cm in diameter. In the centre was a piece of natural stone with a striking pattern of concentric triangular laminations in colours ranging from a light yellowish red to a dark reddish brown. Archaeologists found nine other fragments of this stone, mostly on or near the platform. When the ten pieces were joined together, they formed a triangle about 15 cm high, 6.5 cm wide, and 6.5 cm thick. This triangular stone was evidently originally placed on the platform. It is interesting to note that the Kol and Baiga tribal people who live in this part of the Kaimur hills today make circular rubble platforms and worship similar triangular stones as a symbol of the female principle or as an icon of a goddess. FURTHER DISCUSSION

Ostrich eggshell beads

The ostrich (Struthio camelus sp.), the largest living bird in the world, is today found in its natural habitat only in Africa, where it teeters on the verge of extinction. However, there is clear evidence that ostriches roamed over many parts of Asia, including India, till the end of the Pleistocene or early Holocene. Ostriches may have been hunted for food, and their eggs must also have been eaten. The eggs are big—their size ranges from about 127 × 103 mm to 160 × 129 mm, with an average thickness of 1.97 mm. They weigh between 775 g to 1618 g. The shell is smooth, yellowish white, speckled with black. It is so hard, that you have to use a hammer and saw to break it. The shells could have been used as bowls or containers. Fragments of ostrich eggshell have been found in upper palaeolithic contexts in India. The first discovery was made in the 1860s in the Ken river in Banda district of Uttar Pradesh. Since then,

pieces of ostrich eggshell have been found at Patne in Maharashtra and about 50 discoveries have been made in various parts of Rajasthan, Madhya Pradesh, and Maharashtra. A few of the eggshell pieces have been dated. Patne gives a date of 25,000 years BP; Chandresal (in Rajasthan) gives two dates—38,900 ± 750 BP and 36,500 ± 600 BP; Ramnagar (in Madhya Pradesh) gives a date of over 31,000 years BP. Some eggshell pieces have patterns on them. When examined carefully under the microscope, most of these seem to be the result of natural weathering. However, the fragment found at Patne is clearly engraved with criss-cross patterns made long ago by human hands. Beads and discs for ornaments were also made out of ostrich eggshell. Some of them had a hole through which they could be strung. About 41 Indian sites have given evidence of such beads in Pleistocene contexts ranging from 39,000 to 25,000 BP. For instance, ostrich eggshell beads occur in upper palaeolithic contexts at Patne and Bhimbetka. The Patne beads have a diameter of about 10 mm and the Bhimbetka ones of 6 mm. The Bhimbetka beads were discovered in an upper palaeolithic burial in a rock shelter, on the neck of the skull of a buried man. He must have been wearing a necklace with different kinds of beads; the others had decayed, but the two ostrich eggshell beads survived. Making such beads must have required considerable skill and care, and some scholars have tried to replicate them experimentally. G. Kumar worked with heavily weathered ostrich eggshell and used mesolithic tools, drilling through both sides, to produce two perforated beads. It took him 10–12 minutes. R. G. Bednarik used fresh ostrich eggshell. He found that it was best to work with tools made of coarse-grained quartizites and quartz, and managed to drill through the shell of a complete egg in 70–90 seconds. Through experimentation, he also reconstructed the process whereby beads of this material must have been made. Although the number of surviving beads is small, these must represent a very small proportion of those made and used by prehistoric people. Small beads could not have achieved a decorative result singly or in small numbers. The role and function of such beads must have been nonutilitarian, symbolic, or ideological. They must have been produced with such care and perfection because they were imbued with important cultural meaning. The beads also display an appreciation of an essentially abstract form. Beads made of ostrich eggshell have also been found in upper palaeolithic contexts in Siberia, inner Mongolia, China, and Africa. Clearly, ornaments made out of this material were the fashion in many parts of the prehistoric world. Bushmen of southern Africa are known to have used ostrich eggshell for making beads and as water vessels till recently. SOURCE Bednarik, 1997

F IGURE 2.8 DECORATED OSTRICH EGGSHELL OBJECTS

THE LIFE-WAYS OF PALAEOLITHIC HUNTER-GATHERERS

The life-ways of palaeolithic people living in different parts of the subcontinent were based on their adaptations to their specific environments. However, there were some basic similarities in the lives of these hunting-gathering communities. Ethnographic studies of modern hunter-gatherers can supplement the information from archaeology, although caution has to be exercised while drawing parallels and conclusions. Palaeolithic people lived in shelters made of rock, branches, grass, leaves, or reeds. More and less permanent settlements can be identified and some sites represent specific kinds of activities. Habitation sites such as Bhimbetka and Hunsgi give evidence of continuous occupation over centuries. Other sites indicate temporary camp sites, where people came, lived for some part of the year, and then moved on. Still others were connected with specific activities—e.g., kill or butchery sites and factory sites. As mentioned earlier, some factory sites seem to have attracted many different communities over thousands of years. FURTHER DISCUSSION

Food resources—now and then

Due to the lack of organic plant and animal remains, archaeologists often draw on ethnographic evidence of present communities living in areas that once supported prehistoric populations. Some important case studies have tried to understand palaeolithic sites within their broader environmental and settlement contexts. K. Paddayya’s study of the settlement and subsistence patterns of the lower palaeolithic culture of the Hunsgi valley identified about 40 species of wild edible plants growing in the valley today, including fruits, berries, pods, leafy vegetables, mushrooms, and seeds. The valley does not support any large wildlife today, except perhaps the gazelle and blackbuck. But fossilized bones of wild cattle (Bos sp.) and a horn fragment of a deer were found at the middle palaeolithic site of Hagargundigi on the Bhima river, about 80 km to the northeast. Kodekal, a neolithic site situated 8 km from the Hunsgi valley, yielded remains of three species of deer (the barasingha, gazelle, and spotted deer). It is reasonable to assume that thousands of years ago, such animals were present in the Hunsgi valley as well. The valley still supports a variety of small mammals, birds, reptiles, and aquatic animals. These include the hare, porcupine, birds such as the sandgrouse, partridge, and quail, reptiles such as the monitor lizard, many varieties of fish, and several types of insects. Some of these resources are routinely exploited for food by local communities living in the area today. The present flora and fauna of the Hun-sgi valley gives us an idea of the range of wild plant and animal food available to prehistoric people who lived in this area thousands of years ago. Of course, in those times the area must have had a much thicker vegetation of savannah woodland and must have supported a much richer range of flora and fauna. Paddayya suggests that in view of the fact that the plant resources of the area shrink in the dry summer months, prehistoric people must have had to rely more on hunting animals for food during that period. M. L. K. Murty’s study focused on present-day hunting-gathering tribes of Andhra such as the Yerukulas, Yanandis, Chenchus, and Boyas, as well as incipient agricultural groups such as the Gonds and Konda Reddis. These communities still depend on wild forest food, small game, reptiles, riverine and sea fauna, insects, and honey. Murty listed about 80 edible wild plants used by these people, including fruits, berries, seeds, tubers, pods, pulps, and vegetables. He pointed to a broad congruence of the location of prehistoric hunter-gatherer sites and those inhabited by present-day tribal communities relying significantly on hunting and gathering. This indicates that the ecological niches that were exploited by prehistoric communities who lived by foraging and hunting still manage to support communities who rely on similar subsistence strategies. SOURCE Paddayya, 1985; Murty, 1985

The basic social structure of palaeolithic hunter-gatherers may have corresponded in some ways to what anthropologists call a ‘band society’, although caution always has to be exercised while invoking ethnographic parallels. Bands are small communities, usually consisting of less than 100

people. They tend to be mobile or nomadic to some extent, moving from one place to another, depending on the seasonal availability of the animals they hunt and the plant food they gather. Members of a band are usually related to each other through kinship, and their division of labour is based on age and sex. The exchange of goods is based on rules of reciprocity, not on commercial exchange. Within the band, no single person or persons ‘owns’ the natural resources they all depend on. There are no institutions of formal government, no formal or permanent leaders, not even the powerful chiefs seen in more complex tribal societies. The behaviour of members of the group is not regulated by force but through customs, norms, and social etiquette. One of the stereotypes about the life of hunter-gatherers is that theirs was a constant, relentless struggle for survival with little or no leisure time. The material desires and wants of palaeolithic humans must have been relatively limited and their technology did not permit them to hoard food beyond a point. These two factors meant that their subsistence-related activities ceased when they had obtained enough food. This must have given them some time for other kinds of activities. Ethnographic evidence in fact shows that not all modern hunter-gatherers live a hand-to-mouth existence and many of them have plenty of leisure time to sleep, chat, play games, and relax. Another commonly held view is that hunting-gathering is an inefficient mode of subsistence. This can be questioned on the basis of the long history of this mode of subsistence and its continuation (of course on a much reduced scale) even into our own time. Further, ethnographic studies have shown that many hunting-gathering groups do not fully exploit the natural resource potential of their area and that they consciously practise sensible restraint in their exploitation of the environment in order to conserve its resources. Modern hunter-gatherers tend to obtain a significant amount of their food through gathering rather than hunting. This suggests that the ‘hunting’ part of the term ‘hunter-gatherer’ has perhaps been overemphasized by scholars and the ‘gatherer’ part neglected. This conclusion has important implications for understanding subsistence patterns as well as gender roles and relations in palaeolithic societies. In most modern hunting-gathering communities, men hunt and women gather food, and a similar division of labour probably existed in palaeolithic times. But if plant food had a greater dietary importance, it can be inferred that women must have contributed in a major way to the subsistence base of palaeolithic communities. The artistic, social, and cultic implications of some of the specimens of palaeolithic art have already been mentioned. Modern hunter-gatherers regard themselves as part of a larger world of nature because of their daily and direct encounter with it. Animals, plants, and aspects of the landscape may be treated as kin or foe; they may be worshipped or may form the focus of rituals. Since modern hunter-gatherers maintain some degree of contact with more complex societies, it would be a mistake to assume that prehistoric people had identical beliefs. However, it is possible that there were some very broad similarities arising out of a similar type of subsistence base. The Mesolithic Age MESOLITHIC SITES

The Pleistocene geological era made way for the Holocene about 10,000 years ago. Many environmental changes took place during this transition and there are detailed profiles of climatic patterns for some parts of the subcontinent. For instance, an analysis of soil samples from the site of

Birbhanpur in West Bengal shows a trend of increasing aridity. On the other hand, the study of the salt lake sediments and pollen grains at Didwana in western Rajasthan suggests higher rainfall at this point of time. In eastern Madhya Pradesh, the climate of the early and middle Holocene seems to have been wet and warm, with heavy rainfall in the summer monsoon months and moderate levels of rainfall in winter. A drier spell seems to have set in about 4,000–3,000 years ago.

MAP 2.4 SOME EARLY MESOLITHIC SITES

Towards the end of the Pleistocene or beginning of the Holocene, there were certain changes in the stone tool kits of prehistoric people. People started making and using very small tools referred to by prehistorians as microliths. At sites such as Patne, where there is a long and continuous stratigraphic sequence of prehistoric occupation, the gradual decrease in the size of stone tools can be seen very clearly. The term epi-palaeolithic is sometimes used for the transitional stage of tools that are smaller than those typical of the upper palaeolithic, but smaller than microliths. Changes in tool kits

must have been related to changes in environmental factors, but such detailed connections have not been fully worked out. The term mesolithic is generally used for post-Pleistocene (i.e., Holocene) hunting-gathering stone age cultures marked by the use of microliths. It is not, however, easy to define or identify this phase with precision. Sites such as Patne (in Maharashtra) and Fa Hien Lena, Batadomba Lena, and Beli Lena (in Sri Lanka) have given evidence of microliths in late Pleistocene contexts. Further, microliths are known to have been made and used well into the historical period. The mesolithic economy, like the palaeolithic, was still essentially based on hunting and gathering, but some sites have given evidence of the domestication of animals. Mesolithic sites reflect different levels of sedentariness. Some seem to have been permanent or semi-permanent settlements, or at least settlements that were repeatedly inhabited over long periods of time. Pottery is absent at most mesolithic sites, but it occurs at Langhnaj in Gujarat and in the Kaimur region of Mirzapur (UP). PRIMARY SOURCES

Microliths

Microliths range in length from under 1 cm to 5 cm. The tools are mostly made on short parallelsided blades made of crypto-crystalline silica stone such as quartzite, chert, chalcedony, jasper, and agate. Microliths include miniature versions of some of the upper palaeolithic tool types such as burins, points, and scrapers. But there is also the introduction of tools in regular geometric shapes such as lunates (crescents), triangles, rhomboids, trapezes, and trapezoids. Microliths are usually classified into ‘geometric’ and ‘non-geometric’ types. What was the use of such tiny tools? This question can be answered by supplementing the archaeological data with ethnographic evidence from communities in different parts of the world who still make and use such stone tools in their daily lives. Some microliths may have been used as tools in themselves, but many must have been hafted, singly or in large numbers, onto wooden or bone handles to make composite tools. In some instances, the original hafts have survived. Microliths could have been used to make spearheads, arrowheads, knives, daggers, sickles, and adzes. It is possible that poison was applied to microlithic tips and barbs to add to the lethal effect of the weapons. Microliths were also embedded in a wooden matrix to make sickles for harvesting plants. SOURCE Sankalia [1964], 1982: 69–77; Misra, 1974

F IGURE 2.9 MICROLITHS

One of the features of the Indian mesolithic phase is the spread of settlements to new ecological niches (for site details, see Allchin and Allchin, 1997: 88–110; Chakrabarti, 1999: 98–110). This is generally seen as a result of an increase in population due to more favourable environmental conditions as well as technological innovations. There is a calibrated range of dates from various mesolithic sites, e.g., Bhimbetka (6556–6177 BCE; 4895–4580 BCE), Baghor (7416–6622 BCE;

4246–3991 BCE), Bagor (5418–4936 BCE; 4575–4344 BCE), Sarai Nahar Rai (9958–9059 BCE), and Paisra (6377–6067 BCE).

MICROLITHS FROM VARIOUS SITES

The transition from a hunting-gathering stage to the beginnings of settled agriculture can be traced at Chopani Mando in the Belan valley (Sharma et al., 1980). Excavations revealed a 1.55 m thick occupational deposit, divided into three periods. The first was epi-palaeolithic, while the second and third were clearly mesolithic. Period II was divided into two phases—IIA and IIB. Period IIA had non-geometric microliths such as blades, points, scrapers, and borers, mostly made of chert. In Period IIB, there were a large number of geometric microliths. The microliths continued into Period III, which was also marked by handmade pottery with cord-impressed patterns, anvils and hammer stones, querns and mullers (used for grinding and food processing), and ring stones. There were bones of wild cattle and sheep/goats. Pieces of burnt clay with reed impressions showed that the mesolithic people of Chopani Mando lived in wattle-and-daub huts. The excavations revealed the

outlines of two round huts belonging to Period IIA and five round huts of Period IIB. In Period III, there were remains of 13 round and oval huts, clustered very close to each other. The round huts had an average diameter of 3.3 m, and the oval ones 4.7 × 3.3 m. Outside the huts were three hearths and traces of what seemed to be the bases of storage bins made of bamboo and clay. Wild rice is reported from late mesolithic levels at this site. The three excavated sites of Sarai Nahar Rai, Mahadaha, and Damdama lie very close to each other. Sarai Nahar Rai (in Pratapgarh district, UP) is located on the banks of a dried oxbow lake which marks an old course of the Ganga. Geometric microliths were found here, along with shells and animal bones (of bison, rhinoceros, stag, fish, and tortoise). Within the habitation area, there were 11 human burials in oblong pits—those of 9 men, 4 women, and a child. The age of the men was estimated to be in the range of 16–35 years, and that of the women 15–35 years. One of the buried skeletons had an arrow embedded in its ribs. A multiple burial contained the remains of four persons. Microlithic tools, animal bones, and shells were placed in graves as grave goods. An analysis of the skeletal material revealed that the dental health of the people was on the whole good, but that some of them suffered from osteo-arthritis. The mesolithic level at Sarai Nahar Rai has been dated c. 8400 ± 150 BCE by the radiocarbon method. Mahadaha is also on the banks of an oxbow lake. Excavations revealed a 60 cm thick occupational deposit and distinct areas associated with habitation and butchering. The microliths were made of chert, quartz, chalcedony, crystal, and agate, all of which must have been brought over fairly long distances across the river from the Vindhyas. Twenty-eight burials of thirty individuals, including two instances of a man and woman buried together, were found within the habitation area. The burials were elliptical and their base sloping. The grave goods included microliths, shells, burnt pieces of animal bones, bone arrowheads and rings, and ochre pieces. The bones found in the butchering area included those of wild cattle, hippopotamus, deer, pigs, and turtles. Thousands of animal bones were found in the lake area. The mesolithic people of Mahadaha were tall (men were up to 1.90 m and women 1.62–1.76 m). Their dental health was good, but many of them suffered from osteo-arthritis. Of the 17 males, 7 females, and 3 children identified in the skeletal record, 5 represented persons less than 18 years old, 6 belonged to the 18–40 age group, and only 1 (a female) represented a person between 40–50 years old. These figures provide an idea of average life expectancy. Damdama is situated at the confluence of a small stream belonging to the Sai river system. Within the 1.5 m thick occupational deposit, excavators discovered microliths, bone objects, querns and mullers, anvils, and hammer stones. There were hearths, patches of burnt floor plaster, charred wild grain, and animal bones. There were 4 multiple burials among the 41 human burials. In one of the graves, an ivory pendant was found among the grave goods. Dates for Damdama fall within the early 7th millennium BCE. Recently, domesticated rice has been reported from mesolithic levels at this site. FURTHER DISCUSSION

Animal bones at mesolithic sites

Bones of wild and domesticated animals have been found at some mesolithic sites. However, there is considerable difference of opinion among experts when it comes to the identification of animal species. Bagor (Rajasthan): The mesolithic context here dates to the 5th and 4th millennia BCE. There are differences in opinion about the identification of the animal bones. P. K. Thomas identified domesticated cattle (15.7 per cent of the identifiable bones in the assemblage) and sheep/goats (64.4 per cent; it is not always possible to distinguish between sheep and goat bone fragments). The site also yielded bones of wild boar and pig (3.7 per cent), buffalo (0.8 per cent), blackbuck and gazelle (4.4 per cent), spotted deer (4.8 per cent), sambar (4.3 per cent), hare (0.6 per cent), Indian grey mongoose (0.8 per cent) and Indian fox (0.5 per cent). Other species include porcupine, rat, tortoise, fish, and frog. D.R. Shah does not mention the occurrence of such domesticated species, but lists river turtle and monitor lizard. Tilwara (Barmer district, Rajasthan): According to V. N. Misra, the late meso-lithic level (its dates are uncertain) gave evidence of wild goat, a canid (jackal or dog), pig, spotted deer, hog deer, mongoose, and domesticated humped cattle. Thomas only reported cattle and goat/ sheep from this site. Langhnaj (north Gujarat): The mesolithic context in which the animal bones were found was dated 2550–2185 BCE. Only wild animals were represented. These include a canid (probably wolf), mongoose, rhinoceros, wild boar, chital, hog deer, swamp deer, nilgai, and blackbuck. The presence of wild buffalo or wild cattle has also been suggested. V. N. Misra suggested that the climate of the area during mesolithic times must have been arid. However, the presence of the rhinoceros and the possible presence of the water buffalo go against this theory. The rhinoceros is known to prefer large stretches of marshland and grassland. The animal bones at Langhnaj suggest that the area was covered by a combination of savannah and forest with interspersed wetlands. Kanewal (north Gujarat): This site has given evidence of bones of rhinoceros, buffalo, spotted deer, swamp deer, nilgai, and wild boar. Bones of domesticated cattle, sheep, and goats have also been identified. The occurrence of camel bones is interesting and shows contact with people using these animals. Loteshwar and Ratanpur (north Gujarat): Bones of domesticated sheep, goats, and cattle have been recently reported from these sites. Adamgarh (MP): Bones of domesticated cattle, sheep, goat, pig, and dog have been reported here, along with those of wild animals such as spotted deer, barasingha, sambar, porcupine, hare, lizard, and a species of the genus Equus. There are two very different radiocarbon dates for these finds, one falling in the 6th millennium BCE, the other in the 1st millennium BCE. Due to the uncertain dates and stratigraphy of the finds, the evidence of early animal domestication at Adamgarh has been questioned. Bhimbetka (MP): This site has given bones of domesticated cattle along with those of wild animals such as barasingha, hog deer, and rhinoceros. It is interesting to note that mesolithic

paintings at this site have representations of Indian humped cattle (zebu) as well as its wild progenitor, Bos namadicus. Sarai Nahar Rai, Mahadaha, and Damdama (UP): The faunal evidence from these sites is controversial. K. R. Alur identified wild cattle and sheep/goat. According to U.C. Chattopadhyaya, there is no evidence of sheep or goats, wild or domesticated, at these three sites. Thomas and Joglekar identified over 30 species including cattle, gaur, goat, gazelle, chital, sambar, barking deer, mouse deer, rhinoceros, wild boar, pygmy hog, hippopotamus, elephant, wolf, jackal, sloth bear, porcupine, rat, and bandicoot. No domesticated animals are represented. Chopani Mando (UP): Bones of wild cattle and goat/sheep are reported from this site. SOURCE Chattopadhyaya, 2002

NEW DIRECTIONS IN RESEARCH

Graves, subsistence, and settlement patterns

Umesh C. Chattopadhyaya has explored the connections between the emergence of formal burials among hunter-gatherers and the subsistence patterns and settlements of the mesolithic Ganga valley. His study is based on the faunal remains and burials at Sarai Nahar Rai, Mahadaha, and Damdama. An initial study of Sarai Nahar Rai suggested that the site was occupied seasonally and that in summer, groups living in the Vindhyan stretches migrated into the Ganga valley due to water and food scarcity. Excavations at Mahadaha and Damdama led to a questioning of this hypothesis, largely on the basis of the thickness of the occupational deposits and the occurrence of many heavy non-transportable grinding stones. Chattopadhyaya has carried forward this questioning. This area had diverse food resources. The plant remains at Damdama suggest the availability of various types of wild edible plants. The animal bones found at these sites included those of wild animals exploited for food. The subsistence base of the mesolithic people had a special emphasis on the hunting of swamp deer (Cervus duvauceli) and hog deer (Axis procinus) and must have been supplemented by an intensive use of aquatic resources such as tortoise and fish. Mahadaha and Damdama yielded teeth of hog deer and swamp deer. As the breeding season of these animals is known, April and July can be identified as the months of birth of the two species respectively. On the basis of the analysis of some of the teeth remains, the age of the animal at the time it was killed and the month in which this happened can be ascertained. This gives us an idea about the months/ seasons in which the sites were occupied by humans. Chattopadhyaya’s analysis of such teeth remains shows that Mahadaha and Damdama were occupied during summer as well as winter. The presence of a commensal species (one that depends on humans for

its food) such as bandicoot rat (Bandicota bengalensis) at these sites is also significant, as this species cannot establish itself at a habitation site unless food is available to it all year round. The evidence of burials at the three sites confirms the above conclusions. All three display similar burial practices—the bodies were buried in shallow rectangular graves, usually in an extended position. Male burials outnumber female burials; there are a few instances of child burials at Mahadaha and Damdama. Differences in grave goods suggest some level of social ranking. At Mahadaha, the cemetery-cum-habitation area revealed 35 pit-hearths containing burnt clay, ash, and charred animal bones, which seem to have been associated with funerary rituals. What is most significant for Chattopadhyaya’s argument is the orientation of the burials. With a few exceptions at Damdama, the graves were broadly aligned west–east or east–west. Archaeological and anthropological evidence suggests that it is likely that they were aligned towards the point of sunrise or sunset at the time of burial. The precise points of the east–west orientation would have varied to some extent in summer and winter. Many of the burials at these sites fall broadly within the calculated range of the annual solar path across the horizon, suggesting that burials took place both in summer and in winter. This too indicates that the sites were occupied all year round. Anthropologists have identified a close relationship between the designation of formal areas for the disposal of the dead and the existence of corporate group rights over critical resources. These rights are based on lineal descent from deceased ancestors. The mesolithic burials of Mahadaha are suggestive of this sort of situation. But what were the reliable but restricted resources over which these people may have tried to stake their claim through descent? Chattopadhyaya suggests that they consisted of aquatic resources such as tortoise and fish, which have rich nutritional value and are very productive and reliable food resources. The growing population in the Ganga valley during the mesolithic phase may have led to competition and conflict over these resources. This may have been the impetus for people to come together and function as corporate groups and to stake their claim to territory through burial practices and burial symbolism. SOURCE Chattopadhyaya, 1996

Rock shelters excavated at Lekhakia (in Mirzapur district of southern UP) have yielded blade tools and microliths. There is a clear tendency of tools to become progressively smaller in the upper levels of the deposit. Burials were found, and so was pottery. Baghai Khor is another rock shelter site in the same area. This has a pre-ceramic and a ceramic microlithic phase. Two extended burials were identified, the first belonging to the pre-ceramic phase and the second to the ceramic phase. A 105 sq m section of a mesolithic floor was excavated at Paisra. Apart from micro-liths, there was evidence of large and small fireplaces positioned very close to each other. The thinness of the deposit suggests a short period of mesolithic occupation. Birbhanpur is close to the Damodar river in Burdwan district in West Bengal. Mesolithic stone tools made of quartz, some of chert and chalcedony, were found here. This seems to have been both a

habitation and a factory site. A study has shown that the climate during the mesolithic phase at Birbhanpur was drier than in the immediately preceding phase, which was more wet and humid. NEW DIRECTIONS IN RESEARCH

The journey to get chalcedony

The general picture of the mesolithic phase is one of small, mobile bands of people exploiting the resources available in their environment with their microlithic tools. Recent studies have indicated that the reality was far more complex. Gurcharan S. Khanna’s study of chalcedony at Bagor in eastern Rajasthan has highlighted how a focus on this one raw material reveals patterns of movement, exchange, and resource procurement. Here are some of the findings: The 1. microliths of Bagor are mostly made of chert and quartz, but some are of chalcedony. Tools made of quartz form about 79 per cent of the total stone tool assemblage, those made of chert about 20 per cent, and chalcedony tools about 1 per cent. However, if only the finished tools are considered, the percentages are different—chert tools represent the greater proportion and quartz tools are much less. The proportion of chalcedony finished tools is variable but not insignificant; this stone was preferred for making smaller blade tools. Chalcedony 2. is a member of a group of crypto-crystalline minerals which includes chert, jasper, opal, flint, agate, and carnelian. Compared to other rocks, chalcedony has higher water content and a fibrous grain structure. These qualities make it easy to control flaking and to make standardized tools, especially small ones. This is why mesolithic people may have preferred chalcedony over other stones for making such tools. While 3. chert and quartz are available within walking distance, the chalcedony found in this area is of poor quality. Good quality chalcedony is available in the Deccan Traps, 90 km to the south-east of Bagor. That is probably the area from where the mesolithic people obtained it. There 4. could have been another reason for moving south-eastwards seasonally. Bagor is located on the eastern side of the Aravalli hills, which is an intermediate zone as far as rainfall is concerned. The mesolithic people of Bagor may have moved southeastwards in the dry post-monsoon season in order to find grazing land for their animals (there is evidence of animal domestication at this site). The 5. people of Bagor could have obtained chalcedony from the Deccan Traps either through direct procurement or through exchange with intermediate groups. In 6. the later part of the mesolithic phase, there is evidence of copper. It is possible that the introduction of copper at this site had to do with the interaction of its people with the farmers and metallurgists of Ahar, a settlement which lay at considerable distance to its south.

SOURCE Khanna, 1993

Bagor (in Bhilwara district of eastern Rajasthan) is one of the best documented mesolithic sites. It is located on a sand dune, about 25 km west of Bhilwara in Rajasthan, close to the Kothari river. The three occupational levels represented continuous human occupation over more than 5,000 years. Period I (c. 5000–2800 BCE, according to radiocarbon dates) was mesolithic, Period II (c. 2800– 600 BCE) chalcolithic, and Period III (c. 600 BCE–200 BCE) gave evidence of iron. Microliths were found in the greatest numbers in Period I but continued into the later phases as well. The microliths of Period I were mostly made of locally available chert and quartz. Most of them were

made on blades and they included a large number of geometric microliths such as triangles and trapezes. House floors paved with stone slabs were found, and in some places, there was evidence of roughly circular arrangements of stone that may have marked the outlines of shelters. Certain stone-paved areas with a large number of animal bones were probably butchering areas. Only one burial was unearthed and there was no definite evidence of grave goods. Other discoveries included ring stones (perhaps used as hammer stones to make microliths), pieces of red ochre, querns, and rubbing stones (for grinding food). Bones of wild animals included those of wild cattle, two kinds of deer, pigs, jackals, rats, monitor lizards, turtles, and fish; bones of domesticated sheep/goats and cattle were also reported. There is a possibility that small bits of pottery found at the site may belong to the mesolithic phase. Microliths have been found in the valleys of the Tapi, Narmada, Mahi, and Sabar-mati. One of the important sites is Langhnaj. The occupational deposit here was divided into three periods. Period I was mesolithic and yielded microliths, human burials, bones of wild animals, and some potsherds.

P AINTING OF BOAR, B HIMBETKA (AFTER NEUMAYER, 1983)

DANCERS, LAKHAJOAR

Adamgarh hill near Hoshangabad has already been mentioned among the palaeolithic sites of central India. Its upper layers represented a mesolithic level, which in turn made way for a neolithic–chalcolithic one. Shells found between 15–21 cm from the top of the mesolithic deposit were dated by the radiocarbon method to c. 5500 BCE, so the mesolithic level belongs at least to the 6th millennium BCE. Thousands of microliths were found here, mostly made of chert, chalcedony, jasper, and agate, raw materials which are available in the riverbed of the Narmada about 2 km away. Geometric microliths (triangles and trapezes) were very common. Mace heads or ring stones and hammer stones were also found. The wild animal bones comprised those of the hare, lizard, various kinds of deer, horse, and porcupine. Bones of domesticated cattle, sheep, goat, dog, and pig have also been reported, but this evidence has been questioned. This site has given evidence of pottery at mesolithic levels.

Baghor II in the Son valley has already been mentioned in the discussion of the palaeolithic sites of central India. This site also has a mesolithic phase. The tools are of chert and chalcedony, and geometric microliths occur. Fragments of grinding stones, one hammer stone, and pieces of red ochre were found. There were very few finished stone tools, and as much as 96.7 per cent of the total mesolithic lithic material that was excavated consisted of waste material of stone tool working. This suggests that the tools were made here and taken away to other places. The location of five or six large shelters can be identified by a series of post-holes. Three hoof prints of a sambar deer were preserved in the excavated deposit. Bhimbetka has also already been mentioned in the section on palaeolithic sites. The site shows a gradual reduction in the size of tools. Mesolithic tools include blades and geometric microliths like triangles, trapezes, and crescents. Quartz was used a great deal in the palaeolithic stage, but in the mesolithic phase there was a shift to chalcedony. Bhimbetka is famous for its mesolithic paintings, which will be discussed further on. In peninsular India, microlithic sites found in the vicinity of Mumbai seem to represent coastal mesolithic communities who exploited marine resources for food. Microliths have been found in other parts of Maharashtra as well. Further south, the microliths are mostly made out of milky quartz. They have been found at Jalahalli and Kibbanhalli near Bangalore in Karnataka, in Goa, and at Nagarjunakonda (in southern AP), and Renigunta (in Chittor district, AP). Microliths occur at many places along the east coast of India and seem to mark camps of mesolithic fishing communities. South of Chennai, tiny stone tools, mostly of quartz and chert, have been found on old sand dunes known as teris. On the Visakhapatnam coast, stone tablets and ring stones have been found at sites such as Chandrampalem, Paradesipalem, and Rushikonda. Similar stones are used today by local fishermen in the area as net sinkers. Apart from the coastal areas, rock shelters, flat hilltops, river valleys, and lakesides were also inhabited during the mesolithic phase.

P OLYCHROME BOAR, B HIMBETKA

B ULL, JAORA (AFTER NEUMAYER, 1983)

The evidence from mesolithic sites from different parts of the subcontinent suggests movement and interaction among communities. Factory sites located at sources of raw materials must have been meeting grounds for different groups. The fact that mesolithic tools found north and south of the Ganga are made of the same kinds of stone indicates that either the raw materials or the tools themselves were moved across the river. The mesolithic people of Sarai Nahar Rai, Damdama, and Mahadaha would have had to travel over 75 km to reach the stone resources of the Vindhyas. Clearly, the communities living in the northern alluvial plain and the hill people of the northern fringes of the Vindhyas must have been interacting with each other. In later times, mesolithic communities must have interacted with early agriculturists who lived in their neighbourhood. There are many instances of temporary mesolithic camp sites in various parts of the subcontinent, but sites such as Sarai Nahar Rai, Damdama, Mahadaha, and Chopani Mando were inhabited continuously. This can be inferred from the nature of the archaeological evidence and also certain more specialized studies of the faunal material. The evidence from several sites of formal, ceremonial burials, with the bodies usually laid out in a west–east direction (sometimes the other way around) with grave goods suggests rituals associated with death. The presence of grave goods is often taken as an indication of some sort of belief in afterlife. This may well be so, but there is ethnographic evidence of societies in which certain belongings of the deceased are considered to bring bad luck to the living, and these are therefore buried along with the body. Instances of jewellery found on the body suggest a custom of adorning the body before burial, and may indicate high-rank individuals within the community. THE MAGNIFICENCE OF MESOLITHIC ART

There are very few examples of portable mesolithic art. A chert core engraved with an interesting geometric design was found at Chandravati in Rajasthan. It is assumed to be mesolithic because lots of microliths have been found at the site. A few engraved bone objects have been discovered at sites such as Bhimbetka. A human tooth with faint geometric marks on it was found in a jaw fragment along with some other teeth, and is currently kept at Deccan College, Pune. The first rock paintings in India (and in fact anywhere in the world) were discovered by A. C. L. Carlleyle, an assistant surveyor with the Archaeological Survey of India in 1867–68 at Sohagighat in the Kaimur hills in the present Mirzapur district (UP). Today, over 150 mesolithic rock art sites have been found in various parts of the subcontinent and central India has an especially rich concentration of sites. The paintings are an important source of information regarding the lives of mesolithic communities and show striking thematic similarities across the country.

ANIMAL PAINTINGS: KATHOTIA, LAKHAJOAR (AFTER NEUMAYER, 1983)

In 1957, the archaeologist V. S. Wakankar noticed the Bhimbetka rocks out of a train window while travelling from Bhopal to Itarsi and got off at the nearest railway station to explore. This is how one of the most magnificent rock art sites in the world was discovered. Bhimbetka is one of seven hills marked by a very picturesque natural environment. There are 642 rock shelters here, nearly 400 of which have paintings, engravings, and bruisings. Their style, theme, and worn state indicate that they belong to old times. Mesolithic paintings have also been found at other sites in Madhya Pradesh such as Kharwar, Jaora, Kathotia, and Lakhajoar.

HUNTING SCENE, LAKHAJOAR

F EEDING A PIG AND PREPARING FOOD, KATHOTIA (AFTER NEUMAYER, 1983)

The Bhimbetka paintings have been studied by V. S. Wakankar, Yashodhar Mathpal (1974), and Erwin Neumayer (1983), and the results of their research illuminate many aspects of the lives of the painters. Mathpal identified three main phases of the rock paintings, with further sub-phases within these. The first five sub-phases are mesolithic, the sixth is transitional, and the last three belong to the historical period. Sixteen colours or shades can be identified, with white and light red used most often. The colours were made from minerals which were ground and then mixed with water or some other substance like animal fat, marrow, or egg white. The red was made out of iron oxide (geru), white from limestone, and green may have been made from green chalcedony. Some paintings are monochrome (in one colour), while others are polychrome (in more than one colour). The handles of brushes must have been made from twigs, and the brush itself out of squirrel tail, animal fur, or semal (silk cotton).

LAKHAJOAR: FISHING SCENE

FAMILY SCENE (AFTER NEUMAYER, 1983)

As at most mesolithic rock art sites, animals dominate the scenes at Bhimbetka. Twenty-nine species of animals are depicted, including the chital (this occurs most often), leopard, tiger, panther, elephant, rhinoceros, antelope, deer, and squirrel. Different kinds of birds, fish, lizards, frogs, crabs, scorpions, and small centipedes also make an appearance. Although the Bhimbetka hillsides are still home to many animals, the elephants, rhinoceroses, lions, wild buffalo, gaur, and blackbuck depicted in the mesolithic paintings have disappeared. It is interesting to note that no snakes are depicted in Indian mesolithic paintings, here or elsewhere.

HUNTER WITH BASKET/NET FILLED WITH ANIMALS, JAORA

In mesolithic art at Bhimbetka and elsewhere, animals are represented on their own or as part of hunting scenes. Hunters hunt singly or in groups, sometimes wearing masks and headdresses crowned

with antlers and horns. They are adorned with ornaments such as necklaces, bangles, wrist bands, elbow bands, and knee bands with tassels. Some are unarmed; others carry sticks, spears, bows and arrows, or slings. The hunters are sometimes accompanied by dogs. There are scenes showing traps and snares, others of animals running after hunters. While some of the animal figures are rather abstract, many of them are very realistic. Animals are sometimes shown in outline; in other instances their bodies are decorated with designs. A few paintings are in the ‘x-ray style’, showing the inner organs, including foetuses in the womb of female animals. Apart from hunting scenes, animals appear in more peaceful, sympathetic scenes such as those depicting pregnant animals, a panther or tiger with cubs, stag, and chital running after a fawn, grazing buffaloes, rabbits hopping, and monkeys leaping about. There is a lot of movement in the scenes. There are also fantastic animals—the famous Bhimbetka ‘boar’ has the body of a boar, but a snout like a rhinoceros, the underlip of an elephant, and the horns of a buffalo. Mesolithic paintings at Bhimbetka and other sites also depict men and women, young and old. Male figures often look like matchsticks, women are sometimes given fuller forms. Some men wear loincloths, probably made of leaves, animal skin, or pieces of tree bark. Men wear their hair loose, women braided. Some figures are broad and decorated with geometric designs, and from their attitude seem to represent men of authority. Masked dancers (referred to by prehistorians as ‘dancing sorcerers’) may represent ritual specialists. Hand, fist, and finger prints are also found, similar to those that people make on their houses these days on auspicious occasions. The Bhimbetka paintings reflect a division of labour on the basis of gender. Men hunt and women are shown gathering and preparing food, for instance grinding food on querns. It is difficult to identify what sort of vegetable food was being processed. Basket-like containers must have been used to store food, but no pottery is depicted. Dry gourds and leather bags may have been used to hold water. There are scenes of people collecting fruit and honey. Some scenes depict sexual activity, others show people dancing. The dancers convey a sense of rhythmic movement; occasionally they lose their balance and fall. In Europe, most prehistoric paintings tend to be located in dark, relatively inaccessible parts of the caves, but those in Indian rock shelters are usually in well-lit areas, easily seen. The best paintings were not, however, made in shelters that were living spaces. Some of the big paintings on high surfaces would have required scaffolding and the cooperation of many people. Such paintings, and those made in layers, suggest some kind of ritualistic significance. Prehistoric rock art sites have been found at many other places in India as well. In eastern India, over 55 rock shelters with rock art have been identified in the western districts of Orissa, especially in the Sundargarh and Sambalpur districts. Microliths found in some of the rock shelters have confirmed the fact that there was mesolithic occupation in the area. The richest area for rock paintings are the 12 rock shelters of the Lekhamoda group in the reserve forests of Chhengapahad and Garjanpahad. Excavations in one of these rock shelters revealed a cultural sequence from the mesolithic to the chalcolithic. An interesting feature of the rock art of Orissa is the co-existence of paintings and engravings in the same shelter. Also, the art is mostly non-figurative, with an emphasis on abstract patterns and decorative designs, both geometric and non-geometric. Animals occur infrequently and humans are even rarer. Kerala too has many rock art sites with paintings and carvings. One of the oldest is the cave known as Ezhuthu Guha, situated in the midst of a dense sandalwood forest in Idukki district. In the

earliest stage of rock art in this area, animals were depicted, but no humans. One problem is that although some of the paintings have been assigned to the late mesolithic phase, no mesolithic tools have been found so far in any of the Kerala rock shelters.

P AIR OF HUNTERS, B HIMBETKA

ABSTRACT PAINTING, JAORA (AFTER NEUMAYER, 1983)

Why did prehistoric people make such paintings? Probably for many different reasons—to express their creative urges, to decorate their homes, or to tell a story in pictures. Some scenes may have been picture-stories of memorable events in their lives. Others may have been connected with rituals connected with hunting or fertility. It is impossible to say whether the paintings were made by men or

women, or both. Apart from the scenes of animals and people, there are a few more enigmatic paintings. A very interesting, rather abstract painting has been found in a rock shelter at Jaora (MP). Perhaps it reflects a view of the world consisting of air, earth, and fire. But it is possible that it means something completely different. The mesolithic artist who painted it would have known for sure, but since he/she is not around to tell us, we have to use our imagination to try to unravel its meaning. CONCLUSIONS

Prehistory represents the longest part of the human past, and is associated with the emergence of anatomically modern humans and important developments in stone tool technology and subsistence strategies. Palaeo-environmental studies form the essential background for the reconstruction of the life-ways of prehistoric people. The evidence of the lower, middle, and upper palaeolithic phases in the subcontinent is gradually increasing, but still largely consists of stone tools. Mesolithic communities fanned out into new ecological niches and the evidence of rock art provides valuable information about their lives and aesthetic sensibilities. Palaeolithic and mesolithic people obtained their food through hunting and gathering. However, the animal bones found at some mesolithic sites indicate that the beginnings of animal domestication can be traced to this phase. The major transition from hunting-gathering to food production based on the domestication of plants and animals is associated with the next cultural stage—the neolithic.

www.pearsoned.co.in/upindersingh Further resources

Chapter Three The Transition to Food Production: Neolithic, Neolithic– Chalcolithic, and Chalcolithic Villages, c. 7000–2000 BCE Chapter outline THE NEOLITHIC AGE AND THE BEGINNINGS OF FOOD PRODUCTION WHY DOMESTICATION? THE IDENTIFICATION OF DOMESTICATION AND FOOD PRODUCTION IN THE ARCHAEOLOGICAL RECORD

THE TRANSITION TO FOOD PRODUCTION IN THE INDIAN SUBCONTINENT THE LIFE OF EARLY FARMERS CHANGES IN CULTIC AND BELIEF SYSTEMS CONCLUSIONS

A POT FROM NAL, B ALUCHISTAN

With its arid, mountainous terrain and extreme climate, the Kachi plain in the Baluchistan province of Pakistan appears at first glance to offer a rather inhospitable environment for human settlement. The higher valleys are often covered with snow for up to two months in the year. Rainfall does not usually exceed 10 cm per annum and most of it occurs in winter. However, the river valleys are dotted with many prosperous villages, and a major trade artery connecting the Indus valley with central Asia winds through. The area is inhabited by pastoral nomads and agriculturalists. Wild cereals and wild game are abundant. Farmers dam the non-perennial streams to irrigate their fields with the overflow. Wheat is the main crop, and the area is considered the ‘bread basket’ of Baluchistan. The Kachi plain is also extremely rich in ancient archaeological sites. In 1968, Sardar Ghaus Baksh Raisini drew the attention of archaeologists to a mound near his winter residence on the banks of the Bolan river, about 10 km south of Dadhar, the main settlement of Kachi district. Following up Raisini’s tip, a French archaeological mission, in collaboration with the Department of Archaeology of Pakistan, started excavating the site of Mehrgarh in 1974. The excavations eventually extended over a decade and their results radically altered the understanding of the beginnings of agriculture in the subcontinent.

The world’s first agricultural villages emerged in c. 8000–6000 BCE. West Asia was an early centre of wheat and barley farming, and the earliest domesticated animals in this area included sheep and goats. Early neolithic villages have been identified at Jericho and ‘Ain Ghazal in Jordan, Tepe Guran, and Ali Kosh in Iran, Çatal Hüyük in Turkey, and Cayonu in north Syria. In Southeast Asia, excavations at the Spirit Cave in Thailand revealed 10 different plants species including almond, pepper, cucumber, betel nut, beans, and peas. Although it is not certain whether all of them were cultivated, the wide range of plant remains suggests more than a simple food-gathering community. The excavations at Mehrgarh, which gave evidence of barley and wheat cultivation, and cattle, sheep, and goat domestication, indicated that Baluchistan in South Asia was a third zone of early agriculture. There is a possibility of equally early dates for rice cultivation from the northern fringes of the Vindhyas in the state of Uttar Pradesh in India. As the evidence does not suggest any direct connections between the various zones of early agriculture, they must be considered independent centres. Within the next few millennia, the domestication of plants and animals was being practised in other parts of the world as well. Hemudu in south China has given evidence of rice cultivation and the domestication of water buffalo, dog, and pig in late 6th–early 5th millennia BCE contexts. By 5000 BCE, the people of Mexico were growing corn, beans, squash, gourds, avocados, and chilli pepper, and were domesticating turkeys, dogs, and honeybees. At about the same time, communities living in the Peruvian highlands were cultivating beans, gourds, tomatoes, and potatoes, and may have domesticated the llama and alpaca. In sub-Saharan Africa, the cultivation of finger millet, sorghum, rice, teff, and yams, and the domestication of sheep, goats, and cattle came to be established in various ecological niches. The primary domestication of plants and animals took place in areas where the concerned species were native, but these swiftly spread to secondary areas of domestication in different parts of the world. The Neolithic Age and the Beginnings of Food Production

The domestication of animals and plants was the outcome of a long series of collective experiments involving many generations of men, women, and children, stretching out over hundreds, perhaps thousands, of years. We will never know the names of the people who took part in these experiments and made the critical choices and changes in their strategies of obtaining food. But the processes they set in motion marked one of the greatest achievements of humankind. Archaeological evidence records a fairly late stage in the story of animal and plant domestication, when it was already well underway. Although many details of these processes still elude us, it is possible to reconstruct various aspects of the transition from hunting-gathering to domestication in different parts of the world.

MAP 3.1 CENTRES OF AGRICULTURE

The domestication of plants and animals marked a special kind of human interference in nature and a new stage in the relationship between people, plants, and animals. It involved removing plants and animals from their natural habitat, a process of selective breeding and rearing under artificial conditions under human control for purposes of human gain. There are differences between plant collection and plant domestication, and between animal keeping and animal domestication. When grain is harvested and all of it is consumed, this is a stage of food collection. If, after harvesting, some grain is consumed for food and the rest put aside and later intentionally planted, this is the stage of plant domestication. When certain species of animals are captured and kept, this is a stage of animal keeping. When wild animals are removed from their natural habitat and maintained and bred under artificial conditions by people for their profit, this is the stage of animal breeding or domestication. It is possible to identify gradual shifts in the balance of subsistence strategies from hunting and gathering towards animal rearing and agriculture. For instance, the background to the beginnings of

plant domestication was the transition from simple foraging (food collection) to complex foraging, the latter representing a stage of intensive exploitation of wild plants. The next stage was that of incipient (early) agriculture, which, over time, led to the stage of developed agriculture. In the long run, such shifts were associated with technological changes, greater food availability, a rise in population, an increase in the number and size of human settlements, and more complex social and political organization. Hundreds, probably thousands, of years must have elapsed between the initial domestication of plants and animals in an area and the increased reliance of people on these resources for their food. A distinction can be made between societies in which a small amount of food is obtained through animal and/or plant domestication and those which obtain a significant or substantial amount of food through these activities. It is the latter that can be described as food producing societies. A working definition of a food-producing society is one which meets at least half its food needs for at least part of the year through the domestication of animals and/or plants, in a context wherein animals and plants are not tied to their natural habitat. Of course, since precise statistics are unavailable for ancient societies, the extent to which a group depended on domestication for its food can only be gauged subjectively. In the classification of the stone age, the neolithic age is associated with innovations in stone tool technology, specifically the making of ground, pecked, and polished stone tools and the advent of food production. Changes in stone tools were related to shifts in subsistence strategies. Other features of the neolithic phase include the invention of pottery, a greater degree of sedentary living, the emergence of small and relatively self-sufficient village communities, and a division of labour based on sex. V. Gordon Childe coined the phrase neolithic revolution to highlight the enormous significance of these changes. This was a gradual revolution, which took place several times in different regions, with varied features and results. Why Domestication?

After thousands and thousands of years of hunting and gathering, why did some groups of people start domesticating animals and plants? One of the earliest attempts to answer this question was made by V. Gordon Childe (1952), who suggested that environmental changes at the end of the Pleistocene were the impetus towards food production. Childe argued that about 10,000 years ago, the climate in parts of West Asia became drier due to a northward shift of the summer rains. This desiccation (i.e., drying up) led to a concentration of people, plants, and animals close to water resources such as rivers and oases. This enforced closeness eventually led to new relationships of dependence between humans, plants, and animals, resulting in domestication. Childe’s theory was questioned by Robert J. Braidwood (1960), who rejected the focus on environmental change as the crucial factor leading to agriculture. He pointed out that environmental changes had occurred within the Pleistocene as well and had not led to agriculture. Braidwood argued that domestication took place in certain nuclear zones, which supported a variety of wild plants and animals that had the potential for domestication. In such areas, domestication was the natural outcome of human experimentation and people getting to know their environment better. This theory does not really explain the pressures or incentives that may have led to domestication. There is ethnographic evidence of many hunting-gathering communities who know their environment very

well and are even aware of agriculture, but do not see the point of practising it themselves. There have to be good reasons for a community to radically change its way of life. Braidwood’s theory was rejected by Lewis R. Binford (1968) on the grounds that it could not be archaeologically tested, and that there are some specific, concrete factors that can explain the beginnings of agriculture. Binford asserted that ethnographic evidence indicates that in areas where environment and population have remained constant, a stable balance between the human population and food resources is achieved and people do not have to look for new sources or strategies of getting food. Such groups in fact tend to live at food consumption levels far below the resource potential of their environment. Two factors can upset the balance between people and food: stress produced by environmental change or by demographic (population) growth. Binford identified two kinds of demographic stress—internal demographic stress, which occurs when the number of people within a community increase; and external demographic stress, caused by immigration into an area by people from another area. In the context of the origins of agriculture, Binford emphasized external demographic stress. He argued that at the end of the Pleistocene era, as a result of a rise in sea levels, people living along the coasts migrated to less populated inland areas. This upset the people–food equilibrium in inland areas and gave an impetus to the search for new strategies to increase food supplies. The problem is that evidence of a migration of people from the world’s sea coasts to inland areas at the end of the Pleistocene is lacking. Internal demographic stress may have been a factor in upsetting the people– food balance in some areas, but a question that can be raised is: can we really talk about ‘overpopulation’ and ‘food crisis’ in times when human communities were small and resources abundant? Kent Flannery (1969) shifted the focus from the search for an event that might have led to the beginnings of food production to the process of food production itself and the adaptive advantages of plant and animal domestication over foraging and hunting. He distinguished two types of food procurement systems—negative and positive feedback food procurement systems. Negative feedback food procurement systems involve a balanced exploitation and use of various food resources within an area and discourage any change. Positive feedback systems are those in which the productivity of resources actually increases as a result of human interference and exploitation. Flannery gives the example of the maize plant: When people transplanted maize from areas within its natural habitat to other areas, over time the plants responded to the process of domestication by a series of changes such as an increase in the size of the cob and in the number of grains. Genetic changes resulting from the process of cross-fertilization increased the productivity of this resource, and once people recognized this increased productivity, they turned more and more towards the domestication of maize. This hypothesis explains why people found agriculture more advantageous than food gathering, but it does not explain why the initial experiments in domestication were made in the first place. Recent studies have suggested that the key may in fact lie in environmental change, although not the sort suggested many years ago by Childe. The extinction of big game, which took place in Europe, was not really a factor in zones of early agriculture such as West Asia. Here, gazelles, wild cattle, onagers (wild ass), deer, and wild goats remained the main sources of meat during much of the Pleistocene as well as in the early Holocene. On the other hand, what does seem to be relevant is the fact that in many parts of the world, the Holocene was marked by the onset of a milder, warmer, wetter climate. Such changes may have led to an expansion of the natural habitat area of wild cereals

that had the potential for domestication. Perhaps it was not an environmental crisis but amelioration that was responsible for the beginnings of domestication. Given the limitations of the evidence and the fact that we are looking at very slow, gradual processes that must have varied in pace and detail, we may never be able to fully comprehend the details of the processes of animal and plant domestication or identify the impulses that lay behind them. It should also be remembered that in the case of complex cultural processes, the archaeological evidence often provides little ‘hard data’ on social and political factors that may have had an important role to play. More important than isolating a single factor responsible for the origins of domestication is to try to track down the process as it unfolded in different regions. Given the variety in ecology and resources in the various centres of early plant and animal domestication, it is very possible that different factors may have been involved in different parts of the world.

F IGURE 3.1 THE EVOLUTION OF MAIZE FROM THE WILD GRASS TEOSINTE; THE NUMBER OF GRAINS IN THE COB GRADUALLY INCREASED AND THE HUSKS EVENTUALLY BECAME ENCLOSURES FOR CORN EARS

The Identification of Domestication and Food Production in the Archaeological Record

When wild animals or plants are domesticated over long periods of time, certain morphological changes (i.e., changes in their form) tend to take place. In the case of animals, early domesticates tend to be smaller than their wild counterparts (later, when conditions of feeding and breeding reach an optimal level, their size tends to increase). The face becomes shorter in relation to the cranium. There are changes in dental structure—teeth become smaller, some teeth (such as the premolars and third molars) may disappear. Horns tend to reduce in size. Domesticated cattle have weak muscle ridges and poorly defined joint facets, while in the case of draught animals there is a strengthening of certain muscles. Domestication also leads to a shortening of the animal’s hair and changes in its coloration. Morphological changes of the sort listed above appear only when domestication has been underway for a long time and will not be apparent in the early stages. For example, it has been estimated that it took thousands of years of domestication for such changes to become apparent in the case of the horse, while they were faster in the case of cattle, goats, and sheep. Nevertheless, once such changes manifest themselves, it is usually possible for scientists to study the animal bones and teeth found at an archaeological site and to identify not only the animal they represent, but also whether this animal was wild or domesticated. The task of identifying the bones of a domesticated variety of an animal is made easier if bones of wild or transitional forms are also present at the site. Apart from the direct scientific analysis of animal bones, there are other ways of inferring animal domestication. Animals found outside their natural habitat—for instance, mountain goats found in the plains—suggest domestication. Age and sex ratios reflected in the faunal assemblage can also provide important clues. In the wild, the male–female proportion among animals is 1:1. However, when they are bred, males and castrates are killed quite young and females are killed in old age. These patterns can be identified in the faunal record. Just as in the case of animals, wild and domesticated plant grains and seeds can also be differentiated. Under conditions of domestication, over a long period of time, plants undergo certain morphological changes. For example, the grains of wild wheat and barley are larger than those of

domesticated varieties. Wild varieties of wheat and barley have brittle ears and fragile spikes and their ears break apart immediately on reaching maturity. This is the natural way in which plants maximize their seed dispersal. In the case of domesticated wheat and barley, on the other hand, the ears break up only at the stage of threshing. Not all plants have an equally good chance of surviving or being recognized in the archaeological record. Root crops such as potatoes and yams lack hard parts and are therefore less likely to survive. Further, since they reproduce asexually, they do not necessarily undergo significant genetic changes during the process of domestication, and they may have so many different varieties that it is difficult to distinguish between wild and domesticated ones. Direct evidence of plant domestication can be obtained by a careful analysis of grains or seeds found at a site, especially those that get carbonized due to contact with fire. Even an analysis of impressions of grain or husk on lumps of clay or pottery can help identify domestication. Indirect evidence of animal or plant domestication can be inferred from art remains such as representations of people capturing or tending animals, harvesting grain, or processing food. However, none of these are conclusive. Animal capture could indicate hunting, tending animals could reflect a stage of animal keeping, and harvesting grain and food processing are perfectly compatible with a stage of food collection. Certain kinds of artefacts and tools such as grinding stones and sickles are sometimes taken as indicative of plant domestication, but their evidence is not conclusive. Grinding stones can be used to grind collected wild grain and sickles can be used to reap wild plants. Evidence from the natural sciences—the analysis of pollen grains, molluscs, remains of insects, etc.—can indicate changes in land use and indirectly, the presence or absence of agriculture. However, ascertaining the food-producing status of a community is more difficult and subjective. While some sites give clear evidence of the importance of animal and/or plant domestication in their subsistence base, in many more cases, there is insufficient evidence to make an assessment. In fact, in the Indian subcontinent, sites are often labelled ‘neolithic’ simply on the basis of the presence of ground and polished stone tools. PRIMARY SOURCES

The analysis of ancient plant remains

The study of ancient plant remains is known as palaeobotany or archaeobotany. Botanical remains from ancient sites often include macro-botanical remains such as seeds or grains. These can get preserved through desiccation, waterlogging, or charring. It is possible to collect seeds or grains manually in the course of an excavation. However, this can damage them and smaller pieces may be missed. A more efficient method is the use of the flotation technique. There are different kinds of flotation apparatuses, but the basic principle in all of them is the same. This involves slowly and steadily pouring dried carbonized plant material along with its soil matrix into a liquid medium such as water. The inorganic material will sink to the bottom and the carbonized seeds will float on the surface and can be retrieved. These are then collected and

analysed under microscopes to determine what types of plants they represent and whether these were wild or domesticated. Plant remains can also take the form of micro-botanical remains. Tiny particles of silica called phytoliths are found in certain specific parts of a plant (e.g., the root, stem, or flower). Their recovery from a site can help differentiate between wild and domesticated species. Analysis of plant parenchyma (soft tissue of roundish, thin-walled cells in a plant stem or in the pulp of fruits) can be used for a similar purpose. Palynology—the analysis of pollen and spores—is another important technique. Pollen are the tiny reproductive bodies of flowering plants. Their strong outer exine (shell) can survive in certain kinds of sediments for thousands of years. Scientists can study pollen grains under microscopes and identify the plants they belong to. Changes in pollen profiles in different archaeological layers may suggest climatic change, forest clearance, or agriculture. These days, several new techniques are available, but these are so far used mostly in the West. For instance, it is possible to directly date tiny pieces of squash seeds and maize cobs through the use of accelerator mass spectrometric (AMS) dating. DNA studies can identify the chromosome structure in different plant genotypes. This can help establish links between domesticated and wild species of plants and identify the area where wild progenitors of domesticated species were originally located.

F IGURE 3.2 A FLOTATION APPARATUS

The Transition to Food Production in the Indian Subcontinent

The neolithic age is generally associated with food production, pottery, and sedentary living. The reality is more complex. In the Indian subcontinent, the roots of some of the features associated with the neolithic can be traced to the mesolithic phase. In the last chapter, there were references to the evidence of pottery and animal domestication at certain mesolithic sites. On the other hand, as we shall see, there are some neolithic sites without pottery. The issue of sedentism (i.e., sedentary living) is also complex. As we have already seen, some mesolithic hunter-gatherer communities led a fairly sedentary life. And there were some communities practising animal and/or plant domestication who did not live for very long in the same place. Further, instead of thinking of sedentary and nomadic life as two alternatives, it is necessary to recognize different degrees of sedentism in the lifestyle of various communities. The beginnings of animal and plant domestication did not mean the end of the hunting-gathering way of life. Communities that practised animal rearing and agriculture usually continued to hunt and forage for food. Moreover, there were numerous communities who retained their hunting-gathering way of life and never switched over to domestication at all. This chapter, however, focuses on those that did make the transition. Given the great ecological diversity in different parts of the subcontinent —especially in climate, soil, and the availability of plant and animal species that could be

potentially domesticated—it is not surprising that the details of the various adaptations made by early pastoralists and agriculturists varied quite a bit. One reason why the title of this chapter highlights the beginnings of food production rather than the neolithic is because food production is the most important aspect of the neolithic phase. Secondly, the history of early food-producing settlements in the subcontinent consists of different regional profiles and trajectories. In certain regions (e.g., the northern fringes of the Vindhyas), the foodproducing neolithic culture emerged out of an earlier mesolithic phase. In other areas (such as the north-west), there is no mesolithic phase and the earliest settlements seem to belong to neolithic agriculturists and pastoralists. Another important point to note is that while there are some ‘pure neolithic’ sites, there are many more neolithic–chalcolithic cultures which have elements of the neolithic along with the use of metal (mainly copper). In still other parts of the subcontinent (such as Rajasthan), there is so far little evidence of a neolithic or even neolithic–chalcolithic stage, and the earliest sedentary communities appear in a full-fledged chalcolithic context. Since we are dealing with a vast expanse of time, and in order to convey the idea of the complex and variegated cultural mosaic, in this book, the discussion of food-producing agricultural–pastoral communities of the subcontinent has been divided into three overlapping phases: Phase I—c. 7000– 3000 BCE; Phase II—c. 3000–2000 BCE; and Phase III—c. 2000–1000 BCE onwards. The first two phases are discussed in this chapter, while the third will be discussed in Chapter 5. In the case of sites which have a long cultural sequence, only the earliest phases that fall within the first two chronological phases are discussed here; the later phases will be discussed in Chapter 5. The various geographical zones of early food-producing communities are discussed in terms of their chronology, general features, and specific traits, against the background of the cultural sequence of that particular area (for site details, see Chakrabarti, 1999: 117–40; Allchin and Allchin, 1999: 97– 127). THE EARLIEST VILLAGE SETTLEMENTS IN THE INDIAN SUBCONTINENT, C. 7000–3000 BCE THE NORTH-WEST

Several sites in Baluchistan illustrate the change from a semi-nomadic pastoral life towards settled agriculture. The oldest and best documented evidence comes from Mehrgarh (Jarrige et al., n.d.). This site is located in the Bolan valley in the northern part of the Kachi plain, near the point where the river emerges from the hills through the Bolan pass. The Bolan valley was an important link between the Indus plains and the mountainous valleys of north Baluchistan, and people and animals must have moved along this route from very early times. Excavations at Mehrgarh revealed the remains of ancient settlements scattered over an area of about 200 ha on a low mound and the surrounding plain. Seven occupational levels were identified, giving striking evidence of continuous occupation and of cultural continuity and change over many millennia. The first six levels, i.e., Periods, are relevant for us here. Periods I and II at Mehrgarh are considered neolithic, even though there is a small amount of copper present. The remains of Period I (sub-divided into Periods IA and IB) were located in an 11 m thick deposit at the northern end of the site, on the high bank of the Bolan river. The chronology of this phase is somewhat uncertain due to inconsistent radiocarbon dates. The majority of the dates fall between 6000 and 5500 BP (c. 5000 BCE, calibrated). The problem is that although Period I seems

to have lasted for a very long time, most of the radiocarbon dates for the middle levels of Period IA also fall within the range of 5800 and 5530 BP. Furthermore, the excavators point out that there are also some much earlier radiocarbon dates—9385 ± 120 BP for Period IA; 7115 ± 120 BP for Period IIB; and 6500 ± 80 BP for Period III. This series of earlier dates has the advantage of providing a coherent chronological framework for the Mehrgarh neolithic sequence from the 8th to 6th millennia BCE. The people of Period I (this includes both Periods IA and IB) lived in houses made of handmade mud-bricks with small, rectangular rooms. One of the rooms at the lowest levels of Period I, measuring 2 × 1.8 m, had reed impressions on the floor and a grinding stone. The bricks used for house walls were of a standardized size, with distinctive rounded ends and finger impressions on their upper surface. Some of the structures divided into small units may have been granaries. The stone tools of Period I included thousands of microliths, most of them based on blades. A few ground neolithic handaxes (celts) were also found. Some of the blades were set into wooden handles with a thick layer of bitumen and may have been used as sickles to harvest grain. Grinding stones indicate food processing. There were a few stone vessels and objects such as perforated discs and spatulae incised with a criss-cross design. Bone tools, including needles and awls, were also found, as was a handmade clay female figurine. Mehrgarh I was basically a-ceramic, i.e., it had no pottery; the first few pieces of pottery appeared in Period IB. The people of Period I buried their dead in the open spaces between their houses. The bodies were placed in oval pits in a flexed (bent) position. The bones were often covered with red ochre, suggesting some sort of fertility beliefs. In at least two burials, young goats had been placed near the feet of the body. Grave goods included bitumen-lined baskets and food offerings, and ornaments such as necklaces made of stone or shell beads, bone pendants, and anklets. A copper bead was found in one of the burials. The occurrence of turquoise and lapis lazuli beads is especially interesting. The lapis lazuli could have come from the Chagai hills in north Baluchistan or from Afghanistan. Turquoise could have come from eastern Iran or central Asia. The nearest source of marine shells is the Makran coast, about 500 km away. The presence of such items in the graves indicates that the people of Mehrgarh were engaged in some amount of long-distance exchange. In Period IB, a graveyard consisting of 150 burials covering over 220 sq m was unearthed. The burials were more elaborate than before. A small niche was cut into one side of a pit, and the body and grave goods were placed inside. The niche was then sealed with a wall made of mud-brick, after which the pit was filled up. A few copper beads were found in the burials. There are some instances of double burials and also of secondary burials, where the bones of one or more people were collected and buried after exposing the body to the elements. The significance of these changes in burial practices is unclear. Period II at Mehrgarh, dated c. 6000–4500 BCE, is divided into three sub-phases—A, B, and C. The size of the settlement increased during this period and there were several mud-brick structures divided into small cell-like compartments. Some of these may have been houses, but others may have been used for storage. For instance, double rows of small rooms with a passage in between, with barley seeds on the floors, may have been used to store grain. The stone and bone tool types of Period I continued. There were two sickles made of microliths hafted onto a bitumen matrix. P. Vaughan’s microwear study of stone tools found in an area of Period IIA indicates that most of them were connected with the working of animal products—activities such as butchery, cooking, hide

processing, and the making of bone artefacts. Small amounts of handmade pottery occurred in the early part of Period II and wheel-made pottery appeared in Period IIC. In Period IIB, a copper ring and bead and a small ingot of copper were found. Other finds of Period II included an ivory tusk, pieces of red ochre, grinding stones, and a small unbaked clay figurine of a male torso. There were two flexed burials, the bodies covered with red ochre, unaccompanied by any grave goods. Mehrgarh III belongs to the second half of the 5th millennium BCE and is chalcolithic. There is evidence of a significant increase in craft activities, including large-scale production of wheel-made pottery with painted decorations, marked by innovations and refinement in pottery-making techniques. A pottery-manufacturing area was found, where the bases of three ovens were exposed on top of an accumulation of 6 m of pottery debris. The frequent occurrence of ornaments such as necklaces and bracelets made out of tiny steatite beads indicate that bead making was another important craft. There were also beads of semi-precious stones such as lapis lazuli, turquoise, and agate, as well as of terracotta and shell. Stone micro-drills may have been used to make engravings on shell. There were a few terracotta humped bulls. Terracotta crucibles with traces of copper suggest the beginning of metallurgy. Period III had storage complexes divided into compartments, similar to those of earlier phases. A large cemetery containing the burials of about 99 people shows changes in burial practices. The niches walled in by cigar-shaped bricks, known in Period II, were absent. The heads of some of the skeletons were placed on bricks. There was one collective burial with two wheel-made painted pots as grave goods (pots are not found in any other burial). In another burial, a copper or bronze object that looks like a fragment of a segmented seal was found near the skull. Ornaments, mostly made of steatite micro-beads, occurred frequently among the grave goods. There were also pendants of lapis lazuli, carnelian, turquoise, chrysoprase, agate, terracotta, and seashell. The most remarkable aspect of Periods I–III is that they provide the earliest and most comprehensive evidence of subsistence activities in the region, revealing the transition from hunting and food gathering to a heavy reliance on animal domestication and agriculture. Thousands of plant specimens were collected in the course of the Mehrgarh excavations. These included charred grains and seeds as well as impressions of grain on mud-brick. Barley seems to have been the most important crop. In Period I, the predominant type of barley was six-row naked barley (Hordeum vulgare nudum). There were also other varieties—hulled six-row barley (Hordeum vulgare vulgare) and wild and domesticated hulled two-row barley (Hordeum vulgare spontaneum and Hordeum vulgare distichum). The fact that wild, transitional, and domesticated varieties of barley were found at the site proves that north Baluchistan fell within the natural habitat zone of wild barley and that Mehrgarh was part of a large nuclear area of barley domestication.

F IGURE 3.3 B URIAL WITH STONE BLADES, CORES, AND A CELT AS GRAVE GOODS, MEHRGARH, P ERIOD I

Wheat was another important crop. Grains of domesticated hulled einkorn wheat (Triticum monococcum), emmer wheat (Triticum diococcum), and naked wheat (Triticum durum) were found in Period I. In later periods, a large proportion of the wheat grains belonged to the Triticum sphaerococcum variety. Whether Mehrgarh fell within the natural habitat zone of wild wheat is a matter of debate, as no clear evidence of wild wheat has so far been found in the area. But there is no doubt that the people of Mehrgarh were domesticating this cereal. Seeds of ber (Zizyphus jujube) and dates (Phoenix dactylifera) were also found in Periods I and II. In Period II, in addition to barley and wheat, there were numerous seeds of cotton (Gossypium sp.) found in a hearth. Period III showed continuity with the earlier period, but also a diversification of agriculture. Two new varieties of wheat (Triticum aestivum compactum, Triticum aestivum sphaerococcum) and one of barley (Hordeum hexastichum) and a new cereal—oats (Avena sp.)— were identified. Wheat had become more important than barley. Not much is known about the methods of cultivation practised by the neolithic and early chalcolithic people of Mehrgarh. Farmers must have relied on winter rains and may have channelized water into their fields by building mud or stone embankments similar to the gabarbands made in the region today. Stone sickles made by hafting tiny microliths onto wooden handles with bitumen must have been used for harvesting grain. Neolithic Mehrgarh gives clear evidence of the transition from hunting to animal domestication. The lower levels of Period I were dominated by the bones of wild animals—deer (mostly gazelle, but also some blackbuck, sambar, and chital), nilgai, goat, onager (wild ass), water buffalo, cattle, pig, and perhaps elephant. There is also evidence of domesticated goats, and the decreasing size of sheep and cattle suggests that their domestication too was underway. By the end of Period I, the frequency of bones of gazelles and other wild animals had drastically decreased, while those of domesticated cattle, goats, and sheep had greatly increased. Cattle were now the most important domesticated animal. In Period III, cattle still dominated, but there was an increase in the proportion of sheep and goat bones. Interestingly, Period III also showed an increase in the number of bones of wild animals, suggesting resurgence in hunting activity.

J. R. Lukacs’ study (1985) of the human dental remains shows a low rate of dental caries (cavities) in the early levels. This may have been due to the high fluoride levels in the drinking water available in the area. Other features of the teeth suggest that people had a coarse diet. There is evidence of tooth probing (people poking their teeth either to sooth pain or prise out food). Dental health declined in Period III, and this may have been due to changes in food habits, for instance, the consumption of more refined foods. The evidence from Period IV onwards shows a further expansion of the settlement, diversification of agriculture and crafts, and more and better decorated pottery. In Period IV, there were larger structures, with rooms separated from each other by wide walls and doors with wooden lintels. One door, only 1.10 m high (people must have had to bend down to go through) led into a room crammed with many objects such as stone flakes, blades, grinding stones, pestles, and many bones. Other items found in this room included a storage jar, a crushed basin with ridges and snake designs painted on the inner side, fine goblets, and beautifully painted vessels. The pottery of Period IV included polychrome wares. A new style of terracotta female figurines with a tubular body, pinched nose, and joined legs made its appearance. There are continuities in pottery designs between Periods IV and V. In Period VI, there were some changes—the appearance of a red ware decorated with pipal leaves, and a well-fired grey ware. This is also the time when similar styles of pottery began appearing in various parts of Baluchistan, suggesting an increase in interaction. A large pottery kiln was found in Period VI. A distinctive feature of this period are terracotta female figurines with elaborate hairstyles, heavy breasts, and joined legs, which may have had a cultic significance. Several large mounds in the Kachi plain may represent unexplored sites contemporary to the later periods of Mehrgarh. The Bolan pass leads from Mehrgarh into the Quetta valley, where there are a number of sites. Today, farmers of this valley compensate for meagre rainfall by using water drawn from wells and streams to irrigate their fields. Kile (also spelt Kili) Gul Mohammad and Damb Sadaat are two of the important excavated sites in this area. Kile Gul Mohammad is about 3.2 km from Quetta city, on the banks of the Hannah river. The mound is a small one—about 90 × 55 m. Walter A. Fairservis (1950) conducted a small excavation over a 3.5 sq m area up to a depth of 11.14 m, reaching natural soil. The excavation revealed four periods of occupation—KGM I, KGM II, KGM III, and KGM IV. Radiocarbon dates from the upper levels of neolithic KGM I fall between c. 5000 and 4500 BCE, but the beginning of the settlement could go back to c. 5500 BCE, or even earlier. There was no evidence of pottery at this stage. Bones of domesticated cattle, sheep, goat, and ass/horse were found. There were no cereals, but two sickle blades were discovered. The people of Kile Gul Mohammad may initially have been nomadic pastoralists, but by the end of Period I, they were living in houses made of mud or wattle-and-daub (interlaced rods and twigs plastered with mud). The artefacts included microliths and blades of chert, jasper, and chalcedony. There were a few ground tools and bone points. Handmade and basket-marked pottery made its appearance in KGM II. In KGM III, there was wheel-made pottery, including a fine black-on-red ware with geometric designs painted on it. Remains of mud-brick houses, some resting on stone foundations, have been found. The first copper objects made their appearance in Period III. The uppermost level of Kile Gul Mohammad (KGM IV) was contemporary with the first period of occupation at Damb Sadaat (DS I), and there was a broad similarity in their cultural remains. KGM IV and DS I showed a distinctive type of pottery known as Kechi Beg Ware after the site where it

was first found. This was a well-fired, thin, buff-coloured pottery. The shapes included deep and wide vases, bowls, and jars. The pots were painted with geometric designs in black, sometimes also in red. Calibrated dates for Period II of Damb Sadaat indicate c. 3000 BCE as its midpoint. In this phase, there were multi-roomed mud-brick structures, many with limestone blocks used in the foundations. Hearths for cooking, similar to modern tandoors, were found in houses. The pottery included a type known as Quetta ware—a buff-coloured ware decorated with black painted designs, with shapes such as jars with flaring or straight rims, small-mouthed bowls with a sharp angle between the shoulder and base, and jars with pedestals. There was also a grey pottery known as Faiz Mohammad Grey Ware. This was represented by shallow plates and deep, open bowls, painted with geometric and naturalistic designs. Terracotta objects included cattle figurines, some painted with black stripes, and female figurines similar to those found in Mehrgarh VI. There were also small terracotta models of houses, rattles, and seals. A copper/bronze blade of a dagger or knife, a bone spatula, and an alabaster vessel are other artefacts associated with Period II at Damb Sadaat. Anjira and Siah Damb in the Kalat plateau were excavated by Beatrice de Cardi, the former in 1948 and 1957, and the latter in 1957. Five periods of occupation were identified; the earliest occupation was apparently contemporaneous with Period II at Kile Gul Mohammad. In the Kalat plateau, Period I represented a semi-nomadic settlement, with no traces of structures. The pottery included a fine wheel-made buff ware, sometimes with a burnished red slip (coating). Chert blades were also found. In Period II, there were mud structures made on stone boulder foundations. The pottery included a red-slipped ware and a burnished grey ware. In Period III, the foundations of houses were made of blocks of stone cut into rough squares. The earlier pottery made way for Togau ware, a red pottery with black painted designs. The main shapes were open bowls, and designs of stylized ibexes, birds, and goats were painted on the interior, just under the rim. There was also another kind of pottery (known as Zari ware) with paintings in white with black outlines. In Period IV, the stone used for making houses was properly dressed into square blocks and there was pottery similar to that found at Nal. Period V of the Kalat sites has been co-related with Damb Sadaat III. Mundigak is located on a now dry tributary of the Arghandab river in south-east Afghanistan. Excavations at this site were carried out by J. M. Casal in the 1950s and 1960s. The dates for Period I (which is divided into several sub-phases) fall within c. 4000–3500 BCE. The early settlers seem to have been semi-nomadic, as no structures were found in the lowest levels of Period I. In phase 4 of Period I, there were small oblong cells with walls made of pressed earth. In phase 5, there were larger houses consisting of square or oblong rooms made of sun-dried bricks. Cooking hearths were initially situated outside the houses and later perhaps in the courtyards. There were wells in between the houses. Pottery was found throughout Period I and was mostly wheel-made. There were bone awls, alabaster vases, stone blades, and beads made of stone, lapis lazuli, and frit (a calcined mixture of sand and fluxes). The few copper objects included a needle and a small bent blade. A terracotta figurine of a humped bull was found in phase 3 of Period I. Period II at Mundigak gave evidence of plant remains—club wheat (Triticum compactum) and ber, and there were bones of domesticated cattle, sheep, and goats. Explorations in the Zhob–Loralai area of Baluchistan have identified many early village sites in the plains of the Gomal, Zhob, Anambar, and Thal rivers. Sur Jangal, Dabar Kot, and Rana Ghundai are three important sites in the Anambar valley. The people living at these sites must have been

practising some form of irrigation, otherwise it is difficult to understand how they sustained themselves. The early phase of occupation at Sur Jangal seems to be contemporaneous to Kile Gul Mohammad IV. People lived in small mud houses. The large quantities of cattle bones indicate the importance of cattle rearing. Some of the pottery found at Sur Jangal was decorated with painted designs of humped and humpless cattle. Terracotta items included small house models. There were also goggle-eyed female figurines, similar to those found at other contemporary and slightly later sites in the Zhob valley (such as Periano Ghundai and also at Mehgarh VI and Damb Sadaat III). These figurines have been given the label ‘Zhob mother goddess’, and are assumed by some scholars to have had some sort of cultic significance. Rana Ghundai in the Loralai valley was excavated by Brigadier Ross in the 1930s and reinvestigated by Fairservis in 1950. Five occupational levels were identified. The calibrated dates for Period I gave a range of c. 4500–4300 BCE, while those for the early levels of Period III gave a range of c. 3500–3100 BCE. Period I consisted of a 4 m thick deposit and seems to represent a settlement of a semi-nomadic community. Traces of living surfaces and hearths were found, but there were no well-defined structural remains. Almost all the pottery was handmade and plain. There were bones of domesticated cattle, sheep, and goat. Four teeth, either of a horse or ass, were found. Brigadier Ross, a veterinary officer, was certain that they were horse teeth, but this has been contested by others. Microlithic blades, bone points, and needles with eyes were other artefacts found in Period I. In Period II, the typical pottery was wheel made, with a buff to red surface. Decorations included friezes of stylized humped bulls, and in one instance, blackbuck, all painted in black. The main pottery forms were bowls or cups with a wide shoulder, often with a ring base or hollow pedestal. In Period III, there were some changes in the style of painted pottery. In the valley of the Gomal river (a tributary of the Indus), there are several early sites in Dera Ismail Khan district. Of these, Gumla and Rahman Dheri have been excavated. Gumla was excavated by a team from Peshawar University in 1971. Six cultural phases (i.e., periods) were identified, the first two of which are of interest to us here. Period I revealed a small settlement, just a little over 0.40 ha in size. There were microliths, bones of domesticated cattle, hearths, and large community ovens. Period I was a-ceramic; pottery made its appearance in Period II. Pots with a rough surface were followed by finer pottery painted with geometric designs, cattle, and fish. Terracotta female figurines were also found. There were microliths and a few objects of copper and bronze. Terracotta objects included bangles, cart models, gamesmen, and cattle and female figurines. There are similarities between some of the pottery designs and the female figurines of Gumla and certain sites in Turkmenistan in central Asia.

MAP 3.2 EARLY VILLAGE SETTLEMENTS IN THE NORTH-WEST

There are several sites to the north of Gumla and Rahman Dheri. One of them is Sheri Khan Tarakai, in the Bannu basin, where calibrated radiocarbon dates gave a range of c. 4500–3000 BCE for the earliest levels. Many of the houses here were made of mud-bricks built over stone foundations. Artefacts included ground celts, microliths, saddle querns and mullers, ring stones, and bone tools. Terracotta spindle-whorls and female and bull figurines (some painted) were found. There was evidence of the cultivation of barley. Bones of sheep, goats, cattle, and buffalo were found, as were freshwater molluscs and chank shells from the coast. There were two main types of pottery. One was a coarse handmade pottery with a black slip on the outside and a burnished pinkish buff to cream-slipped interior, with designs (including representations of goats) painted on in black or brown. The body of the other type of pottery had a rusticated surface (i.e., roughened with a thick slurry of clay); sometimes the neck-and-shoulder portion were left smooth and unroughened and were decorated with painting. In the northern part of the Punjab province of Pakistan, the site of Sarai Khola, lying on the edge of the Potwar plateau, revealed a neolithic occupation going back to about the 4th millennium BCE. The site was excavated in 1968–71 by the Pakistan Archaeological Department. Here, in Period I, there

was a handmade plain red or brown burnished pottery with mat impressions on the base. There were ground and polished stone celts, blades, microliths, and lots of bone points. Terracotta wheels and toy carts were also found. The 5 ha site of Nal, located in the Khozdar area which links north and south Baluchistan, was first excavated in 1925. Some of the structures discovered here were made of boulders from a nearby riverbed, while others were made of stone quarried from the nearby hills. Several burials were found, most of them fractional burials in pots, but there were also some complete skeletons laid out in clearly defined and sometimes undefined graves. There was one instance of a child buried in a grave consisting of a small mud-brick chamber with grave goods including a bead necklace and crystal pendant. The typical Nal pottery was polychrome and includes a variety of shapes, many with disc bases— ovoid, narrow-mouthed pots; carinated pots with a narrow mouth; almost straight-walled jars; open bowls; carinated bowls with an inward-turning upper body; and canisters with a flat bottom and a round, straight-edged mouth. Geometric and naturalistic designs (including fish and ibex) were painted onto the pots in blue, red, and/or yellow. The many artefacts found at Nal included stone balls, discs, ring stones and grinding stones; silver foil; beads made of agate, crystal, carnelian, lapis lazuli, and paste; and cattle figurines. Several copper objects and an adze made of copper alloyed with nickel and lead were also found. There are no radiocarbon dates from the site, but Nal pottery is considered contemporaneous with that of Periods I and II at Damb Sadaat and Period IV of Anjira and Siah Damb. Nal-related sites are associated with two types of water management systems. One was the building of stone embankments across hill slopes to block soil washed down by rains; crops were grown on such terraces after the rains were over. The second was a system wherein water that accumulated in low-lying basins was channelized into fields through a system of small dams and canals. Kulli in the Kolwa tract is a 12 ha site, only the upper levels of which have been excavated. Here, there were multi-roomed stone structures. The artefacts included stone querns and rubbing stones, beads made of semi-precious stones such as lapis lazuli, agate, and carnelian, bone bangles, and a small quantity of copper, gold, and glass. The Kulli pottery is profusely ornamented; a typical motif is cattle with an elongated body and large round eyes, usually set in a landscape. Analogous remains have been found at the sites of Mehi, Niai Buthi, Adam Buthi, Nindowari, and Edith Shahr. Adam Buthi, dated 3500–3000 BCE, is the earliest of these sites. Bala Kot is a 2.8 ha site on the Makran coast of south Baluchistan, at the mouth of the Windar river. Period I represented a neolithic occupation dated from the late 5th to early 3rd millennium BCE. The houses were made of mud-bricks. Some of the wheel-made pottery was similar to that found at Nal. Microliths, terracotta figurines of humped bulls, beads (of stone, lapis lazuli, shell, and paste), terracotta, shell, and bone artefacts, and a small number of copper objects were found. There is evidence of the cultivation of barley and the domestication of cattle, sheep, and goats. The bones of buffalo, deer, pig, and hare were found. Apart from Bala Kot, there are other early village sites in the Makran area, such as Miri Qalat and Shahi Tump. In the Cholistan desert of Bahawalpur, a number of early village settlements are located on the alluvial plain of the Ghaggar-Hakra river. This river flowed to the east of the Indus, and although it is now dried up, it must once have been a mighty stream. The typical handmade and wheel-made

pottery found at the earliest settlements in this area included large and small vessels with a coating of mud mixed with pieces of pottery applied to the outer surface; thick and thin pottery with multiple incised lines; and carinated or globular vases with a black slip on the exterior. These pots are known as Hakra wares, and the sites where they are found are known as Hakra wares sites. M. R. Mughal’s (1997) research in this area has revealed that the Hakra settlements go back to the middle of the 4th millennium BCE, if not earlier. As many as 99 Hakra wares sites have so far been identified. They range from small settlements below 5 ha to fairly large ones of 20–30 ha. About 52 per cent of the sites seem to be camp sites, while 45 per cent appear to represent more permanent settlements. Some seem to have been centres of craft specialization. Artefacts found at Hakra wares sites include microliths, grinding stones, terracotta cattle figurines, bangles made of shell and terracotta, and pieces of copper. Bits of copper were found at Valwali, a site where 32 terracotta figurines, including those of the humped bull, were found. Hakra wares have also been found outside the Ghaggar-Hakra valley, for instance at Jalilpur in the Punjab plains of Pakistan, near the left bank of the Ravi. Period I at this site gave evidence of Hakra wares in association with artefacts such as beads of stone, gold, coral and semi-precious stones, chert blades, and bone points. Terracotta net sinkers (used to weigh one end of the fishing net to keep it under water) indicate that fishing was an important part of the subsistence base of the people. Bone remains of sheep, goat, cattle, and gazelle were also found.

F IGURE 3.4 NAL POTTERY (AFTER HARGREAVES, 1929)

Harappa, on the banks of the Ravi, has given evidence of an early period designated the Ravi aspect of the Hakra phase, dated c. 3500/3300–2800 BCE (Meadow and Kenoyer, 2001). Remains of a small village with huts made of wooden posts and walls of plastered reeds were identified. Some mud-brick fragments of what may have been a kiln were found, but there was no evidence of mudbrick structures. Other artefacts included pottery, stone and bone tools, broken necklaces, terracotta spindle whorls, steatite beads, and bangles made of shell and terracotta. The most important evidence were potsherds with pre-firing marks and post-firing graffiti representing the formative stage of the Harappan script.

F IGURE 3.5 KULLI POTTERY FROM NINDOWARI (AFTER CASAL, 1966)

Period IA at Kunal in Hissar district of Haryana has also yielded Hakra wares. In the early level, the settlement was small (about 1 ha). Pottery designs included pipal leaves and a bull with very curved horns. Artefacts included bone tools, micro-blades made of chalcedony, copper fishhooks and arrowheads. People built their houses on an artificially raised area. House floors were made by digging a pit and paving it with rammed earth. The floors were below ground level and walls were plastered with mud. Post-holes around the circumference show the places where wooden posts supported a wattle-and-daub superstructure. No radiocarbon dates are so far available from Kunal. Bhirrana is a recently excavated site in the Fatehabad district of Haryana (Rao et al., 2004–05). Period IA belongs to the Hakra wares culture. People lived in shallow mud-plastered pit dwellings varying from 34 to 58 cm in depth and 230 to 340 cm in diameter. Apart from dwelling pits, pits used for sacrifices or industrial activity and refuse pits were also identified. In addition to the typical Hakra wares, there were other types of pottery such as mud applique ware, incised ware, tan slipped/chocolate slipped ware, black burnished ware, brown on buff ware, bi-chrome wares, black-and-red ware, and red wares. The artefacts included beads of carnelian, agate, jasper, and lapis lazuli; plain and painted terracotta bangles; sling balls of sandstone and terracotta; an unbaked triangular cake; a sandstone quern and pestle; a crucible; a clay hopscotch; and a chert blade and bone point. THE VINDHYAN FRINGES AND OTHER AREAS

The reason why so much detail has been given about early agricultural villages in the north-west is because there is much more data about this zone compared to other areas. However, another early centre of agricultural–pastoral communities lay in the Vindhyan fringes in southern Uttar Pradesh, where over 40 neolithic sites have been identified in the course of explorations in the Belan, Adwa, Son, Rihand, Ganga, Lapari, and Paisuni rivers. Neolithic levels have been identified at several excavated sites such as Koldihwa, Mahagara, Pachoh, and Indari. The key issues are those of dates

and whether the rice remains that have been found at several sites belong to wild or domesticated varieties. The neolithic culture in this area emerged out of a well-established mesolithic phase. Some of the mesolithic features such as microlith blades and the range of heavier stone tools continued, but there are also important new features such as the domestication of cattle and the cultivation of rice. Reference was made in the previous chapter to the discovery of wild rice at mesolithic levels at Chopani Mando in the Belan valley. Recently, domesticated rice has been reported from mesolithic levels at Damdama as well. The fact that wild rice is found in the area even today shows that it fell within the natural habitat zone of this cereal, and this explains the early dates for the domestication of rice.

MAP 3.3 EARLY CENTRES OF AGRICULTURE IN THE SUBCONTINENT

Koldihwa and Mahagara (both in Allahabad district, UP) are two important excavated sites, located on the northern fringes of the Vindhyas on the banks of the Belan river. Koldihwa showed cultural continuity from the neolithic to the iron age. Remains of rice and impressions of rice husk embedded in pieces of burnt clay were found here at neolithic levels. The examination of rice imprints on pottery suggests that the people were familiar both with wild rice and cultivated rice (Oryza sativa). Other discoveries included stone blades, polished stone celts, microliths (mostly

made on chert), querns and mullers (used for grinding), and bone tools. The pottery was handmade and consisted of three varieties—net-marked or cord-marked pottery; a plain red pottery; and a black-and-red ware. Deep bowls and storage jars were the dominant shapes. Some of the red ware showed soot marks, suggesting that these pots may have been used for cooking. There is currently a debate about the dates of the neolithic phase at Koldihwa. Three of the calibrated C-14 dates from the site are early and fall between the 8th and 6th millennium BCE (7505–7033, 6190–5764, 5432– 5051), but the other dates are much later. Mahagara on the right bank of the Belan river (not far from the mesolithic site of Chopani Mando) is another important neolithic site. Floors and post-holes associated with 20 huts were identified here. Reed or bamboo impressions on clumps of mud suggest that hut walls were made of wattle and daub. There were neolithic stone blades, microliths, celts, querns, mullers, and sling balls on floors. Pottery, bone arrowheads, terracotta beads, and animal bones were also found at the site. An interesting discovery was a cattle pen (about 12.5 × 7.5 m) located in the middle of the settlement. This was irregular in plan, with a fence marked by 20 post-holes and spaces left for at least three openings. Inside the fenced area were clusters of hoof marks left by cattle of different ages. The number of such marks suggests that about 40–60 animals may have been penned here. Rows of hoof marks of sheep or goats were also found outside the pen, near the huts, suggesting the frequent movement of animals between the huts and the enclosure. Animal bones included those of cattle, sheep, goat, horse, deer, and wild boar, out of which the first three seem to have been domesticated. The botanical remains included rice husk embedded in pottery. The bone and plant remains suggest that people hunted wild animals, collected wild plant food, and domesticated plants and animals. The site of Kunjhun is in the Son valley in Sidhi district of Madhya Pradesh, not far from Koldihwa. The neolithic settlement here, which goes back to the 4th millennium BCE, yielded wild and domesticated rice. Kunjhun seems to have been a factory site specializing in the making of stone artefacts. Archaeologists identified several areas where stone was heated to improve its colour and workability and then made into blades. Taken together, the evidence from Koldihwa and other sites in its vicinity suggests that the northern fringes of the Vindhyas constituted an early, independent centre of rice domestication. Early agricultural settlements also spread into the central Ganga plain. This is indicated by recent excavations (Tiwari et al., 2001–02) at Lahuradeva in Sant Kabir Nagar district in eastern Uttar Pradesh. The 220 × 140 m mound here stands about 4 m above the surrounding plain, surrounded by a lake on three sides. The site revealed a five-fold cultural sequence from the neolithic to the early centuries CE. Neolithic Period I was sub-divided into Periods IA and IB. In Period IA, there was a cord-impressed red ware and a black-and-red ware. The pots were mostly handmade, with a few wheel-made specimens. Small burnt clay pieces showed that people lived in wattle-and-daub houses. The plant remains included rice and a few wild grasses. Husk marks of rice were found embedded in the core of several potsherds. The rice appears to be a domesticated variety. The calibrated dates for Period IA at Lahuradeva fall within the late 6th and early 5th millennia BCE. Although the evidence is at the moment neither absolutely clear nor substantial, there is a possibility that there were other zones in the Indian subcontinent which saw an early transition from hunting-gathering to agriculture and pastoralism. In Ladakh, the neolithic site of Giak has given a radiocarbon date belonging to the 6th millennium BCE. However, nearby Kiari does not go back beyond c. 1000 BCE.

Pollen studies of the salt lakes of Didwana, Lunkaransar, and Sambhar in Rajasthan indicate a marked increase in cereal-type pollen in this area in c. 7000 BCE. This, along with the discovery of tiny charcoal pieces, may indirectly suggest the clearance of forests and the beginning of agriculture. However, no food-producing sites of such an early date have so far been identified in the area. Cereal pollen in c. 8000 BCE contexts has also been found in the Nilgiri hills in South India. In the Horton plains of central Sri Lanka, pollen analysis suggests incipient cereal plant management along with slash-and-burn techniques of cultivation in c. 17500 BP, and the cultivation of oats and barley in c. 13000 BP. NEOLITHIC, NEOLITHIC–CHALCOLITHIC, AND CHALCOLITHIC COMMUNITIES, C. 3000–2000 BCE

During c. 3000–2000 BCE, village settlements spread to new areas. It can be noted that these settlements were roughly contemporaneous with the urban Harappan civilization, which is the subject of the next chapter. The volume of information for this period is more substantial than for the preceding millennia, and certain distinctive characteristics of the various geographical zones can now be identified. THE NORTH AND NORTH-WEST

In the Kashmir valley, there are several neolithic sites near Srinagar and between Baramulla and Anantnag. These include Burzahom, Gufkral, Hariparigom, Jayadeviudar, Olchibag, Pampur, Panzgom, Sombur, Thajiwor, Begagund, Waztal, Gurhoma Sangri, and Damodara. During the Pleistocene era, the Kashmir valley was a gigantic lake and the neolithic sites are located on the remnants of the ancient lake beds known as karewas. Burzahom, one of the important excavated sites in this region, is located on a terrace of karewa clay above the flood plain of the Jhelum river, 16 km north-east of Srinagar. The site offers a beautiful view of green fields and the Dal lake, which is only about 2 km away. Burzahom is a Kashmiri word meaning ‘place of birch’, and the discovery of burnt birch in the excavations indicates that birch trees grew in the area in neolithic times as well. The site must have been surrounded by forests, with water close by, and the neolithic people must have cut down some of the trees in order to establish their settlement. The site was discovered in 1935 by de Terra and Paterson, who thought it belonged to the Harappan civilization. Its real significance was understood much later, when excavations were carried out by the Archaeological Survey of India in 1960–71 under T. N. Khazanchi. There are four periods of occupation at Burzahom. The first two are neolithic, the third megalithic, and the fourth early historical. Period I was dated by the radiocarbon method to before c. 2920 BCE. A distinguishing feature of Period I at Burzahom is the presence of mud-plastered pit dwellings. Most of the pits were round or oval, narrower at the top and widening out towards the base. The largest is 3.96 m deep, with a diameter of 2.74 m at the top and 4.57 m at the bottom. Post-holes around the circumference of the pits at ground level show where wooden poles would have supported a roof made of pinewood thatched with birch. Some of the deeper pits had a few steps, but these did not extend to the bottom, perhaps because this would have narrowed the space too much. Ladders may also have been used to climb in and out of the deeper pits. Charcoal, ash, potsherds, and hearths made of stone or clay were found inside the pits. There were some square and rectangular pit chambers too, about 1 m deep. One of them measured 6.4 × 7 m. Some of the pit

chambers had stone or clay hearths. It is interesting to note that the square/rectangular pit chambers were found in the centre of the settlement, while the round/oval ones were at the periphery. Close to the living pits were smaller storage pits with a 60–91 cm diameter, containing stone and bone tools and animal bones. Stone hearths near the mouths of some of the dwelling pits suggest that people also lived in the open at ground level, probably during the warm summers.

NEOLITHIC STONE TOOLS, B URZAHOM

NEW DIRECTIONS IN RESEARCH

Did people actually live in the Burzahom pits?

Pits have been found at neolithic levels at Burzahom and Gufkral in Kashmir and at Loebanr III and Kalako-deray in the Swat valley. They have generally been interpreted as winter homes of neolithic people. The steps, ash, charcoal, and potsherds in them have been cited as proof of the fact that people lived here. Pit dwellings are seen as a strategy adopted by neolithic people to cope with the harsh Kashmir winter. It is presumed that people moved to the ground level in summer. Recently, this interpretation has been questioned by R. A. E. Conningham and T. L. Sutherland on the basis of a fresh analysis of pits found at British iron age sites. The British iron age pits were once considered dwellings, but some scholars have rejected this idea. Experiments carried out by P. J. Reynolds showed that as soon as a fire was lit inside such a pit, the atmosphere became intolerably thick with smoke. It is also argued that the lighting of fires inside the pits does not necessarily indicate domestic activities such as cooking or an attempt to warm the living space. The firing of pits could have been in order to prolong their life, to clear mould or damp, or to

speed up the drying of the mud plaster. Moreover, if the pits were dwelling units where fires were regularly lit, their sides should have been black with soot, but this was not the case. An alternative explanation of the British iron age pits is that they may have been underground grain storage units. Conningham and Sutherland suggest that it may be time to reconsider the function of the Kashmir– Swat pits as well. They argue that sites such as Burzahom may not have been occupied all year round, with people living in pits in the winters and moving to ground level in summer. They may have been occupied only during spring and summer, and abandoned during the winter. After the harvest, surplus grain could have been stored and sealed in the underground pits. When winter set in, people may have migrated to the less severely cold areas of the plains or the lower valleys, leaving the sealed grain to be used for sowing next spring. While the majority opinion among scholars currently interprets the Kashmir–Swat pits as dwellings, the hypothesis cited above shows how the same evidence can be interpreted in a different way. SOURCE Conningham and Sutherland, 1997

Other finds of Period I at Burzahom included ill-fired, handmade, coarse pottery in grey, red, brown, and buff colours. The shapes included simple rimless bowls and bottle-shapes with flared rims. Mat impressions on the base of many of the pots showed that they were made on mats. The stone tools included oval and oblong stone axes (some pecked and ground), chisels, adzes, grinding stones, ring stones, and mace heads. Also found were ‘harvesters’—distinctive rectangular stone choppers or knives with two or more holes on the blunt side. Burzahom had a well-developed bone tool industry; artefacts such as points, harpoons, needles (with and without eyes), awls (probably for stitching animal skins), spear heads, daggers, and scrapers were found here. Tools were also made from antlers. No burials were found in Period I, suggesting that people may have adopted some other method of disposal of the dead. In Period II, the people of Burzahom moved out of the pits and built houses on ground level. Some pits were filled up with karewa soil, their surface plastered with mud and covered with a thin layer of red ochre. These formed the floors of huts made of mud, mud-brick, and timber. Several burials were found in Period II, mostly within the habitation area. The dead were usually buried under house floors or in the compounds, in oval pits plastered with lime. Both inhumation and secondary burials were practised. In the case of secondary burials, the bones were sometimes covered with red ochre. In the primary burials, the body was placed in a flexed position. Apart from the occasional beads around the neck of some of the bodies, there were usually no grave goods. Holes in one of the skulls gave evidence of trepanning (boring holes in skulls). Period II at Burzahom continued till at least c.1700 BCE.

B URZAHOM: BONE TOOLS INCLUDING NEEDLE WITH EYE

B ONE ARROWHEAD

P ERFORATED HARVESTER

An interesting feature of Period II of neolithic Burzahom is that humans were sometimes buried along with wild animals such as deer, wolf, ibex, nilgai, snow leopard, and pig, and domesticated animals such as cattle, buffalo, dog, sheep, and goat. The animals may have been killed and buried along with the deceased humans or their meat may have been placed in the grave as part of the grave goods. The interment of dogs with humans suggests that pets were sometimes buried along with their masters. There were also separate pit burials for animals within the habitation area. In one case, five dogs were buried along with antlers.

F IGURE 3.6 B URZAHOM POTTERY

Artefacts from Period II included pottery, mostly handmade. There were a few new shapes and a black burnished pottery, which seems to have been a deluxe ware. The shapes included dish with hollow stand, globular pots, jars, stems with triangular perforations, and a funnel-shaped vase. A distinctive type in the black burnished ware is a high-necked jar with a flaring rim, globular body, and base, with oblique notches incised on the lower part of the neck. Stone and bone tools continued, similar to those of Period I, but they were more numerous and had a better finish. The stone tools included harvesters. A single copper arrowhead was found towards the end of Period II. Microwear analysis of Burzahom neolithic tools (Pant, 1979) has shown that the tools were often re-ground and re-shaped. Some of the handaxes had clearly been used for cutting, chopping, and dressing wood, while others were probably used for chopping meat. Pant’s study also showed that the ring stones functioned as mace heads. Two engraved stone slabs were found in Burzahom Period II. The engraving on one of these is indistinct. Its pattern has been tentatively identified as a hut with a thatched conical roof, to the right of which is the hind portion of some sort of animal, whose tail can be seen clearly. The other engraving is clearer (see p. 129). It covers an area of 48 × 27 cm of a stone slab and depicts a hunting scene. A stag with large antlers is being pierced from behind by a (female?) hunter with a long spear, while another hunter shoots an arrow at it from the front.

Hunting and fishing were important parts of the lives of the neolithic people of Burzahom. This is clear from the animal bones, the engraved hunting scene, and the high percentage of weapons such as spearheads, arrowheads, and harpoons. Initially, there was no direct evidence of agriculture from the site, and scholars interpreted harvesters, stone querns, flake knives, mace heads, and seeds of wild plants as indirect evidence of some level of cultivation. However, more recently, an analysis of botanical remains from different strata of Periods I and II has provided direct evidence of cultivated wheat, barley, and lentils (Lens culinaris). The distinctive features of the Kashmir neolithic include a wide range of stone and bone tools, pit dwellings, perforated ‘harvesters’, and animal burials. Some of these features also occur at sites in central Asia and China. A wheel-made red pot containing 950 beautiful agate and carnelian beads was found in the early levels of Period II. Another globular pot had the painting of what seems to be a horned deity, a motif which occurs at early Harappan levels at Kot Diji. This suggests some contact between the neolithic communities of Burzahom and the Indus area.

B URNISHED GLOBULAR JAR WITH LONG NECK, B URZAHOM

DECORATED STONE HARVESTER, G UFKRAL

The cultural sequence at Gufkral (41 km south-east of Srinagar, near Tral) extends from the neolithic to the historical period. Period I of the sequence is neolithic and is divided into three subphases: Period IA, IB, and IC. There is a calibrated date of 2468– 2139 BCE from Period IB, so Period IA could go back to c. 3000 BCE or even earlier. As at Burzahom, here too, in Period I, there were pit dwellings, circular or oval, wide at the base and narrower above, varying in diameter from 3.80 to 1.50 m at the top. The larger dwelling pits mostly belonged to the earlier phase and were only 20 to 30 cm deep. The dwelling pits were surrounded by storage pits and hearths. Post-holes around the pits and hearths indicated the places where wooden posts were erected to support a superstructure of grass and reed. The bases of houses may have been plastered with mud to prevent the entry of water and snow. In the earlier dwelling pits of Period IA, floors were plastered with red ochre paste. Some pits were subsequently enlarged, and there were also two-chambered dwelling pits. In the early part of Period IA, hearths were rectangular in shape, while in the later phase, circular and rectangular hearths of clay were found. Interestingly, no hearth or fireplace was found inside the dwelling pits. Period IA at Gufkral was a-ceramic. The finds included polished stone tools and a large quern with ochre paste sticking to the depression in the middle. There were tools made of bone and horn, including small arrowheads and a bone needle with an eye. In most cases, the tips of bone tools were charred to strengthen the working edge. Other artefacts include steatite beads and a broken terracotta marble. Bones of wild animals—sheep, goat, cattle, red deer, Himalayan ibex (a wild goat), wolf, and bear—were found. There were also some bones of domesticated sheep and goats. The people of Gufkral were clearly heavily dependent on hunting, but were beginning to domesticate certain animals. Plant remains included barley, wheat, and lentils. The first pottery at Gufkral appeared in Period IB. It was handmade and mostly grey (there were a few red pots), with mat impressions on the base. Big jars, bowls, and basins were the common shapes. The pit dwellings disappeared. A 5–7 cm thick compact clay floor mixed with lime was

found extending over the excavated area. There was also a mud and rubble wall and another compact 70 cm wide wall-like structure. Polished stone tools, as well as bone tools, continued. Bones of red deer, ibex, bear, sheep, goats, cattle, and fowl occurred in Period IB. Many bones had sharp cut marks on them. The proportion of wolf bones decreased and those of dogs increased. The animal bones indicate that although hunting remained important, there was a significant increase in the domestication of sheep, goats, and cattle. The grains of Period IA continued into IB, with the addition of the common pea (Pisum arvense). The presence of large quantities of charcoal and charred wood pieces indicated the occurrence of an extensive fire. A radiocarbon date from Period IB gave a range of 2468–2139 BCE. The upper levels of Period IC at Gufkral were dated c. 1620–1300 BCE, so the beginning of this period can be placed in c. 2000 BCE. In this phase, there were many refuse pits and dumps. Wheelmade pots appeared and included grey, burnished grey, red, and black wares. There were new shapes like long-necked jars and dish-on-stand with triangular perforated designs on the stem. There were stone querns, pounders, and double-holed harvesters. Only one neolithic celt was found. Stone and terracotta spindle whorls with large holes suggest the weaving of woollen cloth. There were terracotta bangles and potsherds with graffiti marks. One copper hairpin with a flattened spiral head was discovered. The largest number of bone tools were found in this period. Animal bones included those of domesticated sheep, goat, cattle, pig, and dog. There were also bones of fish, hare, rodents, hedgehog, and beaver. All the grains of Period IB continued in this period. Hunting continued to decline in importance and the scale of animal breeding correspondingly increased. There are some similarities between the neolithic sites of Kashmir and those in the Swat valley in north Pakistan. The archaeological sequence in the Ghaligai cave in the Swat valley may go back to c. 3000 BCE. Here, at the lowest levels, there was coarse handmade pottery. Some pots had a slip and others a burnished interior. Pebble tools and bone points were also found. Although there are some similarities with the pottery types found at Burzahom Period I, polished stone tools are absent in the Ghaligai cave. A number of grave sites have been explored in the Swat valley. These include Loe-banr, Aligrama, Birkot Ghundai, Kherari, Lal-batai, Timargarha, Balambat, Kalako-deray, and Zarif Karuna. Various kinds of burials have been identified—flexed burials, cremation, urn burials, fractional burials, and multiple burials. Loebanr III and Aligrama have given evidence of wheat and barley. Rice, lentils, and field or common pea were found at Loebanr III, and a grape seed (Vitis vinifera) was also identified. Remains of pit-dwellings, some which must have had thatched roofs on wooden superstructures, were found at Loebanr III and Kalako-deray. Jade beads found at the former site suggest exchange with central Asia. Surface finds of neolithic axes, chisels, and ring stones occur at sites such as Ror, Baroli, and Dehra Gopipur in the Kangra district of Himachal Pradesh. These tools were found along with choppers and flake tools, but the dates of the neolithic context in this area are uncertain. RAJASTHAN

In the areas of Rajasthan, Malwa, and the northern Deccan, the beginnings of settled life are associated with a chalcolithic rather than a neolithic phase. Reference was made in the previous chapter to Bagor in eastern Rajasthan; this site shows a transition from the hunting-gathering mesolithic phase to a chalcolithic and then an iron age phase. Much more substantial evidence of

early sedentary chalcolithic sites comes from areas rich in copper ores. Copper ores occur in many parts of India—Rajasthan, Gujarat, Bihar, Uttar Pradesh, and Andhra Pradesh, but the richest mines are in Rajasthan, Gujarat, and Bihar. There is evidence of the use of copper in certain parts of the subcontinent from about 3000 BCE onwards. Many of the protohistoric cultures discussed in this and the subsequent sections are named after sites where they were first discovered. Archaeological cultures are also sometimes named after a pottery type. This does not mean that this is the only pottery type that occurs, simply that it is a diagnostic type. Cultures can also be named after the region in which they are concentrated. This does not necessarily mean that their sites are not found outside that particular area. For instance, Malwa culture sites are also found outside Malwa in Maharashtra. Similarly, some Ahar culture sites are found in Malwa, outside their nuclear zone in south-east Rajasthan. These are all archaeological cultures, sharing a range of associated material remains. What else they shared, apart from material culture, is a matter of interpretation.

MAP 3.4 THREE MAJOR CHALCOLITHIC SITES OF RAJASTHAN

The Ganeshwar–Jodhpura culture was located in the north-eastern part of Rajasthan. Over 80 sites of this culture have so far been identified. The largest concentration is in Sikar district, but sites also occur in neighbouring Jaipur and Jhunjhunu districts. The site concentration can be connected with the copper ore resources of the Baleshwar and Khetri areas, where traces of ancient copper working have been found. The Ahar culture was located in the south-eastern part of Rajasthan. The profiles of

these sites show that they were part of an important process of metallurgical growth in Rajasthan, the roots of which go back to the 4th millennium BCE. Jodhpura, on the banks of the Sahibi river, is the first site where evidence of the Ganeshwar– Jodhpura culture was identified. The typical pottery here is wheel-made, orange to red in colour, with incised designs. Shapes include dish-on-stand with a thick slip. The calibrated dates from Jodhpura range between 3309–2709 BCE and 2879–2348 BCE. Pottery similar to that found at Jodhpura was later discovered at Ganeshwar, near Nim-ka-Thana. There are three cultural phases at Ganeshwar. The dates for Period I are from c. 3800 BCE onwards, Period II from c. 2800 BCE, and Period III from c. 2000 BCE. Period I reflects a hunting-gathering community using microliths made of chert and quartz. Charred bones, almost all belonging to wild animals, were found. The lower levels of Period I showed a predominance of bones of small animals, while the higher levels were dominated by those of larger animals. Period II was marked by the beginning of metallurgy. A few copper objects were found—five arrowheads, three fishhooks, one spearhead, and one awl. People lived in circular huts with floors paved with pebbles and rock fragments. There were lots of microliths and animal bones. Both handmade and wheel-made pottery was found. There was a profusion of Ganeshwar–Jodhpura ware, a poorly fired pottery made of micaceous clay, with a bright red slip. There were also a few pots made of well-fired, welllevigated clay. Period III had a wide range of pots. Hundreds of copper objects of different types— arrowheads, spearheads, celts, chisels, rings, bangles, balls, etc.—dominated the assemblage, with a corresponding decline in the number of microliths and animal bones. Strangely enough, the reports on Ganeshwar do not mention any direct evidence of copper smelting (furnaces, crucibles, etc.). But the hundreds of copper objects found at this small 1.2–1.6 ha site suggest that it had emerged as a copper-working centre and that its people were supplying these items to communities elsewhere. There are similarities between the wheel-made pottery of Ganeshwar Period II and early Harappan pottery. The early Harappans may have been obtaining their copper from here. This site may also have been one of the major suppliers of copper to the mature Harappan culture. Harappan pottery was found on the surface at two Ganeshwar culture sites. At Ganeshwar itself, there is a reserved slip ware which is only found in the Harappan context at Banawali and a few other places. Double spiral-headed pins from Ganeshwar have been found at some Harappan sites. All this suggests cultural contact between the Ganeshwar and Harappan cultures. In south-east Rajasthan, over 90 sites of the Ahar or Banas culture have been identified in the Banas and Berach river systems, roughly between Udaipur and Jaipur. Some sites also occur in the Malwa plateau of Madhya Pradesh. Ahar, Gilund, and Balathal are the three excavated sites. Ahar was excavated in 1953–54 and 1961–62, Gilund in 1959–60, and Balathal in 1994–98. The typical Ahar pottery is a black-and-red ware with linear and dotted designs painted on in white. Ahar culture sites tend to be located along river banks and generally range in size from a few ha to over 10 ha. However, Ahar itself is at least 11 ha and Gilund is about 10.5 ha. Many of the sites were located within 8–17 km of each other. Ahar is located on the outskirts of Udaipur. Period I is divided into three phases—Ia, Ib, and Ic. The earliest calibrated dates for these three phases are c. 2500 BCE, 2100 BCE, and 1900 BCE. Fifteen building phases were identified in Period I. Ordinary houses were made of mud and rested on stone foundations. Walls were strengthened by bamboo screens or quartz nodules, and roofs were

probably sloping. Floors were made of black clay mixed with yellow silt, sometimes paved with gravel from the riverbed. No complete house plan was exposed, but there were vestiges of a house, about 10.31 m long, partitioned off by a mud wall. Multiple-mouthed ovens were also found. Artefacts included microliths. There were plenty of copper objects—rings, bangles, antimony rods, a knife blade, and four socketless axes. Copper sheet and slag indicated that copper was smelted locally. Saddle querns and beads of semi-precious stones (including one made of lapis lazuli) and terracotta beads or spindle whorls were discovered. Rice grains and bones of cow, buffalo, goat, sheep, deer, pig fish, turtle, and fowl were identified.

MAP 3.5 AHAR CULTURE SITES, RAJASTHAN

An iron ring and nail occurred in Period Ib at Ahar, and iron objects (arrowhead, chisel, peg, and socket) are quite common in Period Ic. Whether the levels at which these artefacts were found were intact or disturbed is a subject of debate. There is every possibility that this constitutes one of the earliest occurrences of iron in the subcontinent. The discoveries at Gilund were broadly similar to those at Ahar. The structural remains included a mud-brick complex, measuring about 30.48 × 24.38 m, and part of a wall made of burnt bricks resting on a foundation of stone rubble. Storage pits were also found. The artefacts included

microliths, fragments of copper, and beads of semiprecious stones. There were terracotta gamesmen and figurines of animals, including humped bulls with long horns. Balathal in Udaipur district is an important Ahar culture site. The first phase of occupation (Period I) is relevant here. The size of the site was about 2 ha. In the early phase of Period I, there were remains of small, circular wattle-and-daub huts with mud-plastered floors and two plastered storage pits. In the later phase, a striking discovery was the remains of a massive mud fortification wall in the centre of the mound. The wall was reinforced in places with stone and there was clear evidence of bastions. The width of the walls ranged from 4.80 m to over 5.0 m, and the fortifications enclosed an area of over 500 sq m. A street (ranging from 2 to 4.8 m in width) running north-west to south-east, along with a small lane, were also exposed. In the second phase of Period I, the houses were larger rectangular units made of mud, mud-bricks, and stone, resting on stone foundations. Three multi-roomed structural complexes were discovered, within which kitchens and storage areas were identified. Two potters’ kilns were also found. The Balathal pottery was of many different types. It included thin red, tan, black-and-red, and buff-coloured pots. There was also a reserve slip ware, in which the pots were first treated with a thin red wash and then with a thick dark red slip, on which designs were made with a comb-like instrument while the slip was still wet. The thick, coarse wares included a red-slipped ware, plain red ware, burnished grey ware, and plain grey ware. There were very few stone microliths or blades. Lots of copper artefacts were found—choppers, knives, razors, chisels, and barbed and tanged arrowheads. There were also bone tools such as points and scrapers; stone querns, grinders, and hammer stones; and terracotta balls and stylized figurines of bulls and sheep. Ornaments included necklaces made of terracotta, steatite, faience, and semiprecious stones like carnelian, agate, and jasper. There were also bangles of copper, shell, and terracotta. The large quantity of animal bones found at Balathal included those of gaur, nilgai, chausingha, blackbuck, fowl, peafowl, turtle, fish, and molluscan shells. Wild animals accounted for only 5 per cent of the bones. Much more numerous were bones of domesticated cattle, buffalo, sheep, goat, and pig. Cattle bones constituted almost 73 per cent of the faunal remains. The plant remains included wheat, barley, at least two varieties of millet, black gram, green gram (moong), pea, linseed, and fruit such as jujube (ber). Cereals and lentils seem to have been grown in large quantities and stored in storage bins, of which several have been found. Grain was ground into flour on stone querns, and the bread was probably cooked on handmade flat pans (tawas) on u-shaped chulhas, similar to those used in the village even today. Calibrated dates indicate that the protohistoric settlement at Balathal goes back to the late 4th millennium BCE. This would make it contemporary with the early Harappan phase at Kot Diji and as early as the Jodhpura–Ganeshwar culture of north-east Rajasthan. Sites of the Ahar culture show the use of a great variety of raw materials including steatite, shell, agate, jasper, carnelian, lapis lazuli, copper, and bronze. Although the shell objects were locally made, the shell itself must have come from the Gujarat coast. The discovery of etched carnelian beads, a lapis lazuli bead, and Rangpur-type lustrous red ware in Ahar Period IC suggest a connection with Harappan sites in Gujarat. THE MALWA REGION

Stone celts discovered in various parts of central India may belong to a neolithic context, but the evidence has not been adequately studied. There is, however, a good deal of evidence concerning the

sequence of chalcolithic farming cultures in the Malwa region, beginning with the Kayatha culture, followed by the Ahar culture, and then the Malwa culture. Calibrated radiocarbon dates place the Kayatha culture in the second half of the 3rd millennium BCE. This culture gets its name from the site of Kayatha in Ujjain district, on the banks of the Chhoti Kali Sindh, a tributary of the Kali Sindh, which is in turn a tributary of the Chambal river. Three types of pottery have been found at Kayatha culture sites. The typical Kayatha pottery is a fine, sturdy, wheel-made ware. It has a thick brown slip, usually from lip to shoulder, and sometimes up to the base. Linear designs are painted on in violet or deep red only on the upper part of the vessel, especially on the rim. The shapes include bowls and basins, vases with a globular profile and concave neck, and large storage jars. Other kinds of pottery associated with Kayatha ware include a buff ware with a thin, fine fabric and linear and geometric designs painted on in red. The shapes are rather limited and include lotas, high and short concave-necked jars, and basins. Thirdly, there is a red ‘combed ware’ with a fine fabric, usually without any slip or wash. It is decorated with multiple wavy and zigzag lines made with some kind of comb-like instrument. The shapes consist only of bowls and basins. As the Kayatha excavations were restricted in scope, no complete house plans were uncovered. But houses were apparently made of mud and reed with mud-plastered floors. The bones of domesticated cattle and horses were found, and the people seem to have eaten tortoises. No grain remains were identified. There was a rich repertoire of stone and copper artefacts. The stone tools included plenty of microliths (blades, points, lunates, etc.) made out of locally available chalcedony. A mace head or ring stone may have been used as an agricultural implement for turning soil. The people were well versed in copper technology. There were two copper axes cast in moulds, a fragmentary chisel, and 28 copper bangles found in two pots. Two beautiful necklaces made of agate and carnelian beads (and one faceted crystal) were discovered in two pots—one consisting of 175, the other of 160 beads. Another pot contained 40,000 steatite micro-beads, strung in threads. The copper axes, bangles, and necklaces were all found in a small area of what must have been a house. It seems that the people who lived here had to leave suddenly, abandoning their valuable possessions on the floor. Kayatha ware is similar in some respects to early Harappan pottery, and there is also a similarity in the steatite micro-beads of these two cultures. The axes found at Kayatha have indentation marks that are similar to those found on Ganeshwar specimens, and it is quite possible that they came from Ganeshwar. All this suggests connections, whose precise nature is difficult to determine. There was an abrupt break in occupation at Kayatha in about 1800 BCE, and the site remained deserted for about a century. When reoccupied, it represented an Ahar/Banas culture phase. THE WESTERN DECCAN

The earliest farming culture in the western Deccan is the Savalda culture, named after the site of this name in the Tapi valley. This culture goes back to the 3rd millennium BCE, and its sites are found between the Tapi and Godavari rivers in north Maharashtra. The typical Savalda ware is a wheelmade chocolate-coloured pottery, of medium to coarse fabric, with a thick, crackled slip. The variety of shapes includes the high-necked jar, dish, dish-on-stand, bowl, basin, ring stand, vase, beaker, and knobbed lid. A remarkable aspect of Savalda pottery is that the designs painted over the thick, crackled slip include tools, weapons, and geometric motifs.

Kaothe is a site belonging to the Savalda culture. It is a 20 ha site, and the shallow 50 cm thick deposit suggests a short-duration, nomadic occupation. The houses seem to have been round or oval, with a sloping roof. Many bone tools and beads made of shell, opal, carnelian, and terracotta were found. Bones of wild deer and domesticated cattle, buffalo, sheep/goat, and dog were identified. Plant remains included a variety of millet and two kinds of pulses—gram and moong. The pottery consisted of a sturdy ware with geometric and naturalistic designs. Daimabad on the banks of the Pravara river (a tributary of the Godavari) in Ahmednagar district of Maharashtra also has a Savalda culture phase. The evidence here did not indicate a semi-nomadic community. There were mud houses, some large and multi-roomed, with hearths, storage pits, and jars. Sometimes there were courtyards in front, and a lane has been traced in one place. The excavations yielded microliths, bone and stone artefacts, and a few beads of shell, carnelian, steatite, and terracotta. A phallus-shaped object made of agate was found inside a house. Plant remains included wheat, barley, pea, lentil, black gram, and green gram. THE MIDDLE GANGA PLAIN AND EASTERN INDIA

In a previous section, reference was made to early evidence of food-producing settlements in the northern Vindhyan fringes at Koldihwa, Mahagara, and Kunjhun, and in the middle Ganga valley at Lahuradeva. Sites of a subsequent period have been found in the Sarayupara plain in the northeastern part of Uttar Pradesh. This is part of the middle Ganga plain, bound on the south and west by the Ghagara and on the east by the Gandak, extending up to the foothills of the Himalayas. An important site in this area is Sohagaura in the Bansgaon sub-division of Gorakhpur district, at the confluence of the Rapti and Ami rivers. The village lies on a mound about 60 ha in area. Excavations in the 1960s and 1970s brought out a six-fold cultural sequence at Sohagaura, ranging from the neolithic (Period I) to the medieval (Period VI). The remains of Period I included small pieces of ill-fired, handmade pottery with a coarse or medium fabric, most of the sherds either rusticated or cord impressed. There are several neolithic and neolithic-chalcolithic sites in the alluvial plains of north Bihar. Five have been excavated—Chirand, Senuar, Chechar-Kutubpur, Maner, and Taradih. All these sites mark 3rd/2nd millennium BCE villages located on the banks of streams and show the presence of full-fledged agricultural villages in the Gangetic plains of Bihar. Chirand (in Saran district) is a huge mound, about 1 km long, situated at the confluence of the Sarayu and Ganga rivers. A 3.5 m thick occupational deposit was excavated here. The beginning of the occupation may go back to before the mid-3rd millennium BCE. Stone celts and hammer stones were made out of quartzite, basalt, and granite. Various other kinds of tools, pestles, querns, and balls were found. Microlithic blades and points were made from materials like chalcedony, chert, agate, and jasper. There were a large number and variety of bone and antler implements such as celts, scrapers, chisels, hammers, needles, points, borers, awls, diggers, and pins. Bone ornaments included pendants, earrings, bangles, discs, and combs, and there were also bangles made of tortoise bone and ivory. The pottery of neolithic Chirand included red, grey, and black wares. There was also a black-andred ware. Most of the pottery was handmade, though there were some examples of the turntable method. Some pots had painted (usually red ochre) and scratched designs on their surface, generally linear or geometric. The exterior of many grey pots was burnished. The shapes included various kinds of vases and bowls. There were different varieties of beads of agate, carnelian, jasper, marble,

steatite, and faience—long tubular, long barrel, short barrel, cylindrical, triangular, and disc-shaped. Some of them were unfinished, indicating they were locally made. No copper objects were found. Terracotta figurines included representations of humped bulls, birds, and snakes. There were also terracotta beads, bangles, wheels, balls, and what seem to be two fragments of a brooch. A small perforated stem had traces of soot inside—perhaps it was a smoking pipe. A few terracotta discs with holes in the centre may have been spindle whorls.

MAP 3.6 VILLAGE SETTLEMENTS IN THE MIDDLE G ANGA PLAIN

The neolithic people of Chirand lived in circular wattle-and-daub huts with rammed floors. In the early stage, floors were below ground level, but later they were at ground level. Hearths were found in the houses. A semi-circular hut had several oblong ovens, perhaps for community cooking. Mud boundary walls of houses were traced. Burnt chunks of clay with reed or bamboo impressions suggest that the houses were destroyed by a fire. Plant remains included rice, wheat, barley, and lentils such as moong and masoor. Lots of bones of animals, birds, and fish were identified, indicating the prevalence of hunting and fishing. Clusters of fish scales and remains of river shells and snails give additional information on the food habits of the people. Animal remains included bones of wild elephants, rhinoceros, and deer, and those of domesticated cattle. Chirand had a later, chalcolithic occupation level as well. Chechar-Kutubpur is a site located on the banks of the Ganga, across the river from Patna, near Biddupur. The neolithic deposit here was divided into three phases (A, B, and C) on the basis of changes in pottery. People lived in circular wattle-and-daub huts with mud floors. There were hearths in the centre of the floors. Lots of bone and antler tools and micro-beads of steatite and chalcedony were found here.

The ancient mound of Senuar (Singh, 2003) is in on the banks of the Kudra river at the foot of the Kaimur range, in Rohtas district of Bihar, not far from Sasaram. There are four periods of occupation at the site: Period I is neolithic, Period II chalcolithic, Period III represented the Northern Black Polished Ware (NBPW) culture, and Period IV belongs to the early centuries CE. The lower levels of Period IB were dated by radiocarbon to c. 1770–1400 BCE; therefore the beginning of Period IA probably goes back to the latter half of the 3rd millennium BCE. Here we are concerned with Period I, which is divided into Periods IA and IB. Period IA at Senuar was a 1.5 m thick neolithic deposit with remains of wattle-and-daub houses. There were three main kinds of pottery—a red ware, burnished red ware, and burnished grey ware. Some of the pottery was rusticated, some had designs made by cord impressions. The shapes included the wide-mouthed shallow bowl, channelled bowl, vase, and spouted vessels. Most of the pottery was wheel-made, but there was also some handmade pottery. Lots of microliths (small bladelets, also flakes and blades) made of chert, chalcedony, agate, quartz, and quartzite were found. There were a few triangular polished celts, stone pestles, saddle querns, hammer stones, and sling balls of various sizes. Bone tools included points, with use marks at the tip. Beads of semi-precious stones were also discovered. The animal bones from Senuar have been carefully studied. The domesticated animals included cattle, buffalo, sheep, goat, pig, cat, and dog. Wild animals included nilgai, antelope, and chital. The charring and cut marks on many of the bones showed that the animals were killed for food. That the people ate shell food from the river is clear from the remains of molluscs and large numbers of shells. Considering that the site is on the banks of a river, it is odd that no fish bones were reported. Carbonized grains show that people grew two crops a year. Rice (Oryza sativa) was the main crop, but people also grew barley, dwarf wheat (Triticum sphaerococcum), sorghum millet, ragi millet, lentil, grass pea (Lathyrus sativus), and field pea (Pisum arvense). Period IB at Senuar was neolithic–chalcolithic and consisted of a 2.02 m thick deposit. House floors were made of well-rammed earth mixed with kankar and potsherds, and there were marks of post-holes in some places. Nineteen copper objects were found, including a fishhook, wire, some rings, a broken needle, and several broken and indeterminate objects. There was also a fragmentary lead rod. Chemical analysis of the copper wire showed that it was made of almost pure copper and that the metal was probably obtained from the neighbouring Rakha mines. The artefacts of Period IB were more or less similar to those of Period IA, but there was a marked improvement in the pottery, especially in surface treatment. Although most of the pots were wheel-made, there were some handmade pieces as well. The vessels had a fine slip and a high grade of burnishing. Post-firing red ochre coloured paintings—earlier only found on the burnished grey ware—were now also found on the burnished red ware. Painted decoration was much more frequent, and pots were often also decorated with thumb or finger impressions, rope, or notched patterns on appliqué bands of clay. There were more stone tools in Period IB than in the earlier phase, including many polished stone celts, mostly made of black basalt. Microliths were also found in large numbers. The material of the tools was the same as in Period IA, but there were a few new shapes. Shell ornaments included triangular pendants. There were lots of finished and unfinished beads of semi-precious stones such as agate, carnelian, and jasper. Twenty-five faience beads were also found. Terracotta artefacts included beads, pottery discs, a bull figurine, and maybe a whistle. Some of the pottery discs may have been wheels for toys or gaming counters used by children. Those with holes may represent

spindle whorls. Apart from the grains that continued from Period I, in Period IB, there were some more plant remains—those of bread wheat (Triticum aestivum), chickpea or gram (Cicer arictinum), and moong (Vigna radiata). There are some cultural similarities between neolithic Chirand and Senuar. The site of Maner is located on the banks of an old course of the Ganga, not far from Patna. The neolithic deposit here was 3.45 m thick and yielded handmade red ware and burnished red and grey wares. The shapes included the long-necked vase, bowl with short stem, lipped bowl, and spouted bowl. Other artefacts included stone microliths, bone points, and terracotta spindle whorls. Taradih is situated close to the Mahabodhi temple at Bodh Gaya. There are two phases of the neolithic occupation here—Period IA had hand-made burnished and un-burnished red wares and cord-impressed wares. Period IB was marked by a handmade burnished grey ware, sometimes with post-firing ochre-coloured painting. The other artefacts included neolithic celts, microliths, and bone tools. There were remains of wattle-and-daub houses with hearths. Bones of cattle, goat, buffalo, pig, sheep, deer, bird, fish, and snail were identified. Plant remains included grains of rice, wheat, and barley. Neolithic tools—ring stones, shouldered celts, and triangular and rectangular axes—have been found in various parts of West Bengal, but the dates of the finds remain uncertain. Kuchai is an excavated site in Orissa which yielded faceted hoes, chisels, pounders, mace heads, and grinding stones. There was also a reddish brown pottery tempered with coarse grit, some with a slip and incised decoration. Neolithic material such as faceted and shouldered celts, bar chisels, rounded butt axes, wedges, and hammer stones occur as surface finds in Mayurbhanj district, but there is a lack of clarity about their dates and cultural contexts.

CELTS FROM NAYAPUR AND KUCHAI ; SHOULDERED CELT FROM KUCHAI

The north-eastern states of Assam, Meghalaya, Nagaland, Arunachal Pradesh, Mizoram, and Manipur have not yet been properly explored for prehistoric sites. Large numbers of polished stone

tools have been found in various parts of the Khasi, Garo, Naga, and Cachar hills, but their cultural context and dates are uncertain. We have to keep in mind the fact that polished celts are even found at historical levels at certain sites. Sarutaru, Daojali Hading, and Marakdola—all in Assam—have been excavated. These sites will be discussed in Chapter 5, as the neolithic levels here may be fairly late. SOUTH INDIA

The dates of the southern neolithic sites mostly fall within the broad time bracket of c. 2900–1000 BCE, but they can be further divided on the basis of chronology and geographical region. The earliest dates so far range between c. 2900 and 2400 BCE and come from Utnur, Pallavoy, Kodekal, and Watgal. These and other early sites are discussed in this section, while the later ones will be discussed in Chapter 5. The widespread palaeolithic and mesolithic occupation in peninsular India was discussed in the previous chapter. At present, there is insufficient information on the dates of the mesolithic phase in the far south, and the connections between the mesolithic and neolithic phases have not been properly worked out. The meagre evidence of neolithic sites along the south-east coast of India is strange, considering that this area has yielded evidence of palaeolithic and mesolithic artefacts. Apart from a neolithic site at Pondicherry on the Tamil Nadu coast, there seems to be an absence of sites in the deltas of the Pennar, Krishna, and Godavari rivers. This may be due to sites being swallowed up by the riverine silts or due to inadequate exploration. However, there are many sites in the middle and lower Krishna valley. In the southern part of the Deccan plateau, where granite hills rise from the black cotton soil, the earliest neolithic villages were generally located on hillsides and plateaux, sometimes along minor streams, and occasionally along the banks of major rivers. A distinctive feature of many sites in this region is that they are marked by ash mounds. Research into the southern neolithic has in fact been dominated by a discussion of the ash mounds. The two key areas are the Raichur doab, between the Krishna and the Tungabhadra, and the Shorapur doab, between the Bhima and the Krishna. Ash mounds have been excavated at Utnur, Kupgal, Kodekal, and Pallavoy. The ash mound sites are large accumulations of ash and vitrified material, created by the repeated burning of heaps of cow dung. They mark neolithic cattle pens which were surrounded by heavy enclosures made of tree trunks. Cattle breeders in parts of central and South India pen their animals in similar enclosures even today. Some of the neolithic pens were attached to permanent settlements, while others may have been temporary camps. The periodic burning of heaps of dung may have been connected with seasonal festivals marking the beginning or end of annual migrations to the forest grazing grounds. Modern pastoralists in peninsular India still burn bonfires on such occasions, and cattle are driven through fire, as it is believed that this will protect them from disease. Excavations at Utnur (in Mahbubnagar district, AP) have shown that the wooden enclosure of the cattle pen here was rebuilt many times, and the dung within it was likewise burnt repeatedly. Cattle hoof-prints were found in the ash. The size of the enclosure indicated that it could have held about 540–800 cattle. Utnur gave evidence of a small amount of ground stone axes, stone blades, and a handmade coarse pottery. The latter included a burnished grey or buff ware (usually plain, sometimes with post-firing designs painted on in red ochre), and also a ware with a red, black, or brown dressing applied to it before burnishing and firing (sometimes with pre-firing black or purple

painted designs). The material culture of Utnur was similar to that of sites such as Piklihal (dated from c. 2100 BCE) and Kodekal. FURTHER DISCUSSION

The mystery of the ash mounds

The first reports of the ash mounds appeared in the 1830s and 1840s. They were described as ‘cinder mounds’ or ‘cinder camps’, and many thought they were of volcanic or limestone origin. T. J. Newbold carried out the first excavation of an ash mound site during this period. In the course of his excavation at Kupgal, he found remains of pottery, animal bones, and a rubbing stone. This convinced him that the mounds were not natural geological formations but were created by people. In the late 19th century, the geologist-prehistorian Robert Bruce Foote became the first to connect the ash mounds with the neolithic culture. On the basis of his excavation at the site of Budikanama (also known as Kudatini) and a chemical analysis of the ash mound material, he argued that the ash mounds were heaps of excessively burnt cow dung, created by neolithic cattle herders. Few were convinced by Foote’s argument. Robert Sewell argued that not all the ash mounds represented cattle camps and that some of them might belong to the medieval period. G. Yazdani suggested that the mounds may have been created by metal workers in gold or iron. There were many others who bought the argument that ancient iron-smelters were responsible for the creation of the ash mounds. In the 1950s, Raymond Allchin and F. E. Zeuner made important contributions towards the understanding of the ash mounds. Zeuner submitted the ash of Kudatini to a chemical and microscopic study. This established beyond all doubt that the mounds were made out of dung, most likely cattle dung. Allchin under took an archaeological survey of the Raichur doab and excavated the habitation site of Piklihal and the ash mound site of Utnur. The Utnur excavation connected the ash mound at this site with a rectangular enclosure surrounded by post-holes, which Allchin interpreted as a cattle pen. Zeuner and Allchin’s investigations indicated that Foote had been right after all. It also became evident that the accumulations of cattle dung had been burnt not once but many times; this repeated burning seems to have been deliberate, not accidental. A number of questions remained: Did the ash mounds represent in situ burning of dung that had accumulated naturally over time, or was the dung collected, deliberately heaped up, and then burnt? Why was it burnt at regular intervals? Was it in order to periodically clean up the cattle pens or did this activity have some sort of symbolic significance? Allchin suggests that the ash fires may represent annual seasonal rituals of purification. Another problematic issue was the relationship between the ash mounds and the settlements. Allchin suggested that there were two kinds of ash mounds—those in or near permanent

settlements (such as Kupgal and Gadiganur) and others not associated with any settlements (including some of the largest ones, e.g., Kudatini and Utnur). On the other hand, on the basis of his excavations at Budihal, K. Paddayya suggested that the ash mound and habitation areas were not two separate, different types of sites, but were, in fact, related to each other. He also argued that the ash mounds were not an in situ accumulation of dung, but that dung and garbage cleared from penning and house areas was piled up here and then burnt. Ash mounds do not occur at all southern neolithic sites. In the Pennar basin in Cuddapah district of Andhra Pradesh, there are neolithic sites but no ash mounds. The mounds are similarly absent from sites in the upper Tungabhadra valley and southern Karnataka. It has been suggested by P. C. Venkatasubbaiah that the absence of ash mounds in the Cuddapah district may be because of differences in subsistence systems. In this area, people practised animal breeding, but they also relied on millet and pulse farming. Due to the importance of agricultural activity, cow dung was used as manure and was therefore not burnt for ceremonial or other purposes. An alternative explanation is that even if agriculture was practised (and there is increasing evidence that it was) at many southern neolithic sites, manuring was not necessary. In such a situation, dung and dung ash could have been used for plastering houses, but they were not the valuable resources they represent for villagers today. The reasons for the presence or absence of ash mounds at southern neolithic sites would, in this case, have more to do with differences in cultural traditions rather than in subsistence practices. The relative dates of the ash mound and non-ash mound sites are not yet fully clear. More investigations are required, and it is likely that not all the ash mound sites represent the same sort of settlement pattern. SOURCE Korisettar et al., 2003

Recent excavations at Watgal and Budihal incorporated new archaeological approaches and techniques, and were marked by an especially careful collection and analysis of faunal and botanical remains. Watgal (Devaraja et al., 1995) is located in Raichur district of north Karnataka. The earliest calibrated radiocarbon date from this site gives a range starting from 2900–2600 BCE, and occupation continued into the 1st millennium BCE. Period I had a microlithic industry consisting mainly of blades and lunates made of chert and quartzite. There were also large flakes of basalt and dolerite. The calibrated date range for Period IIA at Watgal is c. 2700–2300 BCE. This period was marked by increasing diversity in stone tools. There were underground storage pits. Two carbonized seeds of betel nut (Areca catechu) were found. This is the earliest evidence of the use of betel nuts in South Asia. Period IIA was dominated by microliths made of chert. Most of the pottery was handmade, while some may have been made on a slow wheel. It was ill-fired and consisted of coarse red and grey wares, as well as a burnished grey ware with post-firing painting in red ochre. There were other artefacts such as beads made of marine shell. The burials included one urn burial and two extended burials marked by stones, without grave goods.

MAP 3.7 SOME IMPORTANT NEOLITHIC SITES IN SOUTH INDIA

The calibrated range for Watgal Period IIB is c. 2300–2000 BCE. Here, as in the earlier subphase, there were numerous storage pits. The burials included both urn burials and extended burials marked by stones. But there was a new feature—pots appeared as grave goods. The range and number of artefacts were also greater. They included microliths and milling stones, beads of marine

shell, stone, and terracotta, and a shell pendant. A small iron fragment may have been an intrusion from later levels. Animal and human terracotta figurines (one clearly representing the torso of a female) were found. There was a continuity of earlier pottery types, with a slight increase in the amount of wheel-made pottery. Periods III and IV at Watgal are post-2000 BCE and show evidence of copper/bronze and iron. Budihal (in Gulbarga district, Karnataka) has been excavated by K. Paddayya and others (Paddayya, 1993). One of the aims of the excavations was to understand the ash mounds in relation to their ecology and the material evidence around them. The site is located on a sandstone plateau covered with thin brown soil. A complex of four localities (I–IV) within a 400 × 300 m area was identified. Each locality consisted of an ash mound as well as habitational deposit. In the extreme west of the site, an extensive area (about 4.5 ha) was found littered with a huge number of chert tools and waste chert material, and nothing else. Huge sandstone boulders found nearby showed marks of small and big grinding grooves, places where people must have worked at grinding and polishing stone tools. This was clearly a chert blade-working area. It is possible that chert tools made at this site were sent to other neolithic settlements in the Shorapur doab and perhaps even further. Excavations in Locality I (the main part of the site) at Budihal clearly showed that the ash deposits were located in the centre. Within the ash mound area, two distinct parts were identified—a cattlepenning area on the east and a cow dung disposal area on the west. There were several episodes of cattle penning, dung accumulation, and burning. A dozen structures were identified in the 1.34 ha habitational area around the ash deposit. One was a platform-like surface for chert working (chert was available 5–6 km north of the site) and another was a place for storing pottery. The rest were round dwelling units with low walls made of blocks of stone packed in mud. A total of 10 child burials (some in pits, others in pots) were found in the habitational area. The artefacts found from the ash mound and residential area included red and grey pottery, ground stone tools, chert blades, bone tools including axe heads, and beads of shell, bone, and semi-precious stones. Seeds of three types of wild plants were identified through the flotation of soil samples—ber, Indian cherry, and amla (Emblic myrabolans). A few grains of domesticated horse gram were also found. Faunal remains of about 15 domesticated and wild animal species were identified. Bones of domesticated cattle were the most numerous. This shows that the neolithic people of Budihal specialized mainly in cattle rearing and to a lesser extent on sheep, goat, buffalo, and fowl. The bones of wild fauna included nilgai, blackbuck, antelope, monitor lizard, tortoises, birds, fish, crabs, and molluscs. An even more interesting discovery was that of a butchering area within the settlement area, on the southern side of the ash mound. Eleven radiocarbon dates ranging between c. 1900 and 1400 BCE are available for the ash mound and habitational area at Budihal. When calibrated, they give a range of 2180–1600 BCE. NEW DIRECTIONS IN RESEARCH

Community feasting at neolithic Budihal

In one of the trenches excavated to the south of the ash mound, within the habitational area of Locality I at Budihal, the archaeological team discovered patches of floor made of kankar-like

material. Chemical analysis showed these to be made of fine ash, clay, small pieces of potsherds, bone, and charcoal, mixed with water and then rammed together in order to produce a hard surface. This floor seems originally to have covered an area of 200–250 sq m. Strewn over this were huge numbers of animal bones—mostly those of cattle, but also of sheep, goat, buffalo, and wild animals. The large number of bones and stone tools of various kinds, including chopping tools and chert blades, indicated that this was a butchering area. Sandstone blocks found on the floor may have been used for chopping meat. Splinters of bone and bone artefacts show that some bone tools were made on the spot and were probably used for marrow extraction and hide working. Three small pits (20–25 cm wide and 15–20 cm deep) were found in the northern part of the butchering area. These contained ashy soil, pieces of charcoal, and burnt bones. This was probably where people roasted meat. The large size of the butchering floor, its location between the ash mound and the settlement area, the fact that it was plastered to create a hard and permanent working area, the occurrence of such a large number of bones and tools, and the cooking area nearby—all this suggests that the area was used not by a single person but by the entire community or at least a substantial large part of it. Perhaps it was used on special or ceremonial occasions, when animals were killed and their meat shared among those present. SOURCE Paddayya et al., 1995

The Budihal excavations demonstrated the presence of a habitation site directly associated with ash mounds, and Paddayya made some general observations on this basis. He emphasized that neolithic ash mounds and habitation sites were closely related to each other, and that the ash mound sites are best described as neolithic pastoral settlements with ash deposits. Ash mound sites tend to occur in hilly tracts, close to perennial sources of water, with good pasture land but soils too poor for agriculture. Garbage accumulated from the penning of cattle and other animals was dumped along with household refuse at spots close to the settlement and was periodically burnt. The reasons for the cow dung accumulation and burning were in part practical—to keep the settlement clean, to protect people and animals from health hazards posed by vermin-infected dung heaps, and to scare away wild animals. The burning could also have been part of rituals aimed at promoting the fertility of cattle. Some of the ash mounds are so large that the sites could have served as regional or local centres where people came from afar to attend periodic cattle fairs. While the evidence from Budihal is important because it shows the complementary relationship between ash mounds and what seems to be a long-duration habitation site, it is not yet established beyond all doubt that a similar situation prevailed in other places. It is possible to visualize variations among sites—some may have been single, independent sites, others seem to consist of pairs or clusters (e.g., Kupgal, Budihal, Palavoy). Some may represent short-term camps of pastoralists, others more long-term habitation. There are different views on the subsistence base of the southern neolithic sites. One view is that the neolithic people were fully sedentary farmers who made clearances in forests to carry out agriculture. Another view is that while these people may have practised some amount of agriculture,

they were basically nomadic pastoralists. A third view is that they were sedentary pastoralists who did not practise any agriculture whatsoever. Raymond and Bridget Allchin (1997: 104) argue that ash mound sites such as those at Utnur and Kudatini represent seasonal cattle camps. They also suggest that the evidence reflects a transition from cattle pastoralism (represented at the early ash mound sites) towards agriculture (in the later sites). However, the early date from Watgal, which does not have any ash mounds, shows that the ash mound sites were not necessarily the earliest. The faunal remains, ash mounds, terracotta figurines of humped cattle, and rock bruisings of cattle on rocks around some of the settlements testify to the importance of cattle rearing in the southern neolithic. Cattle (Bos indicus) dominate the faunal assemblage, both in the ash mound and non-ash mound sites. Sheep and goat bones also occur, but in much smaller quantities. Horse (Equus) remains have been reported, but it is not clear whether a wild or domesticated species is represented. Bones of water buffalo and pig (probably both wild and domesticated) occur occasionally. Other faunal remains include the bones of wild and domesticated fowl. Till recently, there was not much evidence of agriculture at South Indian neolithic sites. There were the occasional discoveries of charred grain and the indirect evidence of grinding stones, but cattle rearing seemed to dominate the picture. In fact, some scholars argued that the terrain, soil, and dry climate of the area made it unsuitable for agriculture. Recent research has changed this picture and has highlighted the range of plant remains found at southern neolithic sites (Korisettar et al., 2003). Millets seem to have been the staple crop, but grains of pulses and seeds of ber have also been found. Fragments of areca nut, probably wild, were found at Watgal. So far, there is not much evidence of craft or trade activities at these sites. Although copper and bronze objects occur at several sites, there is no indication of the local smelting or working of copper. Did these objects come via exchange or trade from elsewhere? A pair of gold earrings was found at neolithic Tekkalakota and the Kolar fields of Karnataka are the likely source of the gold found in Harappan contexts. This would imply trade between the urban Harappans and the neolithic communities of South India. Marine shell and marine shell artefacts found at Watgal indicate exchange with coastal areas, probably the western coast. We can note the beginning of the chalcolithic phase at sites such as Singanapalli and Ramapuram in the Kurnool district of Andhra Pradesh. Both have been excavated, but there are no full excavation reports yet. The calibrated range of a date from Ramapuram is c. 2455–2041 BCE. This site gave evidence of house floors plastered with lime, wheel-made painted pottery (mostly black-on-red), microliths, and beads of semi-precious stones. The Life of Early Farmers

As mentioned at the beginning of this chapter, hunting-gathering and food production do not represent two ends of a unilinear evolutionary scheme. In some areas, the advent of food production based on animal and plant domestication did not lead to a complete eclipse of the hunting-gathering way of life. Many communities continued to practise these activities, and continue to do so in some parts of the world, even in the 21st century. Further, archaeological data clearly indicates the practise of hunting and/or gathering at most early farming sites. It also suggests relationships of interaction and exchange between early farmers and hunter-gatherers.

The neolithic stage is generally associated with relatively self-sufficient village communities with equilibrium between food production and population. However, the issue is not only one of the quantity of available food. Food is an essential prerequisite for human survival, but it is also much more. The obtaining and consumption of food is generally a social activity; food items may be part of systems of hospitality, gift giving, trade, and social taboos. Food preferences and ways of preparation are important parts of social life, both within the family and in the larger social group. The site of Budihal gives a graphic image of community food preparation and feasting at a neolithic site. Although certain inferences can be made about the social and political organization of early foodproducing communities, it is necessary to recognize the fact that they were not identical to each other. Some sites reflect small communities with a relatively simple social organization, while larger sites represent more complex societies. The details of the subsistence patterns of the communities would have varied, depending on the resource potential of the environmental niche they lived in and on their methods of adapting to it. Differences in material equipment such as tools, pottery, and houses suggest differences in craft traditions and lifestyles. Burial practices and objects of possible cultic significance reflect divergent belief systems and customs. There is a view that compared to the struggle for existence and lack of leisure time that marked the lives of prehistoric hunter-gatherers, the life of farmers was much easier. As indicated in the previous chapter, the first part of such a view can be questioned. Similarly, it would be an oversimplification to think of the life of early farmers as one marked by comfort and ease. Farmers were in fact a vulnerable lot. As is the case today, lack of rain could mean a bad harvest, pests or disease could wipe out an entire crop, and mould and rodents could destroy precious reserves of stored grain. In spite of the differences in the ways of life of early farmers and the need to abandon stereotypical notions, it is possible to identify certain general features of the impact of the transition from hunting-gathering to food production. It was earlier pointed out that elements of sedentary living can be seen among certain hunting-gathering groups, while some farmers and pastoralists retain a migratory lifestyle. Further, there are different views on whether sedentary living preceded or followed the beginnings of agriculture. However, there is no doubt that in the long run, the transition to agriculture did lead to increasing levels of sedentariness among most communities. Studies of nutrition and disease based on an analysis of human bones suggest that hunter-gatherers had a high-protein diet, one that was more varied, balanced, and healthy compared to that of early farmers, whose diet tended to be high in carbohydrates, with an emphasis on cereals or root crops. Sedentary people were also more vulnerable to infectious diseases and epidemics than nomadic groups. This may help explain the high incidence of disease reflected in the bones of certain early farming communities.

SEE CHAPTER 5, P . 234 FOR DETAILS OF DECCAN CHALCOLITHIC SKELETAL REMAINS

F IGURE 3.7 HUNTING SCENE ENGRAVED ON STONE, B URZAHOM

Living for long periods of time in one place would have led to a more enduring relationship between people and their environmental niche. A sedentary life and the diet associated with agriculture would have meant less stress on women during pregnancy and more stable conditions for mother and child after childbirth. Further, high-carbohydrate diets are connected with decreased birth intervals. All these factors would have combined to produce higher birth rates. Sedentary living would have been easier on children and old people, and may have resulted in reduced death rates and increased life expectancy. Due to such reasons, the advent of food production would, in the long run, have led to an increase in population and changes in the age profiles within communities. Food production required new tool kits and equipment. It also involved a new kind of scheduling of subsistence activities and shifts in the contributions of men and women, children, and aged folk. There would also have been a change in the food ethic—hunter-gatherers generally collect as much food as they can immediately consume on a short-term basis. Farmers would have had to produce and store quantities of food for future use. The focus would no longer have been on the acquisition of food to satisfy immediate needs on a daily basis, but rather on strategies that required much more long-term planning. It has been argued that women may have been in the forefront of experiments related to plant domestication. This argument is largely based on ethnographic studies that connect women with horticulture activities. If, in hunting-gathering societies, men generally hunted and women did the food gathering, then it is indeed likely that the early experiments in agriculture were made by women. Further, since pottery was connected to food storage and cooking, tasks that are generally associated with women, they may have had a significant role to play in technical advances related to pottery making. Studies of modern potters have pointed out that making pots is a lengthy process that involves more than the hands of the potter who gives the pot its final shape. Women—and children— may have been involved in these other activities, including collecting and processing clay, collecting

fire wood, piling it in the kiln, and decorating the pots. While ethnographic evidence is never conclusive, in these instances, it is fairly persuasive, and there is good ground to assume the involvement of women in the important cultural advances made in the transition to food production. Although the neolithic stage is generally associated with subsistence-level activities, there is evidence of specialized crafts and long-distance exchange at sites such as Mehrgarh. Kunjhun and Ganeshwar indicate fairly well-developed craft traditions and site specialization. Many sites give evidence of separate areas within the settlement being earmarked for different activities (cattle rearing, craft production, butchering, etc.). This reflects conscious, collective decisions made by members of the community for organizing space and activities. Evidence cited in earlier sections clearly indicates that some neolithic communities were interacting with proto-urban and urban cultures. When larger groups of people started living together in settled villages, they would have had to devise new ways and norms of interaction and co-operation, ones that were different from those associated with bands of hunter-gatherers. The communities of early farmers and pastoralists must have been internally differentiated on the basis of age and sex. At some sites, differences in the sizes of houses and in the quantity and quality of grave goods suggest the existence of social ranks. Among larger groups, the regulation of economic activities and social relations would have required some sort of effective political control and organization. Changes in Cultic and Belief Systems

Changes in subsistence practices would have involved shifts in symbolic and belief systems. One problem is: How are we to define religious or cultic activities, and how can their traces be identified in the archaeological record? In the previous chapter, we noted that some of the palaeolithic and mesolithic art remains may have been connected with magico-religious beliefs and hunting rituals. The cultivation of crops and the domestication of animals must have led to increased concerns with fertility and magico-religious ways of controlling it. Terracotta female figurines found from neolithic levels onwards at certain sites (e.g., in the north-western zone) have often been given the label of ‘Mother Goddesses’. It is very likely that farming communities connected women with fertility because of the fact that women give birth. It is also possible that they worshipped images of goddesses associated with fertility. However, the interpretation of female figurines is very subjective. Were these figurines goddesses, or were they toys, decorative items, or clay portraits of ordinary women? Similarly, were the humped bull figurines found at sites such as Rana Ghundai, Mehrgarh, Mundigak, Bala Kot, Gilund, Balathal, and Chi-rand cult objects? Unless their form or context suggest religious or cultic significance, it is necessary to be cautious while making inferences about the role and function of terracotta figurines. FURTHER DISCUSSION

Female figurines—ordinary women or goddesses?

At one time, scholars tended to use the ‘Mother Goddess’ label for all female figurines found at sites. This was largely because of the belief that the worship of fertility goddesses was an

important part of agricultural societies all over the world, and also due to a tendency to look at ancient remains through the lens of later-day Hinduism, in which goddess worship had an important place. However, scholars are now increasingly aware of the stylistic and technical differences among assemblages of female figurines. Further, all goddesses need not have been part of a single goddess cult, and not all ancient goddesses were necessarily associated with maternity. In the light of such problems, the term ‘Mother Goddess’ should be replaced by the longer but more neutral phrase—‘female figurines with likely cultic significance.’ This does not mean that none of these figurines might have had a religious or cultic significance. It is indeed possible that some were either images that were worshipped or votive offerings that were part of some domestic cult or ritual. However, not all female figurines necessarily had such a function. Whether we are looking at human or animal figurines, in all cases, their possible significance or function has to be assessed, and cannot be assumed. Apart from their form, the context in which they were found is crucial.

F EMALE FIGURINE, MEHRGARH

Purposeful, standardized burials do not appear for the first time in the neolithic or neolithic– chalcolithic phase, but they do increase in number. Such burials imply significance attached to the bodily remains of the deceased. In cases where burials occur within the habitation area, it is difficult to be certain whether the dead were respected or feared, or both. Patterns in the orientation and form of burials show the existence of funerary customs followed by at least some members of the community. Multiple burials may indicate simultaneous death or the strength of kinship ties. The practice of covering bodies with red ochre prior to burial at Mehrgarh suggests a fertility ritual. The

joint burials of humans and animals at Burzahom reflect a close relationship between people and the animals concerned. Simple versus more elaborate graves can be seen as reflections of differences in funerary customs associated with people of different ranks. Food items among the grave goods suggest a belief in afterlife. Secondary burials suggest multi-stage funerary practices and rituals. The social implications of changes in burial practices at certain sites need to be investigated further. CONCLUSIONS

There is considerable variation in the chronology of the early food-producing societies and in the details of their adaptation to their environment. In c. 7000–3000 BCE, food-producing villages emerged in Baluchistan and the northern fringes of the Vindhyas. The number and geographical spread of such settlements increased in c. 3000–2000 BCE. The beginnings of animal and plant domestication did not lead to the extinction of hunting and gathering. One of the striking features of this period was the co-existence and interaction among neolithic, neolithic–chalcolithic, rural chalcolithic, urban chalcolithic, and hunter-gatherer communities. In the long run, the importance of the advent of food production lay not only in its immediate consequences, but also in the potential it created for future changes. In certain areas, the process of food production and its associated cultural developments eventually led to the emergence of proto-urban settlements, and then full-fledged cities.

www.pearsoned.co.in/upindersingh Further resources

Chapter Four The Harappan Civilization, c. 2600–1900 BCE Chapter outline CIVILIZATION AND URBANIZATION: DEFINITIONS AND IMPLICATIONS RECENT DISCOVERIES AND CHANGING PERSPECTIVES HARAPPAN, INDUS, OR SINDHU–SARASVATI CIVILIZATION? ORIGIN: THE SIGNIFICANCE OF THE EARLY HARAPPAN PHASE THE RELATIONSHIP BETWEEN THE EARLY AND MATURE HARAPPAN PHASES THE GENERAL FEATURES OF MATURE HARAPPAN SETTLEMENTS PROFILES OF SOME HARAPPAN CITIES, TOWNS, AND VILLAGES THE DIVERSITY OF THE HARAPPAN SUBSISTENCE BASE HARAPPAN CRAFTS AND TECHNIQUES NETWORKS OF TRADE THE NATURE AND USES OF WRITING RELIGIOUS AND FUNERARY PRACTICES THE HARAPPAN PEOPLE THE RULING ELITE THE DECLINE OF URBAN LIFE THE SIGNIFICANCE OF THE LATE HARAPPAN PHASE CONCLUSIONS

VIEW OF MOHENJODARO (SINDH, P AKISTAN)

In 1826, Charles Masson, an adventurer who had deserted the East India Company army, stood on the mounds of Harappa, a village in Sahiwal district of Punjab. He was convinced that this must have been the very place where, in the 4th century BCE, the Macedonian invader Alexander had defeated king Porus in battle. A few years later, a traveller named Alexander Burnes visited

Harappa. He thought it was an important site, but was clueless about its precise significance. Many decades later, in the 1850s, Harappa was visited by Alexander Cunningham, a military engineer with the East India Company who was keenly interested in archaeology. He conducted a small excavation and discovered the remains of some structures, but was not impressed. When Cunningham re-visited Harappa in 1872, he came as Director General of the newly established Archaeological Survey of India (ASI). He was dismayed to find the mounds badly disturbed by railway contractors who had been busy extracting free bricks. Cunningham found stone tools and ancient pottery, and also obtained a seal with a bull and some strange writing. He was intrigued, but concluded that since the bull did not have a hump, the seal must be a foreign one. He missed a very important clue.

The officers of the Archaeological Survey of India who explored Harappa and Mohenjodaro in the early 20th century were unenthusiastic about the sites. Pandit Hiranananda Sastri reported that he did not think there was any point in excavating Harappa, and D. R. Bhandarkar’s assessment was that Mohenjodaro could not be more than 250 years old! The sites were eventually excavated. In 1920, Daya Ram Sahni started excavations at Harappa and in 1921, R. D. Banerji started excavating Mohenjodaro. But it took a few more years for the true significance of the discoveries at these sites to be understood. The formal announcement of the discovery of the Indus or Harappan civilization was made in in 1924 by John Marshall, Director General of the Archaeological Survey, almost a century after Charles Masson had wandered over the mounds of Harappa and sensed that there was something significant about the place (see Lahiri, 2005 for the details of this fascinating story). The implications of Marshall’s dramatic announcement were enormous. An important and exciting fragment of India’s past had been uncovered, and the beginnings of civilization in India were pushed back some 2,500 years, to a time roughly contemporaneous with the civilizations of Mesopotamia and Egypt.

JOHN MARSHALL, DIRECTOR G ENERAL, ASI, 1902–28

Civilization and Urbanization: Definitions and Implications

The word ‘urbanization’ means the emergence of cities. ‘Civilization’ has more abstract and grander connotations, but refers to a specific cultural stage generally associated with cities and writing. In a few instances, archaeologists have described neolithic settlements as urban on the basis of size and architecture, even in the absence of writing. This is the case with 8th millennium BCE Jericho in the Jordan valley and the 7th millennium BCE settlement at Çatal Hüyük in Turkey. It has also been pointed out that the Mayan civilization of Mesoamerica and the Mycenaean civilization of Greece did not have true cities, while the Inca civilization of Peru did not have a system of true writing. However, apart from a few such exceptions, cities and writing tend to go together, and ‘urbanization’ and ‘civilization’ are more or less synonymous. One of the earliest attempts to define a city was made by V. Gordon Childe (1950). Childe described the city as the result and symbol of a revolution that marked a new economic stage in the evolution of society. Like the earlier ‘neolithic revolution’, the ‘urban revolution’ was neither sudden nor violent; it was the culmination of centuries of gradual social and economic changes. Childe identified 10 abstract criteria, all supposedly deducible from archaeological data, which distinguished the first cities from the older and contemporary villages.

RAKHALDAS B ANERJI, WHO EXCAVATED MOHENJODARO IN 1921

Childe’s observations proved to be the starting point of an important debate on the diagnostic features of urban societies. Some scholars did not agree with his use of the word ‘revolution’ to describe urbanization, as it suggests sudden, deliberate change. Further, his 10 criteria seem to be a loose assemblage of overlapping features, and are not arranged in any sequence of relative importance. For instance, were sophisticated artistic styles as important as an agricultural surplus or a state structure? Further, all 10 features (e.g., exact and predictive sciences) are not directly deducible from the archaeological data. Another objection is that some features, such as monumental architecture, specialized crafts, and long-distance trade are occasionally found in non-urban contexts as well. However, if we consider the 10 characteristics collectively instead of individually, it has to be conceded that Childe did succeed in identifying the most significant features and implications of city life.

Over the years, there have been three different sorts of trends in defining the city. One is to narrow down the diagnostic features, focusing, for instance, on writing, monumental structures, and a large population. A second trend is to identify more specific criteria such as settlement size, architectural features (e.g., fortifications and the use of stone and brick), and a uniform system of weights and measures. A third trend is towards a more abstract definition, highlighting features such as cultural complexity, homogeneity, and far-reaching political control. The various hypotheses that have been put forward to explain the rise of the world’s first cities are reflective of how different scholars view and understand the unfolding of historical processes. Childe emphasized the importance of technological and subsistence factors such as increasing food surpluses, copper-bronze technology, and the use of wheeled transport, sailboats, and ploughs. Scholars such as Robert McC. Adams emphasized social factors, while Gideon Sjoberg asserted that political factors played the pivotal role in the emergence of cities. An important aspect of McC. Adams’ contribution to our understanding of city life is his highlighting the relationship between cities and their hinterlands (see McC. Adams, 1966 and McC. Adams, 1968). City and village are not two opposite poles, but interdependent and interacting parts of a larger cultural and ecological system. While cities were no doubt ultimately sustained by agricultural surpluses produced in villages, the generation, appropriation, and deployment of agricultural surpluses were neither automatic nor purely economic phenomena and were governed by social and political factors. McC. Adams also highlighted the multiple roles played by cities: They were nodes for the appropriation and redistribution of agricultural surpluses. They provided a permanent base for new social and political institutions that regulated the relationships between specialized producers occupying different econiches. They were centres for the safe storage of surpluses, concentration of wealth, and for expenditure on public building programmes by elite groups. They were centres of learning, artistic creativity, philosophical debate, and the development of religious ideas. KEY CONCEPTS

The 10 characteristics of cities, according to Childe The 1. world’s first cities were larger and more densely populated than villages. While 2. the city population may have included some farmers and herdsmen, it also comprised full-time craftspersons, merchants, transporters, officials, and priests. These groups were supported by the surplus food produced by farmers. Farmers 3. had to hand over their surplus produce as tax or tribute to a ruling elite. Monumental 4. public buildings were hall marks of cities and reflected the concentration of social surplus (i.e., surplus produce and wealth generated in a society) in the hands of the elite. There 5. was a trade-off between the ruling class and the rest of society. Rulers lived off the surplus produced by farmers and in return provided them with peace, security, planning, and organization. The 6. invention of systems of recording—writing and numeral notation— helped meet the needs of administration. The 7. invention of writing led to the development of exact but practically useful sciences such as arithmetic, geometry, and astronomy, and the creation of a calendar. Conceptualized 8. and sophisticated styles of artistic expression made their appearance. Cities 9. implied a significant amount of long-distance trade.

10. They also implied a state organization based on residence in a territory rather than on kinship. The state provided security and materials to specialist craftspersons, enabling them to live a settled rather than an itinerant life.

SOURCE Childe, 1950

Gideon Sjoberg (1964) emphasized the close connection between the history of cities and the rise and fall of empires. He argued that political control was crucial in maintaining the social organization of empires and providing the stability necessary for the development of trade and commerce. He also elaborated on the many facets of the city’s functions and features. The concentration of population in a relatively small space in a city allowed a greater level of protection and security than possible in a village. It also facilitated communication and the exchange of goods and services among specialists. Elite groups tended to be concentrated in the city and usually lived near its centre. The city was hence the place where political decisions were taken and military strategies planned. Apart from being centres of intellectual and commercial activity, since elite groups were usually also patrons of the arts, cities also became centres of cultural and artistic activity.

DAYA RAM SAHNI , WHO EXCAVATED HARAPPA IN THE 1920S

Over the years, various factors such as population growth, long-distance trade, irrigation, and class conflict have been suggested as having played an important role in the emergence of cities. Actually, as is the case with all complex cultural phenomena, a variety of factors—social, political, economic, technological, and ideological—must have been involved, in conjunction with each other, and the details of their interplay could have varied from culture to culture. Since archaeology forms the primary source for reconstructing the emergence of the world’s first cities, there is more direct information on the technological aspect rather than other factors, which can be understood only in very general terms. The emergence of cities has to be viewed as part of a longer history of human settlements, both rural and urban. The story of urbanization is one of increasing cultural complexity, a widening food

resource base, greater technological sophistication, expanding craft production, social stratification, and the emergence of a level of political organization that can be described as a state. Recent Discoveries and Changing Perspectives

Over the eight decades or so since the momentous discoveries at Mohenjodaro and Harappa, information about the Harappan civilization has increased enormously. New sites have been discovered, old sites re-excavated, and there are several new interpretations based on the old and new discoveries. The amount of data and information has been steadily growing and continues to grow. Yet, many aspects of the civilization remain mysterious and subjects of vigorous debate. In the initial years after its discovery, the Mesopotamian links were crucial for dating the Harappan civilization, and some archaeologists tended to compare the two (Shaffer, 1982a). This led to many questionable theories about Harappan origins and the nature of the Harappan economy and polity. In recent decades, scholars have become very conscious of the earlier bias and acknowledge the need to view the Harappan civilization independently rather than through a Mesopotamian lens.

MADHO SARUP VATS, WHO EXCAVATED HARAPPA IN THE 1920S AND 1930S

Another feature of the early decades of Harappan studies was an emphasis on urban settlements, especially Mohenjodaro and Harappa. Apart from being the first sites of the culture to be excavated, these two cities seemed to stand out by virtue of their size and architectural features. However, several other sites are now known to be as large or even larger than them, e.g., Lurewala and Ganweriwala in Cholistan, Rakhigarhi in Haryana, and Dholavira in Gujarat. Scholars have increasingly directed attention to the smaller, less imposing sites, including towns and villages. These include the site of Allahdino (near Karachi), a village settlement that measures only about 5 ha, but which reveals all the main features of the Harappan civilization. Another recently excavated site is Balu in Haryana, a small fortified rural settlement that has yielded a rich variety of plant

remains. Profiles of different kinds of Harappan settlements are now available, and the understanding of the networks that connected cities, towns, and villages is slowly growing.

MAP 4.1 DISTRIBUTION OF MAJOR HARAPPAN SITES

Although Harappan sites share certain common features, there are also significant regional and inter-site differences. These are visible, for instance, in the layout of settlements and in the crops that people grew and consumed. There are also differences in the types, range, and frequency of artefacts. For instance, at Allahdino, the typical black-on-red Harappan pottery formed only 1 per

cent of the total pottery finds. The mud-brick platforms in the southern part of the citadel complex at Kalibangan, which have been interpreted as ‘fire altars’, do not occur at most other sites. There are also differences in the frequency of various funerary practices across sites. For instance, postcremation burials were much more numerous at Harappa than at Mohenjodaro. All this suggests a variety of subsistence strategies, food habits, craft traditions, religious beliefs, cultic practices, and social customs. The nature and function of certain structures have also been re-considered in recent years. For instance, there is good reason to question whether the ‘great granaries’ at Mohenjodaro and Harappa were granaries at all (Fentress, 1984). Less acceptable is Leshnik’s suggestion (1968) that the dockyard at Lothal was not a dockyard but an irrigation reservoir. The re-interpretation of structures has important implications for the understanding of the Harappan social and political systems. For instance, the so-called ‘granaries’ used to be cited to support the theory of a strong, centralized state. Recent excavations at Harappan sites reflect the changes in approaches, goals, and techniques within the discipline of archaeology. A good example are the recent excavations at Harappa, conducted by a joint American and Pakistani team. Compared to earlier excavations at the site, these have been marked by much more careful analysis of the cultural sequence and details of various parts of the residential areas. There has also been greater use of scientific techniques, including the analysis of bone and teeth remains, which provide very specific information about the diet and health of the Harappans. The debates about various aspects of the Harappan civilization reflect both the potential of archaeology as a window into the ancient past and the important role of interpretation in this discipline. There are many different theories about almost every aspect of the Harappan civilization. Not all are equally acceptable; each has to be carefully examined. Conclusions can be reached on certain issues, while in other cases, it is necessary to acknowledge the current limits of our knowledge. Harappan, Indus, or Sindhu–Sarasvati Civilization?

The first sites of this civilization were discovered in the valley of the Indus and its tributaries. Hence it was given the name ‘Indus valley civilization’ or ‘Indus civilization’. Today, the count of Harappan sites has risen to about 1,022, of which 406 are in Pakistan and 616 in India. Of these, only 97 have so far been excavated. The area covered by the Harappan culture zone is huge, ranging between 680,000 to 800,000 sq km. Sites have been found in Afghanistan; in the Punjab, Sindh, Baluchistan, and North-West Frontier Province of Pakistan; in Jammu, Punjab, Haryana, Rajasthan, Gujarat, and western Uttar Pradesh in India. The northernmost site is Manda in Jammu district of Jammu and Kashmir, the southernmost is Malvan in Surat district in southern Gujarat. The western-most site is Sutkagen-dor on the Makran coast of Pakistan, and the easternmost is Alamgirpur in the Saharanpur district of Uttar Pradesh. There is an isolated site at Shortughai in Afghanistan. The vast geographical extent of the civilization should make the objection to the terms ‘Indus’ or ‘Indus valley’ civilization obvious. The terms ‘Indus–Sarasvati’ or ‘Sindhu–Sarasvati’ civilization are also used by some scholars. This is because a large number of sites are located on the banks of the Ghaggar-Hakra river, which is identified by some scholars with the ancient Sarasvati mentioned in the Rig Veda. However, the sort of objection to the terms ‘Indus’ or ‘Indus valley’ civilization can

also be applied to the terms ‘Indus–Saraswati’ or ‘Sindhu–Saraswati’ civilization. Since the civilization was not confined to the valleys of the Indus or Ghaggar-Hakra, the best option is to use the term ‘Harappan ’ civilization. This is based on the archaeological convention of naming a culture after the site where it is first identified. The use of the term Harappan civilization does not imply that all other sites are identical to Harappa or that the culture developed first in this place. In fact, Possehl asserts that it is necessary to break the Harappan monolith into sub-regions, which he calls ‘Domains’ (Possehl, 2003: 6–7). Newspapers and magazines sometimes announce the discovery of new sites of the Harappan civilization. This is done on the basis of a checklist of archaeological features. Pottery is an important marker. The typical Harappan pottery is red, with designs painted on in black, and has a certain range of forms and motifs. Other material traits associated with the civilization include terracotta cakes (pieces of terracotta, usually triangular, sometime round, whose precise function is unclear), a standardized brick size in the 1:2:4 ratio, and certain types of stone and copper artefacts. When the basic set of Harappan material traits are found associated with each other at a site, it is described as a Harappan site. The Harappan culture was actually a long and complex cultural process consisting of at least three phases—the early Harappan, mature Harappan, and late Harappan. The early Harappan phase was the formative, proto-urban phase of the culture. The mature Harappan phase was the urban phase, the full-fledged stage of civilization. The late Harappan phase was the post-urban phase, when the cities declined. Other terminology is also used. For instance, Jim Shaffer (1992) uses the term ‘Indus valley tradition’ for the long series of human adaptations starting from the neolithic–chalcolithic stage to the decline of the Harappan civilization. Within this larger sequence, he uses the term ‘regionalization era’ for the early Harappan phase, ‘integration era’ for the mature Harappan phase, and ‘localization era’ for the late Harappan phase. The early Harappan–mature Harappan transition and the mature Harappan–late Harappan transition are also treated as separate, distinct phases. In this book, the simple and straightforward terminology of early Harappan, mature Harappan, and late Harappan will be used. When the unqualified term Harappan culture/civilization is mentioned, the reference is to the urban phase. Before the advent of radiocarbon dating, this civilization was dated by cross-referencing with the Mesopotamian civilization, with which the Harappans were in contact and whose dates were known. Accordingly, John Marshall suggested that the Harappan civilization flourished between c. 3250 and 2750 BCE. When the Mesopotamian chronology was revised, the dates of the Harappan civilization were revised to c. 2350– 2000/1900 BCE. The advent of radiocarbon dating in the 1950s offered the prospect of a more scientific way of dating the civilization, and the number of sites for which radiocarbon dates are available have gradually increased. The 1986–1996 Harappa excavations have given over 70 new radiocarbon dates, but none from the earliest levels, which are submerged in water. D. P. Agrawal (1982) suggested c. 2300–2000 BCE for the nuclear regions and c. 2000–1700 BCE for the peripheral zones, but this is based on uncalibrated radiocarbon dates. Recent calibrated C-14 dates give a time frame of about 2600–1900 BCE for the urban phase in the core regions of the Indus valley, the GhaggarHakra valley, and Gujarat. This is quite close to the dates arrived at through cross-dating with Mesopotamia. The dates of individual sites vary.

Collating the calibrated radiocarbon dates from various sites gives the following broad chronology for the three phases of the Harappan culture: early Harappan, c. 3200–2600 BCE; mature Harappan, c. 2600–1900 BCE; and late Harappan, c. 1900–1300 BCE. Origin: The Significance of the Early Harappan Phase

Issues of origins are always complex and often contentious. In his report on Mohenjo-daro, John Marshall asserted that the Indus civilization must have had a long antecedent history on the soil of India (see Chakrabarti, 1984 for a summary of the various theories). However, there were others who put forward diffusionist explanations. According to E. J. H. Mackay, a migration of people from Sumer (southern Mesopotamia) may have led to the Harappan civilization; other proponents of the migration theory included D.H. Gordon and S. N. Kramer. Mortimer Wheeler argued for a migration of ideas, not people—the idea of civilization was in the air of West Asia in the 3rd millennium BCE and the founders of the Harappan civilization had a model of civilization before them. The fact that city life emerged in Mesopotamia a few centuries before it appeared in the Egyptian and Harappan contexts does not mean that the latter were derived from the former in a direct or indirect way. There are in fact several striking differences between the Harappan and Mesopotamian civilizations. The Mesopotamians had a completely different script, a much greater use of bronze, different settlement layouts, and a large-scale canal system of the kind that seems absent in the Harappan civilization. If the Harappan civilization cannot be explained as an offshoot or offspring of the Mesopotamian civilization, what is the alternative? The story of its origins can, in fact, be traced to the emergence of settled farming communities in Baluchistan in the 7th millennium BCE. Its more immediate prelude was the cultural phase that used to be known as pre-Harappan, and is now usually referred to as the early Harappan phase. Amalananda Ghosh (1965) was the first archaeologist to identify similarities between a preHarappan culture and the mature Harappan culture. Ghosh focused on the pre-Harappan Sothi culture of Rajasthan. He asserted that there were similarities between Sothi pottery and the pottery of (a) Zhob, Quetta, and other Baluchi sites; (b) pre-Harappan Kalibangan, Kot Diji, and the lowest levels of Harappa and Mohenjodaro; and (c) mature Harappan levels at Kalibangan, and perhaps also at Kot Diji. In view of these similarities, he argued that the Sothi culture should be described as protoHarappan. A limitation of this hypothesis was that it was based exclusively on a comparison of pottery, and did not consider other material traits. And in emphasizing ceramic similarities, Ghosh had ignored the many differences between the Sothi and Harappan cultures. The result was an overemphasis on the Sothi element in the account of the emergence of the Harappan civilization. KEY CONCEPTS

The problems with diffusionist theories

Diffusionist theories were popular among archaeologists and historians in the 19th and early 20th centuries and were invoked to explain developments as diverse as the beginnings of agriculture,

the origins of cities, the distribution of megalithic monuments, and similarities in religious ideas. Diffusion is not a theory but a way of theorizing about cultural change. A diffusionist argument can broadly be described thus: The first thing to do is to figure out in which part of the world the change first occurred. This is identified as the point of origin, from where the change is presented as having diffused or spread to other areas. The process of diffusion is variously described as the result of a migration of people, some other form of contact (e.g., trade, invasion) or a more abstract cultural stimulus. Such theories often rest on a number of questionable assumptions and flawed logic: One 1. of these assumptions is that similar discoveries/inventions/ cultural changes in different parts of the world must be connected to each other. This is not necessarily so. As we have seen in the case of the origins of agriculture, at least three independent centres of early agriculture can be identified. Diffusionist 2. theories often take up superficial resemblances between cultures and ignore the differences. They then hold up the superficial resemblances as very significant and as ‘proof’ of diffusion. These 3. theories appear to offer an explanation, but actually do not explain anything at all. Technologies or cultural transformations do not get transported and transplanted into new areas in a simple or automatic way. There has to be a need and acceptance for them in the recipient culture, and a number of preconditions have to be in place. Mere awareness of a different way of life does not lead to people changing their ways of doing things or living their lives. For example, it was pointed out in the previous chapter that there are several hunting-gathering groups who are aware of agriculture but do not practise it themselves. Urbanization is a very complex process and the mere awareness of cities does not necessarily lead to a transformation of village cultures into urban ones. As we shall see further on, a number of things have to be in place before urbanization can happen.

This criticism of diffusionist theories should not be taken to mean that cultures never influence each other. However, in all instances, while making a case for such influence, it is necessary to: prove 1. that there was some contact between the ‘donor’ and ‘recipient’ cultures before the change appeared in the latter; show 2. that there is indeed a striking and significant degree of similarity in the developments in the two cultures; and demonstrate 3. how and why the new technology/practice was transmitted to and absorbed into the cultural fabric of the recipient culture.

The first comprehensive analysis of the evidence from pre-Harappan sites in the greater Indus valley and north Baluchistan was made by M. R. Mughal (1977). Mughal compared the whole range of evidence (pottery, stone tools, metal artefacts, architecture, etc.) from pre-Harappan and mature Harappan levels, and explored the relationship between the two stages. The pre-Harappan phase showed large fortified settlements, a fairly high level of expertise in specialized crafts such as stone working, metal crafting, and bead making, the use of wheeled transport, and the existence of trade networks. The range of raw materials used by the pre-Harappans was more or less the same as that used in the mature Harappan phase (except for jade, which is absent in the early Harappan context). The two things lacking were large cities and increased levels of craft specialization. Mughal argued

that the ‘pre-Harappan’ phase actually represented the early, formative phase of the Harappan culture and that the term ‘pre-Harappan’ should therefore be replaced by ‘early Harappan’. Early Harappan levels have been identified at a large number of sites, a few of which are discussed below. At some sites, the early Harappan phase represents the first cultural stage, at others it is part of a longer cultural sequence. The dates vary from site to site, but the general range is c. 3200–2600 BCE. The early Harappan phase is extremely important, not merely as a stepping-stone to urbanization, but in its own right as well. At Balakot (on the coastal plain of Sonmiani Bay on the Makran coast), Period II is early Harappan. The pottery was wheel-made and painted, some of it similar to the polychrome ware of Nal. There were microliths, humped bull figurines, a few copper objects, miscellaneous artefacts made of terracotta, shell, and bone, and beads of lapis lazuli, stone, shell, and paste. Remains of barley, vetch, legumes, and ber were found and bones of cattle, sheep, goat, buffalo, hare, deer, and pig were identified. Mention was made in Chapter 3 of the site of Nal in the Khozdar area of Baluchistan. Nal- and Amri-related sites represent the early Harappan phase in the southern part of the Indus valley and Baluchistan.

P AINTED DESIGNS ON EARLY HARAPPAN POTTERY, NAL; KULLI

MAP 4.2 SOME EARLY HARAPPAN SITES

Amri in Sindh lies about 2 km from the right bank of the Indus. The settlement goes back to c. 3500 BCE. Period I at Amri is early Harappan and is further sub-divided into four phases—1A, 1B, 1C, and 1D. Period II represents a transitional phase and Period III is mature Harappan. Within Period I, there was a gradual increase in the refinement and variety of pottery. Mud-brick structures, sometimes supplemented with stone, made their appearance. Artefacts included chert blades, stone

balls, bone tools, and a few fragments of copper and bronze. In Period IC, there were multiple cellular compartments, perhaps used for storing grain or as platforms for buildings. The pottery was dominated by wheel-made wares and showed a great variety of forms and painted designs, mostly geometric. The painting was monochrome or polychrome, using brown, black, and ochre. Kot Diji lies about 160 km north-east of Amri, on the left bank of one of the old flood channels of the Indus. Here, there is an early and mature Harappan level with a burnt deposit in between. Early Harappan Period I was dated from c. 3300 BCE. Fortified with a massive wall made of limestone rubble and mud-brick, the settlement consisted of a citadel complex and a lower residential area. House walls of stone and mud-brick were found in the upper levels. Artefacts included objects of stone, shell, and bone; terracotta figurines (including a bull figurine), bangles, and beads; and a fragment of a bronze bangle. There is a great variety of pottery in Period I, mostly wheel-made and decorated with brownish bands of paint. The distinctive pottery is a short-necked ovoid pot, painted with designs such as the ‘horned deity’, pipal leaves and ‘fish scales’. Artefacts similar to those at Kot Diji Period I have been found at other sites as well, and such levels are known as ‘Kot Dijian’.

F IGURE 4.1 AMRI POTTERY

At Mehrgarh, the excavators noted the occurrence of Kot Diji style vessels, fragments of triangular terracotta cakes, very long flint blades, and fragments of perforated jars, which suggest links with the Indus valley by the end of Period VII. However, these links are not so strong as to constitute true Harappan influence. At nearby Nausharo, there is a clear transition from the early Harappan to a

transitional and then mature Harappan phase. The pottery of Period IC (the later part of the early Harappan levels) at Nausharo was similar to that of Mehrgarh Period VIIC. Jarrige (Jarrige et al., n.d.: 87) suggests that these two phases were contemporaneous and can be dated c. 2600–2550 BCE. There are a number of early Harappan sites in the Dera Jat area in the western Indus plains. At Gumla in the Gomal valley, new pottery styles, including some similar to the Kot Dijian, appeared in Period II. Period III was dominated by Kot Dijian pottery forms and designs including the ‘horned deity’. Period IV at Gumla belonged to the mature Harappan phase. Period I at Rehman Dheri in the Gomal valley is early Harappan and its earliest levels are dated c. 3380–3040 BCE. The settlement was over 20 ha in size. Aerial photographs showed a planned, rectangular settlement with a regular grid of streets and houses, surrounded by a massive wall that belonged to a later phase, contemporary with the mature Harappan. However, it is clear that there was a wall made of mud and mud-brick around the settlement in the early Harappan phase as well. The pottery designs show Kot Dijian elements and some of the pots have graffiti. Artefacts included stone blades, copper and bronze tools, and terracotta figurines. Beads of lapis lazuli and turquoise were found, indicating exchange with Afghanistan and central Asia. Plant remains comprised grains of wheat and barley. Bones of cattle, sheep, and goat were identified.

F IGURE 4.2 KOT DIJIAN POTTERY FROM VARIOUS SITES

Similar discoveries were made at several sites in the Bannu basin. The early Harappan settlement at Lewan may go back to the early 3rd millennium BCE. Apart from a small habitation area,

excavations revealed an area measuring about 450 × 325 m, littered with various kinds of stone tools in different stages of production—microliths (mostly of chert) as well as heavy stone artefacts, including various types of querns, stone balls, long triangular stone axes, ring stones, and pointed hammer stones. Lewan was clearly a factory site where various kinds of stone tools were made. Beads and bead making material were also found in a part of this industrial area. Tarakai Qila gave evidence of wheat, barley, lentils (Lens culinaris), and field pea (Pisum arvense), and there were stone blades with the sheen typical of sickles used for harvesting grain. Bones of cattle, water buffalo, sheep, and goat were found. Period II at Sarai Khola in the northern part of Punjab province of Pakistan is early Harappan. There was a transition within this period from pit dwellings to mud-brick houses. The dominant pottery type was Kot Dijian. Stone artefacts included microliths, celts, and chisels. There were other objects such as terracotta figurines, terracotta and shell bangles, beads made of steatite paste, and one of lapis lazuli. Some copper artefacts, including bangles, pins, rings, and rods, also made their appearance. In the previous chapter, mention was made of recent excavations at Harappa in Pun-jab province of Pakistan, which indicate that the first occupation of the site (Period I) belongs to the Ravi or Hakra phase. The settlement of the early Harappan phase at Harappa (Period II) was over 25 ha in area (Meadow and Kenoyer, 2001). It was divided into two mounds, each with massive mud-brick platforms and fortifications. The layout of the houses and streets suggest elements of planning. Remains of mud-brick walls, hearths, and a small circular kiln were found. Craftspeople used a variety of raw materials to produce a diverse range of items. Pottery included types similar to those found at Kot Diji. Other artefacts included chert blades, a few stone celts, terracotta female figurines and bangles, and beads made of lapis lazuli, carnelian, and steatite. There is evidence of writing (on pottery and seals), inscribed seals, and standardized weights. Certain types of artefacts found in the early Harappan phase—including some pottery types, figurines, triangular terracotta cakes, toys, and bangles—continued into the mature Harappan phase. As mentioned in Chapter 3, the first village settlements in the Cholistan tract of the Hakra plain belong to the Hakra wares phase. The next cultural phase in this area is Kot Dijian, i.e., early Harappan. In fact, the greatest concentration of Kot Dijian sites lies in the Cholistan region. In this phase, there was a dramatic change from a nomadic life to permanent settlement. M. R. Mughal’s study (1997) shows a drop in the number of camp sites from 52.5 per cent (Hakra wares phase) to 7.5 per cent. Many of the settlements had kilns, indicating a sharp increase in specialized craft activities. About 60 per cent of the sites are under 5 ha, and 25 per cent are between 5 and 10 ha. There are a few larger sites, namely Jalwali (22.5 ha) and Gamanwala (27.3 ha).

EARLY HARAPPAN POTTERY: ZANGIAN; SHAHI TUMP

Period I at Kalibangan on the banks of the Ghaggar river is early Harappan. Calibrated radiocarbon dates give a range of c. 2920–2550 BCE. The settlement of Period I was about 4 ha in size and was surrounded by massive mud-brick fortifications. Houses were made of mud and mudbrick, and were built around courtyards. There was a standardization of brick size (3:2:1). Hearths, lime-plastered storage pits, and saddle querns were found in houses. Artefacts included stone blades, terracotta cakes, shell bangles, disc beads made of steatite, carnelian, faience, gold, and silver, and over a hundred copper objects. The pottery of Period I showed great variety. Some of the pots were similar to Kot Dijian pottery. The distinctive pottery was red or pink in colour with designs painted on in black, sometimes also in white. The designs included a moustache-like scroll, plants, fish, and cattle. Some of the graffiti on pottery is similar to the script of the mature Harappan phase. One of the most exciting finds in Period I was made to the south of the site—a ploughed field surface, showing the north–south and east–west furrow marks left by a plough hundreds of years ago. There are a number of early Harappan sites in the Indo-Gangetic divide. At Kunal, Banawali, and Rakhigarhi in Hissar district of Haryana, the early Harappan phase is succeeded by a mature Harappan phase. At Kunal, Period IA belonged to the Hakra wares phase. Period IB showed a continuation of the traits of the earlier phase, but also a large quantity of pottery of the type found at Kalibangan I. There was also the first occurrence of sturdy red beakers and jars of the Harappan type. Period IC was transitional between the early and mature Harappan. The below ground-level houses of the earlier phases made way for ground-level houses made of standardized mud-bricks (in

the 1:2:3 and 1:2:4 size ratios). Six steatite seals and one shell seal bearing geometric patterns were found. Large hoards of jewellery, including two silver tiaras, gold ornaments, and beads made of semiprecious stones such as lapis lazuli and agate, were discovered in some of the houses. At Banawali, the early Harappan phase was marked by mud-brick houses with hearths and plastered storage pits in the courtyards. The pottery was similar to that found at Kalibangan I. Artefacts included stone blades, copper objects, beads of gold and semi-precious stones, and a cubical chert blade. Nearby, along the Ghaggar-Hakra, early Harappan levels have been identified at Siswal and Balu in Haryana and Rohira and Mahorana in Punjab. Rakhigarhi gives evidence of a planned settlement and mud-brick structures in early Harappan Period I. The range of pottery types was similar to that of Kalibangan I. Artefacts included uninscribed seals, pottery with graffiti, terracotta wheels, carts, rattles, and bull figurines, chert blades, weights, a bone point, and a muller. A lot of animal bones were found during the excavations, indicating the importance of animal husbandry. A stacked set of hopscotches was found in an open area behind the structural complex. This suggests the possibility that a game similar to pithu, which is popular among children in India and Pakistan, goes back to early Harappan times!

F IGURE 4.3 PAINTED MOTIFS ON PRE-HARAPPAN POTTERY FROM KALIBANGAN

Bhirrana, a recently excavated site in Fatehabad district of Haryana (Rao et al., 2004–05), has given valuable information on the processes leading to the Harappan civilization. Period IA belongs to the Hakra wares culture, Period IB is early Harappan, Period II early mature Harappan, and Period IIB mature Harappan. The remains of Period IB included vestiges of structures made of mudbricks in the ratio of 1:2:3, including a house complex consisting of six rooms, a central courtyard, and chullahs. There were many different kinds of pottery, including the types known from Kalibangan, as well as the bi-chrome wares, a few sherds of light incised wares, and tan/chocolate wares known from Period IA. Other artefacts included copper arrowheads, rings, and bangles; beads of carnelian, jasper, steatite, shell, and terracotta; terracotta marbles, pendant, bull figurine, rattle,

cake, wheel, and gamesmen (small pieces that may have been used as counters in some sort of ancient board game); plain and segmented terracotta bangles; faience bangles; bone objects; and sandstone sling balls, marbles, and pounders. Excavations at sites such as Padri and Kuntasi in Saurashtra have shown the existence of a welldeveloped early Harappan horizon in Gujarat. The site of Dholavira in the Rann of Kutch has early Harappan levels. The settlement was fortified with an imposing wall made of stone rubble set in mud mortar. Buildings were made of standardized (1:2:4) mud-bricks. Pottery included perforated jars and dish-on-stand, and there was evidence of copper artefacts, stone blades, shell objects, terracotta cakes, and stone beads. The Relationship Between the Early and Mature Harappan Phases

In spite of the undeniable evidence of cultural continuity from the early Harappan to the mature Harappan phase, the ‘outside influence’ factor still sometimes resurfaces in different forms. While acknowledging the indigenous roots of the Harappan civilization, some archaeologists still invoke Sumerian influence. Attempts have been made to connect the pottery traditions of the Harappan tradition with those of Mesopotamia and eastern Iran. Lamberg-Karlovsky (1972) suggests that the emergence of an early urban interaction sphere in c. 3000 BCE in Turkmenia, Seistan, and south Afghanistan had an important role to play in Harappan urbanism. Shereen Ratnagar (1981) suggests that Indus–Mesopotamian trade played an important role in the rise and decline of the Harappan civilization. Such theories are difficult to accept in the absence of substantive evidence. Apart from the fact that some features of the mature Harappan culture were already in place in the early Harappan phase, what is also visible is a gradual transition from a variety of regional traditions towards a level of cultural uniformity cutting across regions, a process that the Allchins call ‘cultural convergence’ (Allchin and Allchin, 1997: 163). Some inferences can also be made about the social and political processes that were underway. Specialized crafts imply specialized craftspersons, trade implies traders, and planned settlements imply planners, executors, and labourers. Seals have been found at Kunal and Nausharo and may have been connected with traders or elite groups. The discovery of hoards of jewellery at Kunal, including a silver piece that has been interpreted as a tiara, suggests a fairly high level of concentration of wealth and may also have political implications. The discovery of symbols similar to Harappan writing at early Harappan levels at Padri in Gujarat, Kalibangan in Rajasthan, Dholavira in Kutch, and Harappa in west Punjab shows that the roots of the Harappan script go back to this phase. Another notable feature is the appearance of the ‘horned deity’ at a number of places. He is painted on a jar found at Kot Diji and on several jars found at early Harappan Rehman Dheri, in contexts dated c. 2800-2600 BCE. At Kalibangan Period I, his figure was incised on one side of a terracotta cake, on the other side of which was a figure with a tied animal. All this suggests that the process of ‘cultural convergence’ was also operating in the religious and symbolic spheres. But how did this convergence come about? What led to the transition from the proto-urban early Harappan phase to full-fledged city life? Was it the result of increased inter-regional contact, or long-distance trade? Trade with Mesopotamia has been suggested as a factor, but the importance of this trade has been exaggerated even in the context of the mature Harappan phase. According to Chakrabarti (1995b: 49–52), the catalyst for the transition may have been an increasing level of craft

specialization, instigated especially by the development of copper metallurgy in Rajasthan. He suggests that another crucial factor for the spread of settlements in the active floodplain of the Indus may have been agricultural growth based on an organized irrigation system, but direct evidence of this is lacking. The answer may lie in the emergence of a new, decisive political leadership, significant changes in social organization, or perhaps a new ideology. Unfortunately, such changes are difficult to deduce from the archaeological data.

F IGURE 4.4 HORNED DEITY ON TERRACOTTA CAKE AND POT, KALIBANGAN, P ERIOD I

There are several other gaps in our understanding of the relationship between the early and mature Harappan phases. The information about the earliest levels at sites such as Mohenjodaro and Harappa is inadequate. There are several mature Harappan sites where there is no early Harappan level, e.g., Lothal, Desalpur, Chanhudaro, Mitathal, Alamgirpur, and Ropar. There are several early Harappan sites in the Potwar plateau which do not have mature Harappan levels. In Cholistan, only three of the many early Harappan sites—Chak 76, Gamanwali, and Sandhanawala Ther—continued to be occupied in the mature Harappan phase. Further, there are no early Harappan sites in the active Indus plain. And at sites where there are both early Harappan and mature Harappan levels, the transition from one to the other is not always smooth. At Kot Diji and Gumla, a burnt deposit between the two suggests a major fire. Evidence of burning was also found at Amri and Nausharo. At Kalibangan, the break in occupation may have been due to an earthquake. The General Features of Mature Harappan Settlements

The fact that the Harappan civilization was urban does not mean that all or even most of its settlements had an urban character. A majority were in fact villages. The cities depended on villages for food and perhaps also labour, and various kinds of goods produced in cities found their way into the villages. As a result of the brisk urban–rural interaction, the typical range of Harappan artefacts reached even small village sites. It is not easy to estimate the exact size of ancient settlements, as they are often spread over many mounds and buried under layers of alluvium. Nevertheless, it is clear that the Harappan sites varied a great deal in size and function, from large cities to small pastoral camps. The largest settlements include Mohenjodaro (over 200 ha), Harappa (over 150 ha), Ganweriwala (over 81.5 ha), Rakhigarhi (over 80 ha), and Dholavira (about 100 ha). Lurewala in Cholistan, with an estimated population of about 35,000, seems to have been as large as Mohenjodaro. Other large sites (about 50 ha) are Nagoor, Tharo Waro Daro, and Lakhueenjo-Daro in Sindh, and Nondowri in Baluchistan. Recently, some very large Harappan sites have been reported in Punjab—Dhalewan (about 150 ha) in Mansa district and Gurni Kalan I (144 ha), Hasanpur II (about 100 ha), Lakhmirwala (225 ha), and Baglian Da Theh (about 100 ha) in Bhatinda district, but details are so far lacking. The second rung of Harappan settlements are moderate-sized sites ranging between 10 and 50 ha, such as Judeirjodaro and Kalibangan. Then, there are the even smaller sites of 5–10 ha, such as Amri, Lothal, Chanhudaro, and Rojdi. The many settlements in the 1–5 ha range include Allahdino, Kot Diji, Rupar, Balakot, Surkotada, Nageshwar, Nausharo, and Ghazi Shah. There are also settlements even smaller than these. The streets and houses of Harappan cities were once thought to be laid on a grid-pattern oriented north–south and east–west. Actually, even Mohenjodaro does not show a perfect grid system. Roads in the Harappan cities were not always absolutely straight and did not always cross one another at right angles. But the settlements were clearly planned. There is no strict correlation between the level of planning and the size of a settlement. For example, the relatively small site of Lothal shows a much higher level of planning than Kalibangan, which is twice its size. The details of the plans differ. Mohenjodaro, Harappa, and Kalibangan have a similar layout, consisting of a raised citadel complex and a lower city. At Lothal and Surkotada, the citadel complex is not separate; it is located within the main settlement. In its most fully developed phase, Dholavira consisted of not two but three parts—the citadel, middle town, and lower town. A major difference between the buildings in large cities and those in smaller towns and villages was in the type and combination of raw materials used. In villages, houses were made mostly of mud-brick, with the additional use of mud and reeds; stone was occasionally used for foundations or drains. Buildings in towns and cities were made of sun-dried and burnt bricks. In the rocky areas of Kutch and Saurashtra, however, there was extensive use of stone. The massive fortification walls with a veneer of dressed stone at Dholavira and the remains of stone pillars in the citadel are very distinctive and are not found at any other Harappan site. The fact that some house walls at Mohenjodaro survive upto a height of 5 m is a tribute to the strength of the bricks and the brick-laying skill of the Harappans. There were various styles of laying bricks, including what is known as the ‘English bond style’. In this, bricks were laid together in a sequence of long side (stretcher) and short side (header), with an alternate arrangement in consecutive rows. This gave the wall maximum load-bearing strength. A striking feature of Harappan structures is the uniformity in the average size of the bricks—7 × 14 × 28 cm for houses and 10 × 20

× 40 cm for city walls. Both these brick sizes have an identical ratio of thickness, width, and length (1:2:4). This ratio first makes its appearance at a few sites in the early Harappan phase, but in the mature Harappan phase, it is found in all the settlements. People lived in houses of different sizes, mostly consisting of rooms arranged around a central courtyard. Doorways and windows generally faced the side lanes and rarely opened onto the main streets. The view from the lane into the courtyard was blocked off by a wall. There are remains of staircases that may have led to the roof or a second storey. The fact that some of the houses at Mohenjodaro were two stories high or more is also suggested by the thickness of their walls. Floors were usually made of hard-packed earth, often re-plastered or covered with sand. The ceilings were probably over 3 m high. Roofs may have been made of wooden beams covered with reeds and packed clay.

MOHENJODARO: WELL FLANKED BY HOUSE WALLS

The doors and windows of houses were made of wood and mats. Clay models of houses show that doors were sometimes carved or painted with simple designs. Windows had shutters (perhaps made of wood or reeds and matting), with latticework grills above and below to allow in light and air. A few pieces of carved alabaster and marble latticework have been found at Harappa and Mohenjodaro; such slabs may have been set into the brickwork. Small houses attached to large ones may have been the quarters of service groups working for wealthy city dwellers. In the larger houses, passages led into inner rooms, and there is evidence of frequent renovation activity. Bathrooms and toilets are facilities people use every day but which most books on ancient history rarely discuss. In the case of the Harappan civilization, there is quite a bit of information on this aspect (Kenoyer, 1998: 59–60). Many houses or groups of houses had separate bathing areas and toilets. Bathing platforms with drains were often located in rooms next to a well. The floor of the bathing area was usually made of tightly fitted bricks, frequently set on edge, to make a carefully sloped watertight surface. A small drain led from here, cut through the house wall, and went out into the street, connecting ultimately with a larger sewage drain. Although some people may have used the area outside the city walls to relieve themselves, toilets have been identified at many sites. They ranged from the simple hole in the ground above a cesspit to more elaborate arrangements. Recent excavations at Harappa have uncovered toilets in almost every house. The commodes were made of big pots sunk into the floor, many of them associated with a

small lota-type jar, no doubt for washing up. Most of the pots had a small hole in the base, through which water could seep into the ground. The waste from the toilets was in some cases discharged though a sloping channel into a jar or drain in the street outside. Some people must have had the job of cleaning the toilets and drains on a regular basis.

MAIN STREET

Well laid-out streets and side lanes associated with an efficient and well-planned drainage system are other notable features of Harappan settlements. Even the smaller towns and villages had impressive drainage systems. The sewage chutes and pipes were separate from drains for collecting rain water. Drains and water chutes from the second storey were often built inside the wall, with an exit opening just above the street drain. At Harappa and Mohenjodaro, terracotta drain pipes directed waste water into open street drains made of baked bricks. These connected into large drains along the main streets, which emptied their contents into the fields outside the city wall. The main drains were covered by corbelled arches made of brick or stone slabs. There were rectangular soakpits for collecting solid waste at regular intervals. These must have been cleaned out regularly, otherwise the drainage system would have become choked and a health hazard. The Harappans made elaborate arrangements for water for drinking and bathing. The emphasis on providing water for bathing, evident at several sites, suggests that they were very particular about personal hygiene. It is possible that frequent bathing also had a religious or ritualistic aspect. The sources of water were rivers, wells, and reservoirs or cisterns. Mohenjodaro is noted for its large number of wells. Harappa had much fewer wells but a depression in the centre of the city may represent a tank or reservoir that served the city’s inhabitants. There are a few wells at Dholavira, which is noted more for its impressive water reservoirs lined with stone. Profiles of Some Harappan Cities, Towns, and Villages

A very small proportion of identified Harappan sites have been excavated. And where excavations have taken place, only sections of the settlements have been exposed (for site details, see, for instance, Kenoyer, 1998; Possehl, 2003; and Lal, 1997).

Mohenjodaro in Sindh lies about 5 km away from the Indus; in protohistoric times, the river may have flowed much closer. The site consists of two mounds, a higher but smaller western mound and a lower but larger eastern mound. There is an extensive area to the east that has not yet been explored. The size of the site has been estimated as about 200 ha. On the basis of the density of houses in the excavated area, Fairservis (1967) suggested that the lower city may have housed about 41,250 people. The western mound at Mohenjodaro (known as the citadel) rises up to 12 m above the plain. The structures here were built on an artificial mud and mud-brick platform, about 400 × 200 m. The mound was circled by a 6 m thick mud-brick retaining wall or platform with projections on the south-west and west, and a tower has been identified on the south-east. It has been suggested that the elevated area at Mohenjodaro does not represent a defensive fortification but part of a civic design to create an elevated symbolic landscape. However, the defensive nature of the walls here and at other cities cannot be ruled out.

F IGURE 4.5 CITADEL AND LOWER TOWN, MOHENJODARO

The buildings on the citadel mound of Mohen jodaro are among the things we associate most closely with the Harappan civilization. In the north are the Great Bath, the so-called ‘granary’, and

‘college of priests’. The Great Bath, an example of the Harappans’ engineering skill, measures about 14.5 × 7 m, with a maximum depth of 2.4m. A wide staircase leads down into the tank from the north and south. The floor and walls of the tank were made water-tight by finely fitted bricks laid edge to edge with gypsum mortar. A thick layer of bitumen was laid along the sides of the tank and probably also below the floor, making this one of the earliest examples of waterproofing in the world. The floor slopes towards the southwest corner, where a small outlet leads to a large corbelled brick drain, which would have taken the water out to the edge of the mound. Remains of brick colonnades were discovered on the eastern, northern, and southern sides of the bath and a similar colonnade must have existed on the western side as well. Two large doors lead into the complex from the south and there were also entrances from the north and east. There are a series of rooms along the eastern edge of the building. One of them has a well that may have supplied water to the tank. Immediately to the north of the Great Bath is a large building consisting of eight small rooms with common bathing platforms.

MOHENJODARO: NARROW LANE BETWEEN HOUSE WALLS

G REAT B ATH

Across the street from the Great Bath are the remains of a large, imposing building (69 × 23.4 m) consisting of several rooms, a 10 m square courtyard, and three verandahs. Two staircases led either to the roof or an upper storey. Because of its size and proximity to the Great Bath, it was tentatively identified as the house of the chief priest or several priests, and was labelled the ‘college of priests’. On the western edge of the citadel mound, at the south-west corner of the Great Bath, raised on a tapered brick platform, is a structure that was originally identified as a hammam or hot-air bath, and later as the ‘great granary’. The 50 × 27 m solid brick foundation was divided into 27 square and rectangular blocks by narrow passageways, 2 running east–west and 8 running north–south. The entire superstructure may have been made of wood. A 4.5 m wide brick staircase led from the southwestern edge of the building to the level of the plain. There was a small bathing platform at the top of the stairs and a brick-lined well at their foot. To the north was a burnt brick platform, identified by Wheeler as a loading dock. As it was excavated without recording the artefacts found in the passageways or the rooms, it is difficult to be sure about its function. But the absence of reports of charred grain or storage containers has led some scholars to question its identification as a granary. In the southern part of the citadel mound, there is a large building (27 × 27 m) that has been labelled an ‘assembly hall’. It is roughly square in shape and is divided into five aisles by rows of rectangular brick piers. The lower town to the east, covering over 80 ha, may also have been surrounded by a fortification wall. It was divided into major blocks by four north–south and east–west streets and numerous smaller streets and alleys. The main streets were about 9 m in width, the rest in the range of 1.5–3 m. The houses varied in size, suggesting differences in wealth and status. In the HR area (the sections of Mohenjodaro are named after the excavators: HR stands for H. Hargreaves, DK for K. N. Dikshit), there were remains of a large building where many seals and fragments of a stone sculpture of a

seated man with a shawl over his left shoulder (similar to the so-called ‘priest-king’ found in the DK area) were found. This building was tentatively interpreted as a temple or the house of an important leader. In the western part of the HR area, there was a double row of 16 houses, each consisting of a single room with a bathroom in front and 1 or 2 smaller rooms in the back. These were tentatively identified as shops or workers’ quarters. A number of shops and workshops associated with copper working, bead making, dyeing, pottery making, and shell working were identified in the lower town. There may have been over 700 wells in the city of Mohenjodaro (Jansen, 1989). This gives a very high average frequency of about one in every third house. The wells were 10–15 m deep and were lined with special wedge-shaped bricks. Deep grooves at the top edges show the spots where the ropes attached to buckets rubbed against them. Most houses or house blocks at Mohenjodaro had at least one private well. Many neighbourhoods had public wells along the main street. We can imagine people meeting here, exchanging news and gossip as they waited to fill their pots with water. Chanhudaro is a 4.7 ha site, about 130 km south of Mohenjodaro. Today, the river flows 20 km to its west; in protohistoric times it may have been closer. This is a single mound site with no fortifications. There are mud-brick platforms with remains of various structures. The traces of at least three streets have been identified. The main one was 5.68 m wide, and had two covered drains made of burnt bricks on both sides. Chanhudaro was clearly an important centre of craft activity. Some of the houses yielded raw material such as carnelian, agate, amethyst, and crystal as well as finished and unfinished beads and drills. More striking was the discovery of a bead factory, with lots of finished and unfinished beads, mostly made of steatite. Seal making, shell working, and the making of stone weights seem to have been other important crafts practised here. The mounds of Harappa cover an extensive area of about 150 ha. The Ravi river flows some 10 km away from the site. The higher citadel mound lies to the west, with a lower but larger lower town to its south-east. South of the citadel mound is a cemetery of the mature Harappan phase. The citadel at Harappa was shaped roughly like a parallelogram, about 415 m north–south and 195 m east–west. It was surrounded by a mud-brick wall with massive towers and gateways, and the structures inside were raised on one or more high platforms. Because of the damaged nature of the mound, clear profiles of the main citadel structures, such as those available for Mohenjodaro, are lacking.

F IGURE 4.6 CITADEL AND ADJACENT AREA, HARAPPA

To the north of the citadel complex, a number of structures were located on a mound (Mound F) surrounded by a mud-brick wall. This seems to represent a northern suburb connected with craft activity. One walled complex had at least 15 units (about 17 × 7 m), each consisting of a courtyard in front and a room at the back, arranged in 2 rows with a lane in between. This has been interpreted as workmen’s quarters. To the north of this complex were at least 18 circular brick platforms, with an average diameter of a little over 3 m, made of bricks set on edge. These may have been threshing platforms for grain. A wooden mortar for pounding grain may have been fitted into their centre, as husked barley and straw were found here. The ‘granary’ was located to the north of these platforms. It consisted of 12 units arranged in 2 rows of 6 rooms, divided by a central passage. Each unit measured 15.2 × 6.1 m, with three sleeper walls with air space in between. There was probably a wooden superstructure supported in places by large columns. As in the case of the Mohenjodaro

‘granary’, no grains were reported from this building. Its interpretation as a granary was mainly based on comparisons with structures found in Rome. The lower walled town of Harappa (Mound E) is currently being excavated. A large open area inside the southern gateway may have been used as a market or as a place where goods coming into the city were inspected. Various workshops where shell, agate, and copper artefacts were made have been identified. Outside the southern gateway, a small mound revealed houses, drains, bathing platforms, and perhaps a well. This may have been a halting or resting spot for travellers or traders. Kalibangan (literally, ‘black bangles’) gets its name from the thick clusters of black bangles lying all over the surface of its mounds. This site lies on the banks of the dry bed of the Ghaggar river, in the Hanumangarh district of Rajasthan. It is fairly small, with a perimeter ranging from 1 to 3 km. There is a smaller western mound (known as KLB-1) and a larger eastern one (known as KLB-2), with an open space in between. KLB-1 has evidence of early and mature Harappan occupation, while KLB-2 represents only a mature Harappan occupation. There is also a smaller, third mound, which only has a large number of fire altars. Both the citadel complex and lower town were fortified.

KALIBANGAN: MAIN STREET

HOUSE WALLS

The mature Harappan settlement on the western mound at Kalibangan was divided into two parts by an inner wall with stairs on either side. The southern sector had no houses, but is noted for a series of mud-brick platforms with a row of seven clay-plastered pits. Nearby were a well and bath pavements. The pits have been interpreted as fire altars, i.e., sacrificial pits in which offerings were made into the fire, and the area seems to have been associated with community rituals. The buildings in the northern part of the citadel mound seem to have been houses where people associated with the rituals performed in the southern sector may have lived. There is a burial ground about 200 m west– south-west of the citadel. Apart from regular extended burials, there were also some circular pits with grave goods (pottery, bronze mirrors, etc.), but no human remains. The lower town was a rough parallelogram in plan, enclosed by a mud-brick wall. Several streets were traced here. Oblong fire altars were found in houses, with a central stele (rectangular piece) around which terracotta cakes, ash, and charcoal were found. While corbelled drains made of bricks have been found on the citadel mound, street drains of the Mohenjodaro type were absent in the lower town at Kalibangan. The sewage from houses was discharged into troughs or large jars embedded in the ground outside. The large number of bangles of terracotta, shell, alabaster, steatite, and faience at the site indicate that bangle making was an important craft. Other interesting artefacts include an ivory comb, a copper buffalo or bull, what appears to be a stone phallic emblem with a base, and a terracotta fragment incised with a horned figure. Banawali in Hissar district (Haryana) is a fortified site measuring about 300 × 500 m, close to the dry bed of the Rangoi river. The site shows evidence of the early, mature, and late Harappan phases. Period II represents the mature Harappan culture. A wall divided the fortified area into two sections —a higher citadel area and a lower town. The citadel was semi-elliptical in plan and had its own mud-brick fortifications, surrounded by a moat. A few streets and structures were identified inside. A ramp led from the citadel into the lower town. The mud-brick houses had raised platforms (chabutaras) outside. Baked bricks were used only for wells, bathing pavements, and drains. Excavations revealed a multi-roomed house, where archaeologists identified a kitchen and a toilet with a jar that seemed to have functioned as a washbasin. Since many seals and weights were found in this house, it may have belonged to a wealthy merchant. There was another big house with a large number of beads of gold, lapis lazuli, and carnelian, tiny weights, and a ‘touchstone’ showing steaks of gold. This must have been a jeweller’s house. Interestingly, seals were only found in the lower town, not in the citadel complex. Lots of stone weights in small denominations were found at the site, as was a terracotta model of a plough. Several houses at Banawali gave evidence of fire altars. In

one place, these altars were associated with an apsidal structure which may have had some sort of ritualistic function.

B ANAWALI: EASTERN GATE

CROSS-SECTION OF DEFENCE WALL

APSIDAL STRUCTURE

Five mounds have been identified at Rakhigarhi (Hissar district, Haryana). The citadel mound, surrounded by a mud-brick fortification wall, had platforms, a brick well, fire altars, some streets, and drains of various sizes. A lapidary workshop was identified, with remains of about 3,000 unfinished beads and roughly cut pieces of stone, mostly carnelian, chalcedony, agate, and jasper; bead polishers for smoothening the beads; and a hearth for heating the stones. In another part of the site, bones, antlers, ivory pieces, and finished and unfinished bone points, combs, needles, and engravers gave clear evidence of bone and ivory working. A cemetery revealed eight burials consisting mostly of brick-lined pits; in one case there was a wooden coffin. At Bhirrana in Haryana, Period IIA has been described as early mature Harappan and Period IIB as mature Harappan. The mature Harappan settlement was surrounded by a massive fortification wall made of mud-brick. Three multi-roomed house complexes were exposed. One of them, in the central part of the mound, consisted of four rooms. Two house complexes, separated from each other by a lane, were exposed in the eastern part of the mound. One of these consisted of 10 rooms with a verandah and a courtyard; terracotta cakes mixed with ash and clay were found on the floors. Yet another house complex in the north-western part of the mound consisted of six rooms, a kitchen, a central courtyard, three additional courtyards, and an open verandah. The floors were paved with mud-brick, and the brick walls were plastered with mud. A circular tandoor and chullah were found in one of the courtyards, and another chullah was discovered in the kitchen. Charred bones and the skull of a bovine animal were found next to one of the chullahs. A 4.80 m wide street ran north– south along the fortification wall. Three lanes were also identified. The artefacts included a fragment of a thick, sturdy red ware with an incised female figure, whose pose is reminiscent of that of the bronze Mohenjo-daro ‘dancing girl’.

WELL AND DRAINS, LOTHAL

Lothal is located between the Sabarmati river and its tributary, the Bhogavo, in Saurashtra in Gujarat. The sea is now about 16–19 km away, but at one time, boats from the Gulf of Cambay could

have sailed right up to the place. It was a modest-sized settlement (280 × 225 m), roughly rectangular in plan, surrounded by a wall which was initially made of mud and later of mud- and burnt bricks, with the entrance on the south. There was a burial ground in the north-west, outside the enclosing walls. The citadel (called the ‘Acropolis’ by the excavator S. R. Rao) was roughly trapezoidal in plan and consisted of an area elevated on a mud-brick platform in the southern part of the site. Remains of residential buildings, streets, lanes, bathing pavements, and drains were traced here. To the south of the residential area was a complex identified as a warehouse, where goods may have been packed and stored. Sixty-five terracotta sealings with impressions of reed, woven fibre, matting, and twisted cords on one side and impressions of Harappan seals on the other were found here. Some of the houses in the main residential area were quite large, with four to six rooms, bathrooms, a large courtyard, and verandah. A few had fire altars—small pits with terracotta cakes or round lumps of clay and ash. The streets were paved with mud-brick, with a layer of gravel on top. Houses belonging to artisans such as coppersmiths, bead makers, etc. were identified on the basis of the occurrence of kilns, raw materials, and finished and unfinished artefacts. One of the streets was identified as a ‘bazaar street’, the rooms lining it interpreted as shops.

LOTHAL DOCKYARD

The most distinctive feature of Lothal is the dockyard, which lies on the eastern edge of the site. This is a roughly trapezoidal basin, enclosed by walls of burnt bricks. The eastern and western walls measured 212 m and 215 m respectively in length, while those on the north and south measured 37 m and 35 m. The dockyard had provisions for maintaining a regular level of water by means of a sluice gate and a spill channel. A mud-brick platform along the western embankment may have been the wharf where goods were loaded and unloaded. An alternative interpretation of this structure as a water reservoir is not convincing. Dholavira is located on Kadir island in the Rann of Kutch in Gujarat. In protohistoric times, water levels in the Rann may have been higher than they are today, allowing boats to sail from the coast

right up to the site. The architecture of Dholavira shows a large-scale use of sandstone, combined in places with mud-brick—a feature of the Harappan sites of Gujarat. The layout of this settlement is unlike that of any other Harappan site. It is surrounded by an outer fortification wall made of mudbrick with a veneer of stone blocks on the outer face, with imposing bastions and two major gateways in the middle of the northern and southern walls. Within the outer walls, at least three different sections were identified. There was a small ‘castle’ area, a ‘bailey’ area to its west, and a larger ‘middle town’ to the north, all with their own enclosing walls. A lower town lay to the east. An interesting feature is a large open area (called the ‘stadium’) between the castle–bailey and the middle town, which may have been used for special ceremonial occasions. There was also substantial evidence of habitation outside the fortification wall, which may represent a suburb of the city. The site seems to be looking out towards the sea and it must have been an important stopping point on busy maritime trade routes.

DHOLAVIRA: TANK

NORTHERN GATE

The fortified acropolis covered an area of 300 × 300 m, with gateways in the centre of its four walls. Remains of limestone pillar bases and pillar fragments with a highly polished surface were found in the eastern gateway. This discovery has taken the history of monumental stone sculpture/architecture in the subcontinent back from the 4th century BCE (the Maurya period) to the 3rd millennium BCE. In one of the side rooms of the northern gateway of Dholavira lay what seems to be a fallen signboard. An inscription had been made with white gypsum paste inlaid into a wooden board. The wooden board had fallen flat on its face, and although the wood decayed, the gypsum was found intact. The symbols, each measuring about 37 × 25–27 cm, perhaps announced the name of the city or the title of its ruler. The acropolis had a large well, an elaborate drainage system, and large buildings which may have had administrative or ritualistic functions.

F IGURE 4.7 P LAN OF DHOLAVIRA

The middle town of Dholavira was surrounded by a 360 × 250 m wall with four gateways. The lower town gave evidence of houses and areas where various types of craft activities such as bead making, shell working, and pottery making were carried out. Outside the city walls, there was evidence of additional habitation and burials. The cemetery area revealed rectangular pit burials lined with blocks of stone, but there were no skeletal remains. These may have been memorials to the dead. The city had an impressive and unique water harvesting and management system. It can be noted that this area receives less than 160 cm of rain every year and is very prone to droughts. The site is flanked by two streams—the Manhar and Mandsar. Dams were built across these to channelize their water into reservoirs. Several large, deep water cisterns and reservoirs (at least 16) located in the citadel and lower town preserved precious stores of rain water.

DHOLAVIRA CITADEL: EASTERN GATE WITH PILLAR FRAGMENTS

WELL

MASSIVE DRAIN

Allahdino is a small (1.4 ha) unfortified village site of the Harappan civilization, about 40 km east of Karachi. Houses made of mud-brick, often resting on stone foundations, were laid out in a west–

south-west to east–north-east orientation. A large multi-roomed building on a large mud-brick platform in the north-eastern part of the excavated area seems to have had some special significance. Another building was associated with three wells. The wells at Allahdino had very small diameters, and their mouths ranged from 60 cm to 90 cm. This may have been to enable the ground water to rise higher due to hydraulic pressure. It has been suggested that well water may have been used to irrigate the nearby fields. The artefacts found at Allahdino included a large number of copper items, seals, terracotta toy carts, and triangular terracotta cakes. The most spectacular discovery was a small terracotta jar containing a profusion of gold, silver, bronze, agate, and carnelian ornaments. These included a massive belt or necklace consisting of 36 long carnelian beads and bronze spacer beads and a multistrand necklace of silver beads. The discovery of ornaments of precious metals and stone at a village site shows that at least some of the inhabitants of this Harappan village were very rich.

www.pearsoned.co.in/upindersingh PHOTOGRAPHS OF HARAPPAN SITES AND ARTEFACTS

The Diversity of the Harappan Subsistence Base

The Harappan civilization covered an enormous area within which there was great ecological variety—alluvial plains, mountains, plateaux, and sea-coasts. The resource potential of this area was rich enough to generate the food surpluses that are an important aspect of urbanization. The diversity of the subsistence base may also have been an important sustaining factor—if one food resource failed, people could turn to others. Agriculture was the mainstay, supplemented by animal husbandry and hunting. Riverine and marine food resources were tapped, where available. The sources of information on the subsistence patterns of the Harappans consist of plant remains, animal bones, artefacts, motifs on seals and pottery, and analogies with modern practices. Subsistence is closely related to environment, and the nature of the Harappan environment is the subject of continuing debate. Archaeologists such as Mortimer Wheeler and Stuart Piggott suggested a wetter climate in Harappan times on the basis of the following arguments: (a) the large number of burnt bricks found at Harappan sites would have required large quantities of fuel, which would only have been possible with a heavy forest cover, supported by heavier rainfall; (b) the gabarbands (embankments) constructed in the Baluchistan area suggest heavier rain; (c) the depiction of animals such as the tiger, elephant, and rhinoceros on seals indicates a forest and grassland vegetation that could only have been supported by heavier rainfall; (d) the elaborate drainage system of the cities was geared towards carrying off rain water. The first and last points can be refuted most easily. It is not easy to estimate just how much wood (and forest) would have been required to make the burnt bricks, and the Harappan drains were largely part of a system of sewage disposal. Many scholars hold that climatic conditions in the greater Indus valley have remained more or less constant since Harappan times. However, some studies suggest otherwise. Plant palynologist Gurdip

Singh (1971) analysed pollen from the three salt lakes of Sambhar, Didwana, and Lunkaransar, and the freshwater Pushkar lake, and constructed a profile of rainfall in this part of Rajasthan from c. 8000 BCE to 1500 BCE. He concluded that there was an increase in rainfall in c. 3000 BCE and a decrease in 1800 BCE. However, a recent study of the Lunkaransar lake (Enzel et al., 1999) suggests that it had dried up by 3500 BCE and that the climate had become drier long before the emergence of the Harappan civilization. The issue of the nature of climatic conditions in Harappan times thus remains unresolved. Given the area covered by the civilization, naturally there were regional variations in the plants grown by farmers. Wheat has been found at Mohenjodaro and Harappa; barley at Mohenjodaro, Harappa, and Kalibangan; and sesamum at Harappa. Harappa has also given evidence of watermelon seeds, peas, and dates. Rice occurs at Harappa, Kalibangan, Lothal, and Rangpur. Millets have been identified at Harappa, Surkotada, and Shortughai. Grapes were known, so was henna (mehendi). Cotton may also have been grown. Detailed evidence of the plant economy of the early and mature Harappan phase is available from Balu (in Haryana) (Saraswat and Pokharia, 2001–02). The crop remains identified here included various types of barley, wheat, rice, horse gram, green gram, chickpea, field pea, grass pea, sesamum, melon, watermelon, date, grapes, and the earliest evidence of garlic. Apart from the wide range of cereals, pulses, vegetables, and fruits grown by the Harappans, another striking point is the similarity of the past and present plant economies in the various regions.

TERRACOTTA PLOUGH FOUND AT B ANAWALI

Modern cropping practices provide some clues to protohistoric patterns. Today, in Sindh, rainfall levels are low, but the Indus brings down flood waters and silt. The fertile land requires no deep ploughing, irrigation or manuring. Sesamum and cotton were probably sown in June/July and reaped in September/October, as kharif (summer) crops. Crops such as wheat and barley would have been sown in November and reaped in March/April as rabi (winter) crops. In Gujarat, rice is a kharif crop, and it must have been so in Harappan times as well. Reference has already been made to the discovery of a ploughed field at early Harappan levels at Kalibangan. The continuing use of the plough into the mature Harappan phase can be inferred. Terracotta models of ploughs at Bahawalpur and Banawali give further evidence of the use of this implement. The fact that no actual ploughs have survived is no doubt because they were made of wood. Farmers must have built bunds (embankments) of mud or stone to divert river water, as they do today in areas like Baluchistan. Irrigation canals have been found at Shortughai. Fairservis suggested that a well and associated drains at Allahdino may represent an irrigation system, but the evidence is far from conclusive. Similarly, Leshnik’s hypothesis that the dockyard at Lothal is actually an

irrigation reservoir is not convincing. Even if the Harappans did dig canals in the alluvial plains, it would be very difficult to identify them. However, H. P. Francfort (1992) has identified remains of a small-scale canal network in the Haryana area, and some of the ancient canals traced in the GhaggarHakra plain may belong to the Harappan phase. Bones of wild animals have been found at Harappan sites. These include many varieties of deer, pig, boar, sheep, goat, ass (?), and pig. Bones of tortoise and fish have also been found. Rhinoceros bones occur only at Amri, although this animal is depicted on numerous seals and in terracotta figurines. Elephant and camel bones occur in very small quantities, although the elephant appears on seals. Tigers are represented often in figurines, leopards more rarely. Rabbits, peacocks, pigeons, ducks, monkeys, and wild fowl are represented in figurines and paintings on pottery. The Harappans exploited riverine and marine resources where these were available. At coastal sites in Gujarat, molluscs provided an important protein-rich element in people’s diet. The discovery of marine catfish bones at Harappa suggests that coastal communities may have traded in dried fish in inland cities. Harappan sites have also yielded remains of domesticated animals such as humped and humpless cattle, buffalo, sheep, and goat. Cattle and buffaloes were the most important domesticated animals. They would have been used for meat, milk, and also as draught animals. Goats and sheep could have been used for meat, wool, milk, and as pack animals (they are still used to carry loads of salt and grain in some of the Himalayan stretches). Dog figurines suggest the domestication of this animal. The issue of the horse is controversial and hinges on the stratigraphic context in which the remains have been found and the identification of the species they belong to. For instance, it is not easy to ascertain whether the bones in question belong to the half-ass (Equus hemionus khur) or domesticated horse (Equus caballus). Horse remains have been reported at Harappa, Lothal, Surkotada, Kuntasi, and Kalibangan, and at superficial levels at Mohenjodaro. Sàndor Bökönyi (1997) examined the equid bone samples from Surkotada and concluded that at least six of them probably belonged to the true horse. His conclusions were challenged by Meadow and Patel (1997). Brigadier Ross (1946) reported horse teeth at pre-Harappan levels at Rana Ghundai, but this identification was questioned by Zeuner (1963). While horse bones may not be completely absent at Harappan sites, they are not prolific either. NEW DIRECTIONS IN RESEARCH

Animal bones at Shikarpur

Shikarpur is a Harappan site in Kutch district in Gujarat, excavated by the Gujarat State Department of Archaeology in 1987–90. The excavation was a small one. It revealed an over 3 m thick deposit, of which the lower layers (layers 10–19) represent an early Harappan phase and the upper layers (layers 1–9) the mature Harappan phase. The animal remains found at the site were sent to the Archaeozoology Laboratory at Deccan College, Pune. The preliminary results of the detailed investigations by P. K. Thomas, P. P. Joglekar, Arati Deshpande-Mukherjee, and S. J. Pawankar have given important information about the subsistence patterns of the Harappans in Gujarat:

A total of 15,483 pieces of bone were unearthed in the excavations. It was possible to identify 53.46 per cent of them, i.e., 8,267 fragments. There were cut marks and signs of charring on some of the bones, indicating slaughtering and cooking. The faunal assemblage consisted of 47 species—23 mammals, 3 birds, 2 reptiles, 5 fish, 13 molluscs, and 1 crustacea. The wild animals included wild buffalo, nilgai, chowsingha, blackbuck, gazelle, various kinds of deer, wild pig, wild ass, jackal, hare, and rhinoceros. The domesticated animals included cattle, buffalo, sheep/goat, horse, pig, and dog. The bones of domesticated animals comprised over 85 per cent of the total faunal assemblage in both the early and mature Harappan phases. Cattle bones were most numerous. In the early Harappan phase, 77.48 per cent of the bones were of cattle, while in the mature Harappan phase, their percentage was 77.84 per cent. Sheep/goat bones (it is difficult to distinguish the two) amounted to 11.26 per cent of the early Harappan phase, and were reduced to 4.63 per cent in the mature Harappan phase. Buffalo bones were 4.28 per cent and 4.61 per cent in the early and mature Harappan phases respectively. Dog bones were only found in the mature Harappan phase, and that too in very small quantities (0.116 per cent). Very few horse bones were found (0.13 per cent), and these occur only in the mature Harappan phase. The evidence shows that the consumption of meat of domesticated animals was an important part of the diet of the people of Shikarpur. The contribution of wild and aquatic animals varied considerably in different layers. The analysis of bones and teeth showed that domesticated animals were killed at different ages. Most of the cattle and buffaloes lived up to the age of maturity—about 3 years—and were killed at various ages up to the age of about 8 years. The fact that some were older than 8 years suggests that they were also valued for secondary products and used for draught purposes. Sheep/ goats were killed at relatively younger ages—between 6 months to their respective ages of maturity, suggesting they were primarily reared for meat. Towards the end of the mature Harappan phase at Shikarpur, there seems to have been an increase in the exploitation of wild animals. It is not clear whether this was the result of a decline in agricultural production, failure of rains, population pressure, or a combination of several such factors. SOURCE Thomas et al., 1995

Harappan Crafts and Techniques

Earlier writings tended to contrast the plainness of Harappan artefacts with the opulence of their Egyptian and Mesopotamian counterparts. Nowadays, the technological sophistication and beauty of some of the Harappan artefacts are recognized. There is a great variety of standardized, massproduced craft items at Harappan sites. The artefacts are far greater in quantity and range, and show greater technical finesse than those found in earlier cultural phases. While some sites specialized in

the production of a single or a few items, others such as Harappa manufactured a wide range of goods. Craft activity was often localized in a certain part of the settlement. Ceramics include all items involving the heating of clay such as bricks, terracotta, and faience. The Harappan pottery reflects efficient mass-production. Pottery kilns were found at Mohenjodaro, Harappa, Nausharo, and Chanhudaro. The pots were fired in funnel-shaped up-draft closed kilns, although open-firing kilns may also have been used. There is a great variety of pottery, including black-on-red, grey, buff, and black-and-red wares. Most pots were wheel turned. Both fine and coarse fabrics occur and their thickness varies. The typical Harappan pottery is a fine, sturdy, wheel-made ware with a bright red slip, decorated with painted black designs. Polychrome painting is rare. The red colour for the slip was made from red ochre (iron oxide, known as geru), while black was made by combining dark reddish-brown iron oxide with black manganese. Distinctive shapes include the dish-on-stand, vase with s-profile, small vessel with knobbed decoration, large slender-footed bowl, cylindrical perforated jar, and goblet with pointed foot. The decorative patterns range from simple horizontal lines to geometric patterns and pictorial motifs. Some of the designs such as fish scales, pipal leaves, and intersecting circles have their roots in the early Harappan phase. Human figures are rare and crude. At the earliest levels of Mohenjodaro, a burnished grey ware with a dark purplish slip and vitreous glaze may represent one of the earliest examples of glazing in the world. Although there is a certain level of uniformity in pottery styles and techniques across the Harappan culture zone, there are also differences between regions.

F IGURE 4.8 HARAPPAN POTTERY

Inferences can be made about the functions of some of the Harappan pots. The large jars may have been used to store grain or water. The more elaborately painted pots may have had a ceremonial use or may have belonged to rich people. Small vessels may have been used as glasses to drink water or other beverages. The function of the perforated jars is not clear. One suggestion is that they may have been wrapped in cloth and used for brewing fermented alcoholic beverages. Another possibility is that they may have had a ceremonial or ritualistic use. Shallow bowls probably held cooked food; flattish dishes were used as plates. Cooking pots of various sizes have been found. Most of them

have a red- or black-slipped rim and a rounded bottom; the lower part of the pot is often strengthened by a thick slurry or clay mixed with ground pottery or chaff. The rims of the cooking pots are strong and project outwards to help pick them up or move them around. Some of the forms and features of the pots used by the Harappans can be seen in traditional kitchens even today. Apart from ceramic vessels, the Harappans also made and used metal ones.

MINIATURE PERFORATED POT; BEAKER; POT WITH POINTED BASE (WITH SEAL IMPRESSION); RING STAND

Harappan sites have yielded a profusion of terracottas. There are figurines of animals such as bulls, buffaloes, monkeys, and dogs. There are toy carts with solid wheels. Human figurines include male figurines and more numerous female figurines of various types. The Harappan craftspersons also made terracotta bangles. Terracotta masks have been found at Mohenjodaro and Harappa. Faience is a paste made out of crushed quartz and coloured with various minerals. The Harappans made faience bangles, rings, pendants, miniature vessels, and figurines (including those of monkeys and squirrels). Another distinctive Harappan craft was the making of hard, high-fired bangles known as stone ware bangles. These were highly burnished red or grey-black, with a standard inner diameter of 5.5–6 cm, and usually had tiny letters written on them.

P OTTERY DESIGNS

TERRACOTTAS: HUMAN AND ANIMAL FIGURINES; MASK; CIRCULAR AND TRIANGULAR CAKES

Stone work was another important craft. Reference was made earlier to the stone masonry and fine polished pillars at Dholavira. More visible at all Harappan sites were the mass-produced chert blades made by the crested guided ridge technique. Some of these may have been used as knives for domestic use, others as sickles. Harappan stone quarries have been identified in the Rohri hills of Sindh. Some of the stone blades may have been obtained from contemporary hunter-gatherer communities. The fact that stone flakes and cores occur in many houses at Mohenjodaro suggests that at least some of the tools were made by people in their homes. The Harappan civilization is marked by a large number of copper objects. Apart from making artefacts out of pure copper, Harappan craftspersons alloyed copper with arsenic, tin, or nickel. Copper and bronze artefacts included vessels, spears, knives, short swords, arrowheads, axes, fishhooks, needles, mirrors, rings, and bangles. The axes were flat, without a shaft hole, and were probably hafted in a split and bound handle. The number of pure copper artefacts was far greater than alloyed bronze ones. Usually, tools like knives, axes, and chisels, which needed hardened edges, were alloyed. Alloys increased over time—for instance, at Mohenjodaro, bronze tools increased from 6 per cent to 23 per cent from the lower to the higher levels. The small proportion of alloyed objects compared to those of pure copper may suggest cultural preference rather than technological backwardness. Sixteen copper furnaces were found at Harappa, and copper workshops were found at Lothal. A large amount of copper oxide was discovered in a brick-lined pit at Mohenjodaro. That metal objects were considered precious is clear from the fact that they were buried in hoards for safekeeping by their owners. One hoard found at Harappa consisted of a large cooking pot with a bronze cover. Inside were several types of copper tools and weapons, including various types of axes, daggers, spearheads, arrowheads, chisels, and a bowl. Some of the objects were unused, others used and worn. Beautifully worked gold and silver jewellery including necklaces, bracelets, brooches, pendants, and earrings have been found at Harappan sites. A hoard of jewellery made of gold, silver, and semi-precious stones was found at the small village site of Allahdino. The Harappans used silver to emboss conch shells and to make vessels. Lead was used to make plumb bobs and in copper casting. It may be noted that two metal objects found at Lothal contain 39.1 per cent and 66.1 per cent iron. The latter can be called an iron object. What this suggests is that the Harappans (at least those of Gujarat) may have had some familiarity with iron smelting. Seal making was another important Harappan craft. Most of the seals are square or rectangular. The average size of the square seals is about 2.54 cm, but there are larger ones, a little over 6.35 cm. Some have a perforated boss at the back for handling and suspension. A few cylindrical and round seals have also been found. Most of the seals are made of steatite, but there are a few silver, faience, and calcite ones as well. Two fine silver seals with the unicorn motif were discovered at Mohenjodaro, and some copper and soapstone ones were found at Lothal. To make the stone seals, the stone was sawed and shaped with knives, and then carved, using fine chisels and drills. The seal was coated with an alkali and heated, giving it a white lustrous surface. The carving is in intaglio— i.e., it is a sunken engraving, with the impression appearing in relief. Motifs include the elephant, tiger, antelope, crocodile, hare, humped bull, buffalo, rhinoceros, and the one-horned mythical

animal referred to as a unicorn. There is often a small feeding trough or stand in front of the animal. There are also composite animals, human figures, and plants. Most of the seals have a short inscription. Some rectangular seals have writing, but no motif.

CHERT BLADES; STONE GAMESMEN

COPPER ARROWHEAD AND CELT

STONE SEALING; SEAL

Bead making was a craft known in earlier cultures, but in the Harappan civilization new materials, styles, and techniques came into vogue. A new type of cylindrical stone drill was devised and used to perforate beads of semi-precious stones. Such drills have been found at sites such as Mohenjodaro, Harappa, Chanhudaro, and Dholavira. The Harappan craftspeople made beads out of steatite, agate, carnelian, lapis lazuli, shell, terracotta, gold, silver, and copper. The Harappan long barrel cylinder beads made out of carnelian were so beautiful and valued that they found their way into royal burials in Mesopotamia. Tiny micro-beads were made of steatite paste and hardened by heating. Beads were also made of faience. FURTHER DISCUSSION

Sculpture in stone and metal

Apart from utilitarian items made of stone and metal, a few pieces of stone and metal sculpture have been found at Harappan sites. Most of them are small, but they display fine artistic skills and sensibilities. They include the stone bust (17.78 cm high) of a male figure found at Mohenjodaro, which has been labelled the ‘priest-king’. Two fine stone torsos of a male figure (about 10 cm high) were found at Harappa, a seated stone ibex or ram (49 × 27 × 21 cm) at Mohenjodaro, and a stone lizard at Dholavira. The only large piece of sculpture is that of a broken, seated male figure from Dholavira.

Two bronze female figurines were found at Mohenjodaro. One of them has become famous as the ‘dancing girl’. This figurine was found in a small house in the southwestern quarter of the city (in the HR area) during the 1926–27 excavations. The figure is 10.8 cm high and was made by the lost-wax method. The lost-wax method involves first making a wax model and then covering it with a clay coating, leaving some holes as passageways. When the clay-covered moulds are heated in ovens, the wax melts out. Molten bronze is then poured in, and takes the place of the wax. When the mould has cooled, the outer clay envelope is chipped off and the craftsperson can then put the finishing touches to the solid bronze statue. This technique is still used in certain parts of India. But to get back to the ‘dancing girl’: She represents a very thin woman standing with her right hand on the back of her hip and left hand resting on her left thigh, just above the knee. She may have once held some object in this hand. She is naked. She wears a necklace and has 24–25 of bangles on her left arm and just 4 on her right arm. Her arms are unnaturally long. Her head is tilted back, and she has a defiant, nonchalant air about her. Her hair is swept back in a low, loose bun at the nape of her neck. John Marshall named her the ‘dancing girl’ because he thought she had the air of a semi-impudent ‘nautch girl’, hand on hip, beating time to the music with her feet. The name has stuck. But the ‘dancing girl’ may not have been dancing at all, and even if she was, she may not represent a professional dancer.

THE ‘DANCING GIRL’

Bead making factories with tools, furnaces, and beads in various stages of preparation have been found at Chanhudaro and Lothal. At Bagasra in Gujarat, there is evidence of the production of artefacts of shell, faience, and beads of semi-precious stones (agate, carnelian, amazonite, lapis lazuli, and steatite). Clay-lined silos, varying from 0.30 to 1 m in diameter and 0.15 to 0.30 m in depth, were used to store semi-precious stones. The bead-making tradition in Gujarat today gives us clues on how the Harappan craftspeople may have made their beads. Beads, bracelets, and decorative inlay work of shell show the existence of craftspersons skilled in shell working. Bangles were often made from conch shell. Chanhudaro and Balakot were important centres of shell work. Further evidence of site specialization comes from Gujarat. An intensive surface survey and excavations at Nageshwar (in Jamnagar district) have shown that this site was exclusively devoted to shell-working and specialized in making bangles. Evidence of shell working also comes from Kuntasi, Dholavira, Rangpur, Lothal, Nagwada, and Bagasra. This craft was clearly very important in the Gujarat region of the Harappan culture zone. Bone working was another specialized craft. Beads, awls, and pins were made out of bone. There are a few examples of ivory carving in the form of combs, carved cylinders, small sticks, pins, gamesmen, and a carved plaque.

SHELL LADLE, LOTHAL

JEWELLERY: NECKLACES OF CARNELIAN BEADS, GOLD; BANGLES OF TERRACOTTA, COPPER, STONEWARE, LAPIS LAZULI BEADS; GOLD SPIRAL PIN; GOLD AND TERRACOTTA BEADS

It can be inferred from the available evidence that the Harappans made cotton and woollen textiles. The terracotta figurines wearing clothes (shawls, skirts, etc.) reflect the kinds of clothes people wore. Mesopotamian texts mention cotton as one of the imports from Meluhha (an area which included the Indus valley). Traces of cotton cloth were found at Mohenjodaro, preserved over the centuries due to their being in contact with a corroding silver jar. Several examples of cotton thread and cloth were identified on copper tools. At Harappa, cotton threads were found wrapped around the handle of a small copper mirror in a burial and also around the handle of a curved copper razor. Recent excavations at Harappa have given evidence of woven textile impressions on the inside of faience vessels. The uniform thickness and uniformity of the weave suggest the use of spinning wheels. Various kinds of spindle whorls for spinning thread have been found at Harappan sites.

Weaving may have been a cottage industry practised in villages, and also to some extent in the cities. Impressions on clay floors and fired clay lumps suggest traditions of making baskets and mats out of reeds and grasses. The Harappan crafts display an impressive level of standardization. Kenoyer (1998: 149–50) has suggested that state control may have been responsible for the high level of standardization in crafts that were considered to have a value in maintaining the socioeconomic or ritual order and which used non-local raw materials and highly complex technologies (e.g., the making of seals, stoneware bangles, and stone weights). Leaving aside pottery and bricks, crafts using local materials and simple technologies tend to show greater variation. Standardization extended to units of weights and measure. Cubical weights made of chert, chalcedony, black stone, etc. have been found at all excavated sites, and their accuracy all over the Harappan culture zone is remarkable. The system is binary in the smaller weights (1:2:8:16:32:64) and decimal in the higher weights (with a ratio of 160, 200, 320, and 640). The largest weight found at Mohenjodaro weighs 10.865 g. A shell scale was found at Mohenjodaro and an ivory scale at Lothal; a shell object found in Saurashtra was probably used to measure angles.

STONE WEIGHTS, DHOLAVIRA

What is the explanation of the high level of standardization in crafts such as pottery-making and brick making? Does it imply centralized control by merchants or rulers? Some element of central direction is suggested, but its nature and degree are far from certain. If not direct, it may have taken the indirect form of facilitating or controlling the flow of at least some of the raw materials and finished goods. On the other hand, the level of standardization could also indicate the fanning out of hereditary craft specialists over large areas, or a well-developed network of internal trade. It is possible that craftsmen and traders may have been organized in corporate groups similar to guilds, but there is no proof of this. NEW DIRECTIONS IN RESEARCH

The making of long carnelian beads

The city of Khambhat (Cambay) in Gujarat is one of the largest centres of stone bead- making in the world today. Mark Kenoyer, Massimo Vidale, and Kuldeep K. Bhan conducted an ethnoarchaeological study, examining the techniques used by modern bead makers of this place.

They supplemented this with experimentation and an analysis of the remains of bead manufacture at the site of Chanhudaro in south Pakistan. The results throw light on how the Harappan craftspersons may have made their beautiful long barrel cylinder beads. The process must have been something like this: Long nodules of carnelian (a reddish orange variety of agate) were brought from Gujarat to Chanhudaro. The best were chosen and separated. These were dried in the sun for many months and then heated in shallow ovens to make the stone easier to work. The heating also deepened the red colour. The bead roughouts were made using a copper-tipped stake and an antler or horn hammer, using indirect percussion or pressure flaking techniques. Larger nodules were cut lengthwise and chipped to make bead roughouts. These roughouts were then partially ground on grooved sandstone or on quartzite grinding stones. Then came the drilling of holes through the beads. This was done using a special cylindrical drill made out of a rare metamorphic rock which was heated to make an extremely hard and durable tool. This material has been given the name of ‘Ernestite’, after the archaeologist Ernest J. H. Mackay, who was the first to discover the drills and understand their significance. It could have taken a craftsperson a whole day of work—heating, chipping, and grinding—to make a drill. The Harappan bead makers used many different sizes of drills (at least six sizes) to make a single bead. The drilling was probably done with a hand-held bow drill. The friction would have produced intense heat, so the work may have been done under water, or at least by dripping water continuously on the drill hole. The study conducted by Kenoyer and his team showed that even with these superior drills, it would have taken over 24 hours or three 8-hour days of steady drilling to perforate a single 6 cm long bead. The beads on the belts found at Mohenjodaro and Allahdino vary from 6 to 13 cm in length. It would have taken 3–8 days to make one of the longer beads, probably more, considering that the bead makers of Khambhat take long breaks after a couple of hours of work, as it is a very strenuous and tiring process. Once the beads were perforated, there was a laborious polishing process. Taking the process from start to finish, it would have taken over 480 work days to make a belt of 36 beads of the kind found at Allahdino. Even if more than one worker was put on the job, it would still have taken up to a year. These beads must have been highly valued and worn only by the rich. For people who could not afford the expensive long carnelian beads, Harappan craftspeople made imitations in terracotta and painted them red. Kenoyer, Vidal, and Bhan also analysed the archaeological patterns of manufacturing waste and finished artefacts, the structural evidence, and settlement layout in order to make inferences about the way in which bead manufacture was organized and controlled. Why did the Harappans transport carnelian nodules from Gujarat to Chanhudaro, instead of getting at least some of the preliminary work, such as discarding poor quality nodules, done near the source of the raw materials? The evidence suggests that all stages of carnelian bead manufacture at Chanhudaro were centralized and controlled by a powerful and wealthy group of merchants. This also

explains the uniformly good quality of the raw materials used and the high level of standardization. This is in contrast to evidence from the Moneer area at Mohenjodaro, which is suggestive of short-term production by several entrepreneurs. SOURCE Kenoyer et al., 1995

Networks of Trade

The discovery of the Harappan civilization generated a great deal of interest in Harappan– Mesopotamian trade links. This is because before the advent of radiocarbon dating, these links gave vital clues for dating the Harappan culture, and also due to the prevailing interest in cross-cultural comparisons. Over the years, however, many scholars have come to the conclusion that Harappan– Mesopotamian trade may not have been as substantial as earlier held. Other areas such as the Persian Gulf have been identified as important zones of interaction as far as the long-distance trade of the Harappans is concerned. However, it is clear that trade networks within the Harappan culture zone and those linking the culture with other areas in the subcontinent were extremely significant; they are crucial for understanding the structure of the Harappan civilization as well as its striking level of cultural homogeneity. The importance of such trade is clear from the very wide range of raw materials and finished goods that found their way to different parts of the vast Harappan culture zone. This was an age before the advent of coinage, and the vibrant trade of the Harappans was based on barter. One of the important aspects of Harappan trade is the identification of the sources of major raw materials used by the Harappans. The best way of doing this is to scientifically analyse the artefacts and to compare the results with raw materials from various possible sources. Unfortunately, there are not enough studies of this kind so far. Another method is to plot the location of the known resources of various raw materials, especially those closest to the Harappan culture zone. Proof that these were being used in protohistoric times would, of course, give clinching evidence. Unfortunately, this is not usually available, and the earliest evidence of the exploitation of these resources is often contained in 18th/19th century textual references. In spite of its limitations, this kind of exercise is useful in helping identify probable sources of raw materials used by the Harappans. The discovery of factory sites in the limestone hills of Sukkur and Rohri indicates that chert blades were mass produced here and sent to various Harappan settlements in Sindh. The Khetri deposits of Rajasthan must have been an important source of copper. Reference was made in Chapter 3 to the links between the copper-manufacturing Ganeshwar–Jodhpura culture and the Harappan civilization. Lead and zinc probably also came from Rajasthan. Tin is available in the Tosam area of modern Haryana, but other possible sources are Afghanistan and central Asia. Gold may have come from the Kolar fields of Karnataka, where it may have been obtained via trade from the neolithic people who lived there. These neolithic herders may also have been exporters of cattle. (Fine disc beads, probably of steatite paste, found at Piklihal may have been obtained from the Harappans.) Gold could also have been panned from the sands of the upper Indus. Most varieties of semiprecious stone used for bead manufacture came from Gujarat. The exception is lapis lazuli, which was probably obtained from Afghanistan, although it also occurs in the Chagai hills in Baluchistan.

Traders must also have been engaged in a brisk trade in grains and other food products, transporting these between villages and cities. Two-wheeled carts were an important mode of transport for people and goods. Bronze and terracotta models of carts have been found at various sites. No carts survive, but their tracks have been found at several sites, indicating spans roughly similar to those used today. Traders must also have transported their merchandise across long distances in caravans of pack animals such as oxen, sheep, goats, and donkeys. Towards the end of the mature Harappan phase, there is evidence of the use of the camel. The use of the horse seems to have been very minimal. Boats are depicted on seals and moulded tablets, and clay models have been found at Harappa and Lothal. River boats had cabins, ladders leading to the roof, and a high seated platform on the stern for navigation. Seafaring boats had a sharp keel, pointed prow, high flat stern, and mast and ropes for sails.

TERRACOTTA CART, HARAPPA

Several routes of trade and communication connected the various parts of the Harappan culture zone—Baluchistan, Sindh, Rajasthan, Cholistan, Punjab, Gujarat, and the upper doab. These routes can be reconstructed by studying the geographical landscape, settlement patterns, and the distribution of raw materials and finished products. Lahiri (1992: 112–43) points out that major trade routes connected the following areas: Sindh and south Baluchistan; coastal Sindh, upper Sindh, and the central Indus plains; the Indus plains and Rajasthan; the regions lying to the north of the Indus and Harappa; Sindh and east Punjab; east Punjab and Rajasthan; and Sindh and Gujarat. Some of the routes were already well defined in the early Harappan phase—e.g., the Baluchistan–Sindh route via the Kirthar mountains, and the route from east Punjab and Rajasthan via the Cholistan tract. The route connecting north Afghanistan, the Gomal plain, and Multan with a feeder route going to the Taxila valley also continued to be important. Certain routes that were being used in the earlier period became more important in the mature Harappan phase—e.g., the routes within Sindh, between Sindh and the central Indus plains, and between Sindh and Baluchistan via Kutch and Kathiawar. It is likely that the Indus saw a certain amount of riverine traffic. There was also a coastal route linking the Gujarat sites such as Lothal and Dholavira to sites such as Sutkagen-dor on the Makran coast. The location of some of the important sites can in fact be explained in relation to the trade routes of the time. For instance, Mohenjodaro lay at the intersection of the water-route of the Indus and the east– west land route that linked the Quetta valley and the Bolan river to Kot Diji and the western Nara.

MAP 4.3 HARAPPAN ROUTES OF INTERNAL TRADE (AFTER LAHIRI, 1992)

The main sources of information on long-distance trade include a number of Harappan or Harappan-related (i.e., similar to Harappan types) artefacts found at sites outside the subcontinent, and foreign objects found at Harappan sites. These are supplemented by textual sources in the case of Indus–Mesopotamian trade (see Chakrabarti, 1990). A number of Harappan and Harappan-related objects have been found in south Turkmenistan at sites such as Altyn Depe, Namazga, and Khapuz. These include ivory dice, two types of metal objects (a spearhead and ladle), an ithyphallic terracotta, perforated ware, a segmented bead, and a silver seal. The most definite evidence comes from Altyn Depe, in the form of a rectangular Harappan seal bearing the Harappan script. The sites in Iran which have yielded Harappan and Harappan-related artefacts are Hissar, Shah Tepe, Kalleh Nisar, Susa, Tepe Yahya, Jalalabad, and Marlik. The main evidence consists of seals and carnelian beads (both the etched and long barrel cylinder types). The most important evidence of trade with Afghanistan comes from an isolated Harappan trading outpost at Shortughai.

Many years ago, a round seal with a short-horned bull motif and Harappan writing was found at Failaka in the Persian Gulf. In recent years, there has been a substantial increase in the evidence of Harappan trade contacts with the Persian Gulf area. Harappan and Harappan-related artefacts (including a piece of ivory, a linga-shaped object, a circular mirror, and seals with Harappan motifs and/or writing) have been found at Rasal-Qala on the island of Bahrain. Excavations near Hamad in Bahrain yielded a typical Harappan seal and carnelian beads in burials. A seal with the bull motif and Harappan script was found at the site of Hajjar. From Failaka, apart from the ‘Persian Gulf seal’ mentioned above, there was a flat, round seal with the Harappan script. Jar fragments with Harappan writing have been found at many sites in the Persian Gulf. These were probably containers used to transport perishable goods from the Harappan culture zone to this region. The Harappans were also trading with the Oman peninsula. An etched carnelian bead of the Harappan type was found at Umm-an-Nar. There are similarities between certain other types of objects found at this site (a square steatite seal, fragments of pottery, carnelian beads, a cubical stone weight, etc.) and Harappan artefacts. Maysar, an excavated copper-smelting site, has yielded evidence (e.g., pottery decorations and motifs on a seal) that suggests Harappan influence. The major imports from Oman may have included chlorite vessels, shell, and perhaps mother-of-pearl. Copper has been mentioned as another Omani export to the Harappans, but this is unlikely, as the metal was available closer, in Rajasthan. As for Harappan exports to Oman, the items that survive in the archaeological record include beads, chert weights, and ivory objects. There is literary as well as archaeological evidence for Harappan trade with Mesopotamia. Mesopotamian records of the time of king Sargon (2334–2279 BCE) refer to ships from the lands of Dilmun, Magan, and Meluhha tied along the quay of the capital city, Akkad. Dilmun can be identified with Bahrain, and Magan with the Makran coast and Oman. Meluhha may have been a generic term for areas lying to the east of Mesopotamia, including the Indus valley, or it may refer specifically to the Indus valley. The archaeological evidence for Harappan–Mesopotamian trade consists mainly of a few Harappan or Harappan-related seals and carnelian beads at Mesopotamian sites such as Kish, Lagash, Nippur, and Ur. Carnelian beads (both the etched type and the long barrel-cylinder type) were also found in the royal graves at Ur. Certain motifs such as the bull on Mesopotamian seals have been cited as reflecting Harappan influence. Cylinder seals (which are common in West Asia) with Harappan-type motifs suggest interaction between merchants of these two areas. The absence of Mesopotamian seals and sealings in the Harappan context suggests that Mesopotamian traders were not directly involved in the Harappan–Mesopotamian trade interactions.

MAP 4.4 LONG-DISTANCE TRADE ROUTES

Carnelian beads were clearly an important Harappan export to West Asia. Textiles and conch shell objects were other possible exports. Ivory and ivory objects may have been exported by the Harappans to Afghanistan, Turkmenistan, and perhaps the Persian Gulf. Mesopotamian texts mention the following items as imports from Meluhha: lap1/22/2016is lazuli, carnelian, gold, silver, copper, ebony, ivory, tortoiseshell, a chicken-like bird, dog, cat, and monkey. Mesopotamia’s general exports included fish, grain, raw wool, woollen garments, and silver. It is possible that wool and silver found their way to Meluhha, but there is no archaeological proof of this. There are two very different assessments of Harappan–Mesopotamian trade. Ratnagar (1981) highlights the importance of this trade, especially the trade in lapis lazuli, and even argues that its decline was a reason for the decline of the Harappan civilization. Notwithstanding the long list of items mentioned in texts, the fact remains that there are very few Harappan artefacts found in Mesopotamia and even fewer Mesopotamian artefacts found at Harappan sites. A few Mesopotamian-type stone weights have been reported from Mohenjodaro and Harappa. Three motifs found on some Harappan seals are seen by some scholars as reflecting Mesopotamian influence—the whorl design, a man grappling with two animals, and the gatepost motif. The evidence as a whole is not very substantial. Chakrabarti (1990) and Shaffer (1982b) argue that Harappan trade with Mesopotamia was not direct, extensive or intensive. This trade does not seem to have been particularly important for the development or sustenance of the Harappan civilization. Among the Harappan imports via long-distance trade, lapis lazuli was probably an import from Afghanistan (or it could have been obtained closer from the Chagai hills of Baluchistan). Jade must

have come from Turkmenistan. Tin may have been obtained from Ferghana and eastern Kazakistan in central Asia. Carved chlorite and green schist vessels were a popular item of trade in West Asia and the Persian Gulf, and a few fragments have been found at Mohenjodaro. These may have been imported from southern Iran or from Baluchistan. Very few West Asian artefacts have been found in Harappan contexts. A seal of the Persian Gulf type was found at Lothal as a surface find. A lapis lazuli bead from Mohenjodaro and a pendant with lapis lazuli inlay found at Cemetery-H levels at Harappa were possibly imports from West Asia. A cylinder seal (as mentioned earlier, cylinder seals were common in West Asia) with Indian motifs was found at Kalibangan. Harappan objects in Mesopotamia can be dated from the Early Dynastic IIIA period (c. 2600/2500 BCE) to the Isin–Larsa period (c. 2000/1900 BCE) in the Mesopotamian sequence, which corresponds to the entire span of the mature Harappan phase. The finds from other parts of West Asia also belong roughly to this period. However, the discovery of a Harappan seal at the site of Nippur in a 14th century BCE context suggests that Harappan contact with Mesopotamia may have continued, although in a diminished form, into the late Harappan phase. The continuation of some amount of trade with the Persian Gulf region is suggested by two Harappan seals found at Failaka in a 14th century BCE context, and a late Harappan seal found at Bet Dwarka. The latter has Harappan writing and a three-headed animal motif similar to that found on certain Persian Gulf seals. The importance of overland routes from the Harappan civilization through Afghanistan is evident from the location of Harappan sites near each of the passes and routes that lead through Baluchistan into Afghanistan. Pathani Damb is near the Mula pass, Nausharo near the Bolan pass, Dabarkot in the Gomal valley, and Gumla and Hathala in the Derajat, along the route via the Gomal pass. The Gomal route seems to have been the most important. Two main overland routes connected the Harappan civilization with West Asia. The northern one passed through northern Afghanistan, north Iran, Turkmenistan, and Mesopotamia, crossing sites such as Shortughai, Tepe Hissar, Shah Tepe, and Kish. A southern route passed through Tepe Yahya, Jalalabad, Kalleh Nisar, Susa, and Ur. The maritime route to Mesopotamia may also have been used. It is likely that sites such as Sutkagen-dor, Balakot, and Dabarkot (the latter two may at that time have been located at the coast instead of some distance away) were important points along this route. Lo-thal (10 km away from the Gulf of Cambay) and Kuntasi (on the Phulki river, 4 km from the coast), Dholavira (in the Rann of Kutch), and the sites along the coast of Kutch no doubt played an important role in maritime trade. FURTHER DISCUSSION

Shortughai—a Harappan trading post in Afghanistan

Shortughai is located near the confluence of the Oxus and its tributary, the Kokcha, in north-east Afghanistan. It is a small site, only about 2 ha. The cultural deposit is 2.5–3 m thick, within which four periods of occupation have been identified. Period I (50 cm thick) was dated by radiocarbon to the end of the 3rd millennium BCE. The discoveries of Period I included the following: pottery with Harappan designs, terracotta cakes, fragments of toy carts, copper and bronze objects, pieces of gold and lead, a discoidal

gold bead; lapis lazuli, agate, carnelian, steatite, small barrel-shaped agate beads; long tubular and etched carnelian beads; flint micro-blades and drill heads; shell bangles; and mud-bricks of the typical Harappan size. Harappan graffiti occurred on the rims of jars and on beakers. There was a square Harappan seal with the motif of a rhinoceros and the Harappan script. The discovery of so many typical Harappan artefacts and manufacturing techniques proves that this was not a site which had mere contact with the Harappan civilization, but a site belonging to the Harappan civilization. Shortughai also has some unique features. A ploughed field covered with flax seeds was found in an area unsuitable for irrigation, showing the practice of dry farming. Small irrigation canals drawing on the water of the Kokcha, located about 25 km away, were found in other parts of the site. What were the Harappans doing at Shortughai? This site seems to have been connected with the lapis lazuli mines nearby. However, lapis lazuli objects are not particularly numerous at Harappan sites. A second possibility is that Shortughai owed its importance to its proximity to the tin mines of Afghanistan and Ferghana. A third possibility is that it had a role to play in camel trade. SOURCE Chakrabarti, 1990: 1–2, 86–89

The argument that the quantum of Harappan long-distance trade was not great is persuasive. Unlike the resource-poor area of Mesopotamia, the Harappan culture zone was rich in a variety of natural resources. Food requirements and most of the raw materials required by Harappan craftspersons could have been met by resources available within the Harappan culture zone. The diverse, well-developed craft traditions meant that most of the finished goods required by the Harappans were likewise available from within this area. A few raw materials and products were obtained from other parts of the subcontinent and from areas such as Afghanistan and central Asia. Very few essential items had to be imported from distant places. Harappan trade must have involved highly organized merchant groups as well as nomadic peddlers in the mountainous stretches. The extent of state control over this activity is a matter of debate. The Nature and Uses of Writing

Among the biggest mysteries about the Harappan civilization are the language (or languages) the Harappans spoke and their writing system. It is likely that people living in various parts of the Harappan culture zone spoke different languages and dialects. The writing on the seals was probably in the language of the ruling elite. Some scholars have suggested that this language belonged to the Dravidian family of languages, while others have argued in favour of the Indo-Aryan family. However, there is so far no consensus on the affiliation of the Harappan language or on the decipherment of the script. A total of about 3,700 inscribed objects have been found at Harappan sites (for details, see Mahadevan, 1977, Parpola, 1994). Most of the writing appears on seals and sealings (seal

impressions), some on copper tablets, copper/bronze implements, pottery, and other miscellaneous objects. About 50 per cent of the inscribed objects have been found at Mohenjodaro, and the two sites of Mohenjodaro and Harappa together account for about 87 per cent of all inscribed material. Most of the inscriptions are very short, with an average of five signs. The longest one has 26 signs. The script seems to have emerged in a fully evolved state and does not show any significant changes over time. This conclusion may, however, be the result of the inadequacies of earlier excavations, which did not record the stratigraphic context of all objects, making it difficult to sort out earlier and later samples of writing.

THE DHOLAVIRA 'SIGNBOARD'

There are 400–450 basic signs and the script is logo-syllabic—i.e., each symbol stood for a word or syllable. It was generally written and meant to be read from right to left (this is reversed on the seals). This is evident from that fact that in inscriptions, the letters are cramped on the left side, where space had clearly run out, and from overlapping letters scratched onto pottery. There are a few instances, however, of writing from left to right. Longer inscriptions that consisted of more than one line were sometimes written in the boustrophedon style—with consecutive lines starting in opposite directions. What was the connection between the motifs on the seals and the writing? What was the extent of literacy among the Harappans? What was writing used for? In order to understand the uses of writing in the Harappan civilization, it is necessary to try to interpret the functions of the inscribed objects. Writing appears very frequently on the seals. Some of these were impressed onto small moist clay tablets known as sealings, probably by merchants to authenticate their bales of merchandise. The evidence of textile impressions on some sealings supports this interpretation. However, more seals than sealings have been found, and the seals are generally worn at the edges and not inside. This suggests that some of the so-called seals may have had other functions. They may have been tokens used in the buying and selling of goods. They may also have been worn as amulets or used as identification markers (like modern identity cards) by well-to-do people like landowners, merchants, priests, artisans, and rulers. Those no longer in use must have been intentionally broken so that they could not be misused by anybody. Tablets with narrative scenes may have had a religious or ritualistic function. The so-called ‘seals’ were thus used for multiple purposes. Writing also appears on miniature tablets made of steatite, terracotta, and faience. Since these objects were not used to make impressions, unlike the seals, the writing on them was not reversed. Many of the objects were discovered at Harappa and other large cities. Rectangular copper tablets with writing and animal motifs were found at Mohenjodaro, while a few tablets with raised writing were found at Harappa. The limited number of places where they occur suggests a restricted use. Interestingly, there are many duplicates of both the miniature and copper tablets. The evidence of writing on pottery suggests a wider use in craft production and economic transactions. Harappan potters sometimes inscribed letters onto pots before firing. At other times, inscriptions were made on pots after they were fired (this is termed ‘graffiti’). Even if the potters who made the marks on their pots were themselves illiterate, they must have been able to recognize

the symbols. Pointed goblets sometimes have seal impressions, which may have indicated the name or status of the person for whom the pot was made. Items like copper and bronze tools, stoneware bangles, bone pins, and gold jewellery were sometimes inscribed. A copper vessel found at Mohenjodaro contained a large number of gold objects. These included four ornaments with tiny inscriptions, all apparently written by the same hand, probably giving the name of the owner. Some of the writing inscribed or painted on personal possessions such as bangles, tools, beads, and bone rods may have had some sort of magicoreligious or ritualistic significance. The Dholavira ‘signboard’ may or may not indicate a high level of urban literacy, but it does indicate a civic use of writing. It is likely that a very small proportion of Harappan written material survives, and that people wrote on perishable material as well. The evidence of a common script all over the vast Harappan culture zone shows a high level of cultural integration. The virtual disappearance of the script by c. 1700 BCE suggests both a close connection of writing with city life and the lack of sufficient downward percolation of writing.

HARAPPAN SEALS

THE SCRIPT–THE 'JAR' SIGN

AND THE 'MARKER' SIGN

ARE THE MOST FREQUENTLY OCCURRING SYMBOLS

Religious and Funerary Practices

The basic elements of what can be loosely described as ‘Harappan religion’ were outlined by John Marshall in 1931. Although some aspects of Marshall’s interpretation can be criticized—especially his tendency to read elements of later Hinduism into the evidence—he did succeed in identifying several important features of Harappan religion. Hypotheses about this issue are bound to be subjective, especially in view of the fact that the script is undeciphered. The worship of female goddesses associated with fertility has long been held as one of the major features of Harappan religion. This conclusion is based on the following factors: (a) the concerns that agricultural societies are invariably known to have with fertility; (b) cross-cultural parallels with other ancient civilizations; (c) the importance of goddess worship in later Hinduism; and (d) the discovery of a large number of terracotta female figurines that were labelled ‘Mother Goddesses’. Certain representations on seals are also relevant. For instance, a seal showing a nude woman, head downwards, with her legs apart and a plant issuing from her vagina is often interpreted as a prototype of Shakambhari, the Earth Mother. Describing all female figurines as representations of a single great ‘Mother Goddess’ associated with fertility and maternity clearly over-simplifies the situation. The attributes of the figurines and the contexts in which they were found have to be considered carefully before assigning them a religious or cultic significance. As pointed out in an earlier chapter, not all female figurines necessarily represented goddesses (let alone a single goddess), and not all goddesses necessarily

had maternal associations. Some of the Harappan female figurines may have had a cultic significance and may have been part of household rituals. Others may have been toys or decorative items. A study of the Harappan terracottas by Alexandra Ardeleanu-Jansen (2002) has underlined the great variety in the form of female figurines. The type which is frequently interpreted as having a religious significance is a slim female figure with a distinctive fan-shaped headdress, wearing a short skirt. She is heavily ornamented with necklaces, armlets, bangles, anklets, and earrings. Some of the figurines have cup-like attachments and flowers on either side of the head. In certain cases, the cup-like attachments have traces of black residue, suggesting that they were used to burn oil or some sort of essence. Such figurines may have been religious images worshipped in households, votive offerings made to a deity, or part of the paraphernalia of domestic rituals. It is interesting to note that such figures do not appear on Harappan seals and tablets or in stone or metal sculpture. There is also a matronly, pot-bellied type of female figurine who may represent either a pregnant woman or a prosperous woman. She is naked and sometimes wears some jewellery and a turban or head-dress. Both the ‘matronly type’ and the ‘slim type’ of female figurines may hold a baby in their arms. The ‘matronly type’ can stand without support, while the youthful, ‘slim type’ needs support. It is interesting to note that female figurines—including those with possible religious significance—are found in large numbers at sites such as Mohenjodaro, Harappa, and Banawali, but not at sites such as Kalibangan, Lothal, Surkotada or Mitathal. Most of the terracotta figurines (including the female ones) were found broken and discarded in secondary locations. None were found in a context that could be interpreted as a temple. This was one of the reasons why Marshall suggested that they were votive offerings rather than cult images. The fact that so many of them were broken suggests that they may have been part of a ritual cycle and were made for short-term use for certain specific occasions. The relationship between the female figurines and the male and animal figurines with which they are associated needs to be explored.

F EMALE FIGURINE WITH FAN-SHAPED HEADDRESS

Marshall suggested that the Harappans also worshipped a male god represented on a steatite seal discovered at Mohenjodaro, usually referred to as the Pashupati seal. This shows a male figure with a buffalo horn head-dress seated on a dais with his legs bent double under him, heels together, toes pointed down. His outstretched arms are adorned with bangles, his hands rest lightly on his knees. He is flanked by four animals—an elephant, rhinoceros, water buffalo, and tiger. Beneath the dais are two antelopes or ibexes. Marshall thought the male figure was three-headed and ithyphallic (with erect penis). He saw a striking resemblance between this deity and the Shiva of later Hindu mythology, who is also known as Mahayogi (the great yogi) and Pashupati (lord of the animals).

F EMALE FIGURINE, B ANAWALI

Another aspect of the fertility-related beliefs of the Harappans was the worship of male and female creative energy in the form of stone icons of lingas and yonis (representing the male and female sexual organs respectively). A number of such stones were identified by John Marshall. Many years later, George Dales argued that the contexts in which these stones were found do not suggest cultic significance. Some of the ring stones had lines on them and may have had architectural use, either to guide masons in pillar building or to measure angles. Alternatively, they may have been used to make astronomical calculations. Marshall himself had suggested that some of the lingashaped objects may have been grinders or unfinished weights. Dales made his arguments forcefully; however, a terracotta piece which closely resembles a linga with a yoni-pitha (yoni base) has recently been found at Kalibangan. The Harappan seals, sealings, amulets, and copper tablets depict a number of trees, plants, and animals, some of which may have had cultic significance. The pipal (Ficus religiosa) tree appears often and may have been venerated. Sometimes, there is a figure peering out from between its branches, possibly a tree-spirit. A seal found at Mohenjo-daro shows a row of seven figures with long braids standing in front of a pipal tree which has a horned figure standing in it. It is not clear whether the figures are male or female, but because they are seven in number, scholars have speculated that there may be a connection with the later traditions of the seven rishis or the seven mothers. Some of the animals depicted on seals and sealings—for instance, the humped and humpless bull, snake, elephant, rhinoceros, antelope, gharial, and tiger—may have had cultic significance. The bull, a symbol of male virility in many ancient cultures, seems to have been particularly important. We can note the steatite bull statuettes discovered at certain sites, including a very sophisticated terracotta bull found at Mohenjodaro. It is possible that some of the terracotta animals on wheels may have been cult images rather than toys. Two Harappan sealings appear to represent animals being carried in processions; one of them resembles a bull or cow. The composite animals (tiger– human, bull–elephant, ram–bull–elephant, etc.) and the ‘unicorn’ depicted on some seals and sealings may also have had some sort of religious or mythological significance. Some of the terracotta, shell,

faience, and metal tablets may have been amulets. Their motifs, such as the svastika, may have been associated with a protective function or auspiciousness. Terracotta masks and puppets found at Mohenjodaro and Harappa include those in the form of real and mythical animals, and these may have been used in religious, political, or politico-religious rituals.

Man, god, or goddess?

THE 'P ASHUPATI SEAL'

Marshall concluded that this seal showed that the Harappans worshipped a god who seems to have been a proto-Shiva. This conclusion has not gone unchallenged. The questions that have been asked include the following: Is 1.the figure really sitting in a yogic posture of ritual discipline? Is 2.he really three-headed? Is 3.he ithyphallic? Is 4.the figure a male? Shiva 5. as Pashupati in later Hindu mythology protects domesticated cattle, while the figure on the seal is associated with wild animals. In view of this difference, can the two really be connected?

The figure has been variously identified as a chieftain, a divine bull-man, Indra, or the demon Mahisha of the Puranas. M. K. Dhavalikar and Shubhangana Atre (see Atre, 1985–86) have suggested that it represents a goddess—a ‘lady of the beasts’. Notwithstanding all these alternative interpretations, the basics of Marshall’s interpretation are still persuasive. The figure can be accepted as that of a male seated in a yogic posture, although it is not certain that he was three-headed. The similarities between the deity—for he seems to be no ordinary man—and certain attributes of the later-day Shiva remain striking. Of course, we do not know what name the Harappans gave him. We can recall here the ‘horned deity’ that appears on a Kot Diji pot, Kalibangan terracotta cake, and the Padri jar. This indicates that the worship of a horned deity goes back to the early

Harappan phase.

FURTHER DISCUSSION

The ‘fire altars’

The citadel complex at Kalibangan consists of a northern and southern unit, separated from each other by a wall. In the southern sector, archaeologists found five or more mud-brick platforms, separated from each other and from the back of the fortification wall by streets. Steps or ramps led up to the platforms. On one of these platforms, there was a row of seven clay-lined pits, each about 75 × 55 cm. These have been identified as ‘fire altars’, i.e., pits in which offerings were made into the fire as part of sacrificial rituals. Ash, charcoal, the remains of a rectangular clay piece, and terracotta cakes were found in them. To the west of this row of pits, within easy reach of whoever sat in front of them, was the lower half of a jar containing ash and charcoal, embedded into the ground. Nearby was a well and the remains of bath pavements with attached drains, all made of burnt bricks. A ‘fire altar’ and a well were discovered on another platform in the southern sector of the citadel complex. There was also a 1.25 × 1 m brick-lined rectangular pit, containing cattle bones and antlers. This suggests the practice of animal sacrifice. The southern sector of the Kalibangan citadel complex seems to have been a place where sacrificial rituals of a congregational character were performed. The northern part of the citadel complex contained houses. B. B. Lal suggests this may have been where the priests who performed the rituals lived. ‘Fire altars’ have also been reported at Banawali, Lothal, Amri, Nageshwar, and Vagad in Gujarat and at Rakhigarhi in Haryana. But it is only at Kalibangan and Banawali that they may have signified some community event; in the other cases, they seem to have been associated with domestic rituals. Again, as in the case with female figurines, the fact that the ‘fire altars’ have been found at a few sites but are absent at most, indicates variations in religious practice within the vast area of the Harappan culture. SOURCE Lal, 1984

The Great Bath was probably the scene of an elite ritual activity involving ceremonial bathing. A triangular terracotta cake found at Kalibangan has a carving of a horned deity on one side and an animal being dragged by a rope by a human on the other. The latter has been tentatively interpreted as suggesting the practice of animal sacrifice. A Kalibangan cylinder seal shows a woman flanked by two men who hold her with one hand and raise swords over her head with the other; this may represent a scene of human sacrifice. The most striking evidence suggesting ritualistic practices comes from the ‘fire altars’ found on the citadel mound at Kalibangan. Harappan cemeteries have been located at sites such as Harappa, Kalibangan, Lo-thal, Rakhigarhi, and Surkotada. The most common method of burial was to place the body of the

deceased in an extended position, with the head towards the north, in a simple pit or brick chamber. Grave goods including food, pottery, tools, and ornaments were placed along with the body, but they were never too many or lavish. Clearly, the Harappans preferred to use wealth in life rather than bury it with their dead. At Harappa, there was a coffin with a shroud made of reeds. Symbolic burials with grave goods but no skeletons were found at Kalibangan. Fractional burials (where the body was exposed to the elements and the bones then gathered and buried) were found at Mohenjodaro and Harappa. These two sites also gave evidence of urn burials suggestive of cremation. Multiple burials of men and women were discovered at Lothal. The religious and funerary beliefs and practices of the Harappans show great variety. While there are dangers in viewing these through the lens of later-day Hinduism, it is interesting to note that the Harappan civilization does display a few features reminiscent of later traditions, except, however, the important element of temple worship. Not a single structure found at any Harappan site can conclusively be identified as a temple.

HARAPPAN SEALS WITH DEPICTIONS OF TIGER AND ELEPHANT

The Harappan People

What did the Harappan people look like? What sorts of clothes and ornaments did they wear? How did they relax and have fun? Terracotta, stone, and bronze sculptures (some of which have been described in earlier sections) help answer such questions. The form of human terracotta figurines was connected to their function, stylistic conventions, and audience, and they may not be realistic representations of what all or even most Harappans looked like. Nevertheless, they do help insert three-dimensional people into our picture of the Harappan civilization. The human terracottas can be divided into female and male figurines, those whose sex is not clear, a few that have both female and male attributes (e.g., a figurine from Harappa which has breasts and

a beard), and a few males in feminine dress. Going by the figurines, Harappan women wore a short skirt made of cotton or wool. They wore their hair variously in braids, rolled into a bun at the back or side of the head, arranged in separate locks or ringlets, and wrapped around the head like a turban, or left loose. What looks like a fan-shaped headdress could actually represent hair stretched over a frame made of bamboo or some other material. At Harappa, it is supplemented by flowers or flower-shaped ornaments. Such hairstyles or headdresses could indicate women of distinction or deities. Female figurines wear ornaments such as necklaces, chokers, hair ornaments, bangles, and belts. We can recall the beautiful jewellery found at many Harappan sites. Male figurines are usually bare headed, though some are turbaned. Most of them are nude, so it is difficult to say what sort of clothes men wore. Certain stone sculptures suggest the use of a dhoti-like lower garment and an upper garment consisting of a shawl or cloak worn over one shoulder and under the other. There are various hairstyles—braids, buns, and hair hanging loose. Most of the male figurines have beards, in styles ranging from the ‘goatee’ to the more common combed and spreadout style as in the case of the ‘priest-king’. There is some degree of overlap in male and female hairstyles and ornaments, but also some differences. For instance, men and women both wear bangles and necklaces, but men rarely wear multi-strand necklaces made of graduated beads. Children of all cultures and all times play with toys, and Harappan children were no exception. Terracotta toys of various kinds have been found at Harappan sites. They include balls, rattles, whistles, gamesmen, carts with moveable parts, and animals on wheels. There are spinning tops made of terracotta and shell. Some have a shallow depression, while others have a copper tip to make them spin around a long time. Clay marbles have been found in courtyards of houses. Miniature terracotta cooking vessels, beds, and other toy furniture have been found, with which children must have played house. There are figurines of children playing with toys. One of them holds what seems to be a clay disc. Many clay discs have in fact been found at Harappan sites, and it is possible that these are remnants of a pithu-like game played with a ball and piled-up pieces of clay or stone. Lots of terracotta figurines of dogs have been found at Harappan sites, some with collars, suggesting that people kept dogs as pets. Some of the terracotta figurines of people and animals have a comic appearance, reflecting a sense of humour. The social implications of the worship of female deities are complex. Although such worship reflects the ability to visualize divinity in feminine form, it does not necessarily translate into power or a high social position for ordinary women. While some of the female figurines found at Harappan sites may represent goddesses, many seem to represent ordinary, mortal women. Terracotta figurines of women at work are few. Figurines depicting women grinding or kneading something (food/clay?) have been found at Nausharo, Harappa, and Mohenjodaro, suggesting the association of women with food-processing activities. In ancient societies, childbirth was a process fraught with danger. Some of the fat female terracotta figurines may represent pregnant women. Recent excavations at Harappa have yielded a burial with a woman and baby, perhaps a case of death in childbirth. Some female figurines found at Harappan sites carry a suckling infant on the left hip; others show women carrying infants close to their breast. An unusual terracotta figurine found at Nausharo (Period ID) shows a male with feminine headdress holding an infant. Tiny terracotta figurines of small children have been found at most sites. Were all of them toys or could they be votive objects? Can a statistical analysis of the child figurines help us identify whether there was a cultural bias in favour of male or female children? This is a very interesting question, but answers can only be speculative.

TERRACOTTAS: FIGURINE; GAMES AND DICE

P ERFORATED BIRD-SHAPED RATTLE; BULL WITH MOVEABLE HEAD; CART

NEW DIRECTIONS IN RESEARCH

How healthy were the Harappans?

The early excavations at Harappa focused on architecture and artefacts. The more recent excavations carried out during the 1980s and 1990s reflect the advances in the field of archaeology and included a careful collection and scientific analysis of bone remains. The results give us important information about the health and nutrition of the Harappans. Cemetery R-37 is located in the southern part of the site. Excavations were carried out under the supervision of J. M. Kenoyer. A team of four physical anthropologists—K. A. R. Kennedy, John R. Lucacs, Nancy Lovell, and Brian Hemphill—had the special job of carefully excavating the skeletons and removing them to the laboratory for analysis. Ninety skeletons were recovered from the cemetery. Most of them represented females. The number of skeletons in different age ranges were as follows: Children (< 16 yrs)

:

15

Young adults (17–34 yrs)

:

35

Middle-aged adults (35–55 yrs)

:

27

Older adults (> 55 yrs)

:

13

The general health of this sample of the Harappan population was quite good. The skeletons showed a low incidence of traumatic injury, chronic infectious diseases, and neoplastic diseases (tumours). There were no traces of nutritional inadequacy such as rickets, scurvy, or anaemia. There were, however, three cases of arrested growth lines, suggesting that growth during childhood was halted temporarily. This could have been due to malnutrition or some serious illness. The most common ailment suffered by the people buried in this cemetery was arthritis. Signs of this appeared in the spine and in the joints of knees, hands, and feet. There were also several instances of severe arthritis in the neck, which may have been the result of unusual stress on the neck vertebrae, perhaps due to carrying heavy loads on the head over a long period of time. The teeth of the people were analysed and the dental pathology profile was what would be expected in a community of agriculturists. The most common dental problem was gross enamel hypoplasia (pitted or missing enamel) and the least common was hypercementosis (excessive deposit of cementum, a calcified hard tissue covering the root surface). Dental caries (cavities) were present in 43.6 per cent of the individuals examined. The dental caries rate was worked out as 6.8 per cent, which is a high rate typical of agricultural groups. Tooth loss, calculus (hardened plaque or tartar), and alveolar resorption (wasting away of the bony socket) occurred with moderate frequency. There were differences between males and females in the incidence of tooth loss and enamel hypoplasia. But the frequency of dental abscesses, calculus, and alveolar resorption were more or less the same for men as for women. The study showed that the Harappans buried in Cemetery R-37 were relatively healthy agriculturists. A statistical analysis of the crania of the skeletons shows biological similarity among the people buried in the cemetery, a similarity between them and the skeletons found in the late Harappan Cemetery-H, and with the modern populations inhabiting this area today. This shows a broad biological continuity between the inhabitants of the area from mature Harappan to late Harappan into more recent times. SOURCE Dales and Kenoyer, 1991: 191–99, 210–12

Early studies of Harappan skeletons focused on classifying the Harappans into racial types. More recent studies have abandoned the old, rather arbitrary racial classifications. They have asked different questions and given an interesting set of conclusions. Kenneth A. R. Kennedy’s study (1997) of skeletons found at Harappan sites shows biological heterogeneity between the different regions, and similarity with the people who live in these areas today. This means that the Harappans of Punjab resembled the present-day Punjabis in appearance, while the Harappans of Sindh resembled the modern inhabitants of Sindh. Kennedy also identified the incidence of malaria among the Harappans. There is the larger question of the analysis and assessment of the structure of Harappan society. The absence of deciphered written evidence is a major handicap, and inferences have to be made very carefully on the basis of archaeological data. The people who lived within the Harappan culture zone comprised villagers and city folk. Harappan society included occupational groups such

as farmers, herders, hunter-gatherers, craftspeople, fisherfolk, merchants, sailors, rulers, administrative officials, ritual specialists, architects, carpenters, brick masons, well diggers, boat makers, sailors, sculptors, shopkeepers, sweepers, garbage collectors, and so on. Some farmers may have lived in the cities and tilled their fields nearby. Terracotta net sinkers and arrow points found at Mohenjodaro and Harappa suggest that the city population included hunters and fisher-folk. The level of social differentiation may not have been as great as in Mesopotamia and Egypt, but differences in house sizes and the hoards of jewellery do indicate a concentration of wealth and differences in social and economic status. The affluent social groups would have comprised rulers, land owners, and merchants. Class and rank differences based on occupation, wealth, and status must have existed. However, claims that the caste system existed in Harappan society are highly speculative. The Ruling Elite

Political organization includes a range of issues related to the exercise of power and leadership in a society. The debate on the nature of the Harappan political system has focused largely on whether or not a state existed, and if so, what sort of state it was. A great deal depends on our definition of a state and the interpretation of the archaeological evidence. Cultural uniformity does not necessarily mean political unification; therefore there is the additional question of whether the evidence suggests the existence of one state or many. Many scholars have observed that the elements of warfare, conflict, and force in the Harappan civilization seem weak compared to contemporary Mesopotamia and Egypt. Weapons are not a dominant feature of the artefacts found at Harappan sites. There are few depictions of conflict between people in the narrative reliefs on terracotta and faience tablets. However, fortifications, especially the imposing ones at sites such as Dholavira, cannot be overlooked. It is indeed possible that the element of force in the Harappan culture has been underestimated. Force and conflict could not have been completely absent in such a large area over such a long period of time. That the Harappan civilization lasted for some 700 years and its artefacts, traditions, and symbols seem to have continued more or less unchanged through this long period, suggests a strong element of political stability. There must have been groups of rulers in the various cities. Just who they were and how they were related to each other remains a mystery. These groups would have been responsible for the maintenance of the city facilities—walls, roads, drains, public buildings, etc. Some of the seals may bear names, titles, and symbols of these elites and could throw important light on the Harappan rulers, if the writing could be read. One of the earliest hypotheses regarding the Harappan political structure was put forward by Stuart Piggott and was supported to some extent by Mortimer Wheeler (for details of the various theories, see Jacobson, 1986). Piggott suggested that the Harappan state was a highly centralized empire ruled by autocratic priest-kings from the twin capitals of Mohenjodaro and Harappa. This view was based on a number of features, including the level of uniformity in material traits, the use of a common script, and standardized weights and measures. Mohenjodaro and Harappa seemed to clearly stand out in the midst of the other settlements. Urban planning and monumental public works implied the mobilization of a specialized labour force. The ‘granaries’ at Mohenjodaro and Harappa fitted in with a view of the Harappan rulers as exercising a high level of control over everything,

even maintaining buffer stocks of grain to tide over times of food scarcity. The apparent lack of internecine warfare between the settlements suggested that they were united under a single rule. This view of the Harappan state soon came in for criticism. Walter A. Fairservis (1967) argued that the Harappans did not have an empire, not even a state. He pointed to the absence of evidence of priest-kings, slaves, standing armies, or court officials. According to him, Mohenjodaro was a ceremonial centre, not an administrative one. He argued that the sort of control reflected in the Harappan civilization could have been exercised by an elaborate village administration. Later, Fairservis modified his views to some extent and agreed that there may have been some element of centralized control and a class structure. But he still maintained that force did not play a significant role and that interdependence, religion, and tradition were responsible for regulating social behaviour. Another view of the Harappan political system came from S. C. Malik (1968), who argued that the lack of imposing monuments and supreme gods goes against the idea of a strong, centralized state. The Harappan polity, according to Malik, is an example of what Elman Service described as the chiefdom stage, transitional between a kinship society and civil state society. KEY CONCEPTS

Defining a state

The word ‘state’ is used very often in historical and anthropological analysis; therefore, it is important to know the various meanings attached to it. Here are some of the frequently cited and used definitions: According to Elman R. Service (1975:14), a state is characterized by the existence of civil law and formal government that are ‘institutionalized, enacted, official’, and which ‘employ, threaten, or imply the actual use of force’. For him, the essential ingredients of a state are the power of force and authority. Ronald Cohen (1978: 69-70) identified the state as a specific type of political system characterized by a centralized bureaucracy and dominant control of the mechanisms of force by a central authority. He further emphasized that an important difference between a chieftaincy and state was the latter’s ability to counter forces of political fission (breakaway groups or splintering). The central element in Morton H. Fried’s (1978) conception of the state is social stratification based on differential access of the members of a society to basic productive necessities. Fried makes a distinction between pristine states and secondary states. A pristine state is one which emerges from indigenous stimuli, usually with no pre-existing models. A secondary state is one which has the model of an already existing state at hand and whose origins are related to pressures from this already existing state.

Henri J. M. Claessen and Peter Skalnik (1978) define an early state in the following way: a centralized socio-political organization for the regulation of social relations in a complex, stratified society, which is divided into at least two basic strata or emergent social classes—the rulers and the ruled—and in which the relations of political dominance and tributary obligations between the rulers and the ruled are legitimized by a common ideology founded on reciprocity (mutual relations of give and take). They also suggest that early states can be divided into three types on the basis of increasing levels of complexity—the inchoate early state, the typical early state, and the transitional early state. Since state formation is a gradual process, it is often difficult to say precisely when something that can be called a ‘state’ appeared. Elman Service suggests that the transitional period between a pre-state kinship society and a state society should be considered a distinct stage in itself called the chiefdom stage. This is characterized by ‘centralized direction, hereditary hierarchical status arrangements with an aristocratic ethos, but no formal, legal apparatus of forceful repression’. He adds that leadership in a chiefdom was exercised by an authority that possessed neither formal legal power nor a bureaucracy. There were social ranks, but no classes. Part of the problem in defining a state is that the many different kinds of state systems that have existed in history make it difficult to formulate a universal definition. For instance, although Fried directs attention to the element of social stratification in state societies, his emphasis on centralization simply does not fit all states. Apart from the problem of definition, in the case of early states, there is also the problem of identifying levels of social and political complexity on the basis of archaeological evidence. Recent studies of the state have questioned various aspects of the older evolutionary models and terminology. For instance, Norman Yoffee has challenged various ‘myths’ related to the evolution and nature of the earliest states. These myths include the ideas that all these states were basically similar: that they were ruled by powerful totalitarian elites who exercised a monopoly of control over goods, services, and information; that they were marked by territorial integration of large areas; and that their social structure can be understood by invoking modern ethnographic parallels. SOURCE Claessen and Skalník, 1978; Yoffee, 2005

The two trends in recent writings are, paradoxically, a return to the idea of a Harappan empire and a complete rejection of such an idea. Ratnagar (1991) analysed the archaeological evidence and used cross-cultural parallels with other early state societies to conclude that we do seem to be looking at a Harappan empire. The strongest critique of such a view has come from Jim Shaffer (1982b). Shaffer questions the level of homogeneity in the Harappan civilization and suggests that it could have been the result of a well-developed network of internal trade rather than a strong, centralized government. He underlines the absence of huge royal tombs, palaces, and temples, and the absence of marked social differentiation of the kind visible in ancient Egypt and Mesopotamia.

At Harappan sites, artefacts of various types are distributed throughout the occupational levels rather than clustered in elite residences or structures. All the typical Harappan artefacts (including ornaments of precious metals and semi-precious stones, seals and sealings, and the script) occur in small village settlements. This suggests an equality of access to wealth or the symbols of wealth among village and city dwellers, which goes against the idea of a centralized empire. The fact that some form of state structure did exist in the Harappan civilization cannot be denied. The absence of marked social or economic differences and tombs or palaces of the Egyptian or Mesopotamian kind does not mean that a state did not exist, rather that it was a different sort of state. The communications system, standardization in artefacts, site specialization, mobilization of labour for public works, the establishment of the trading outpost of Shortughai—all these things indicate a level of economic complexity and the existence of a state. So does the level of cultural homogeneity and the use of a common system of writing across areas in which many different languages and dialects must have been spoken. The levels of social differentiation indicate some degree of class stratification. Some of the buildings on the citadel complex seem to have had an administrative function. Centralized control is apparent in the Harappan civilization. The questions are: How much and by whom?

A priest-king?

In ancient Mesopotamia and Egypt, rulers are portrayed extensively in stone reliefs and sculptures; their palaces, tombs, and temples further proclaim their power. The Harappan case is strikingly different. The stone bust of a male figure found at Mohenjodaro has been given the label ‘priest king’. The figure is that of a man with a close-cropped beard, half-closed eyes, and a fillet with an encrusted diadem around his head. An armlet with a similar but smaller ornament is tied around his right arm. A robe decorated with a trefoil design passes over his left shoulder and under his right arm. However, whether he represents a priest or king or both is far from certain. The same is the case with a large damaged seated figure found at Dholavira. While large

houses have been found at Harappan sites, none of them matches our idea of a palace, although it is possible that certain buildings on the citadels of cities such as Mohenjo-daro were the functional equivalent of palaces.

Jacobson (1986) suggests that the Harappan state was an early state with the following characteristics: a sovereign or sovereigns closely linked to a mythical character and seen as benevolent; a military component lacking the dominance characteristic of more mature states; and weakly developed economic stratification. According to Possehl (2003: 57), Harappan society was highly disciplined and had a strong corporate element; the Harappans may have been ruled by councils rather than kings. Kenoyer (1998: 100) suggests that the Harappan state must have comprised many competing classes of urban elites, such as merchants, ritual specialists, and those who controlled resources such as land and livestock, with different levels and spheres of control. Kenoyer also suggests that the animals on the square stamp seals represent totemic symbols standing for a specific clan, perhaps along with some additional information. At least 10 clans or communities are represented by these animals—the unicorn, humped bull, elephant, water buffalo, rhinoceros, humpless bull with short horns, goat, antelope, crocodile, and hare. The unicorn motif is found at almost all sites where the seals have been found, including in Mesopotamia. At Mohenjodaro, over 60 per cent of the seals have this motif, while it occurs on about 46 per cent of the seals at Harappa. The large number of unicorn seals at major cities led Ratnagar to suggest that the unicorn was the symbol of the Harappan ruling elite. Kenoyer, on the other hand, argues that the ‘unicorn clan’ probably represented the aristocracy or merchants who had an important executive role in the government. It is in fact the less frequent motifs such as the bull, elephant, rhinoceros, and tiger that may have been symbols of the most powerful rulers at the apex of the Harappan power structure. While Mohenjodaro stands out in some ways (for instance, no other site has a structure comparable to the Great Bath), there are other large Harappan cities such as Rakhigarhi, Lurewala, Ganweriwala, and Dholavira. Were they provincial centres knit together through a well-worked-out system of political control? Were they the capitals of separate states? Were they city-states? In the past, scholars tended to simply presume highly centralized political structures, whereas now there is a greater acceptance of the possibility of decentralization. It is not, however, certain whether we need to think in terms of a Harappan empire or a number of separate, perhaps inter-related states. Another possibility that cannot be ruled out is that there may have been several states with different kinds of political organization. The Decline of Urban Life

At some point of time, things started going wrong in the Harappan cities. Decline had set in at Mohenjodaro by 2200 BCE and the settlement had come to an end by 2000 BCE. In some places, the civilization continued till 1800 BCE. Apart from the dates, the pace of decline also varied. Mohenjodaro and Dholavira give a picture of gradual decline, while at Kalibangan and Banawali, city life ended all of a sudden (see Lahiri, 2000 for the various theories regarding Harappan decline).

A 'UNICORN' SEAL

One of the most popular explanations of the decline of the Harappan civilization is one for which there is least evidence. The idea that the civilization was destroyed by Aryan invaders was first put forward by Ramaprasad Chanda (1926)—he later changed his mind—and was elaborated on by Mortimer Wheeler (1947). Wheeler argued that references in the Rig Veda to various kinds of forts, attacks on walled cities, and the epithet puramdara (fort destroyer) given to the god Indra must have a historical basis and reflect an Aryan invasion of the Harappan cities. He identified a place called Hariyupiya in the Rig Veda with Harappa. Wheeler also pointed to certain skeletal remains found at Mohenjodaro as proof of the Aryan massacre. He subsequently modified his hypothesis, to the extent that he acknowledged that other factors such as floods, decline in trade, and over-utilization of natural resources may have had a role to play. But he insisted that the ultimate blow was given by an Aryan invasion. The Cemetery-H culture, he suggested, represented the culture of the Aryan invaders. Many scholars such as P. V. Kane (1955), George Dales (1964), and B. B. Lal (1997) have refuted the invasion theory. The evidence from the Rig Veda, a religious text of uncertain date, is far from conclusive. Moreover, if there had been an invasion, it should have left some traces in the archaeological record. There is, in fact, no evidence of any kind of military assault or conflict at any Harappan site. The 37 groups of skeletal remains at Mohenjodaro do not belong to the same cultural phase and, therefore, cannot be connected to a single event. Not one of these skeletons was found on the citadel mound, where we would have expected a major battle to have taken place. The fact that there is a sterile layer between the mature Harappan and Cemetery-H levels goes against Wheeler’s hypothesis that the latter represents the settlement of the Aryan invaders. Moreover, K. A. R. Kennedy’s analysis (1997) of the skeletal remains does not show any discontinuity in the skeletal record in the north-west at this point of time, making it clear that there was no major influx of new settlers with a different physiognomy. The Harappan civilization was not destroyed by an IndoAryan invasion. Natural disasters, not necessarily sudden or single, did have a role to play. Several layers of silt at Mohenjodaro give evidence of the city being affected by repeated episodes of Indus floods. M. R. Sahni (1956), and later Robert L. Raikes (1964) and George F. Dales (1966), argued that the floods at Mohenjodaro were the result of tectonic movements. Dales suggested that these may have

occurred at a place called Sehwan, about 90 miles downstream from Mohenjodaro, where there is evidence of rock faulting. The theory is that tectonic movements led to the creation of a gigantic natural dam that prevented the Indus from flowing towards the sea, turning the area around Mohenjodaro into a huge lake. The theory of several such episodes of flooding induced by tectonic movements is not, however, convincing. Neither is H. T. Lambrick’s hypothesis (1967), based on what he himself describes as purely circumstantial evidence, that the Indus changed its course, moving some 30 miles eastwards, starving Mohenjodaro and its inhabitants of water. While Mohenjodaro may have got worn out due to repeated episodes of naturally occurring floods, Harappan sites in the Ghaggar-Hakra valley were affected by gradual desiccation. The Sutlej or the Yamuna once flowed into the Ghaggar. Tectonic movements led to river capture—either the Yamuna joined up with the Ganga system or (what is more likely) the Sutlej was captured by the Indus, drastically reducing the water flowing into the Ghaggar. M. R. Mughal’s (1997) study of settlements in this region shows a drastic reduction in the number of sites as the river dried up. A sudden rise in the Arabian Sea coastline of west Pakistan could have caused floods and a rise in soil salinity. Such an uplift along the coast and in the lower Indus valley could also have seriously disrupted the coastal communications and trade of the Harappans. Reference has already been made to the debate on the nature of the climate, especially rainfall, in protohistoric times. On the basis of his study of pollen from Rajasthan lakes, Gurdip Singh (1971) suggests a connection between the onset of a drier climate and the decline of the Harappan civilization. However, a study of the sediments of the Lunkaransar lake indicates that the onset of drier conditions in this area may have happened well before the emergence of the Harappan civilization. Whether climatic change played a role in the decline of the Harappan civilization therefore remains unclear. The issue of environmental change can be connected to the ways in which the Harappans were treating their environment. Perhaps they were over-exploiting it through over-cultivation, overgrazing, and excessive cutting of trees for fuel and farming. This would have resulted in decreasing soil fertility, floods, and increasing soil salinity. Making estimates of population, land, food, and fodder requirements on the basis of modern data, Fairservis suggests that the civilization declined because the growing population of people and cattle could not be supported from resources within the Harappan culture zone. Shereen Ratnagar (1981) has argued that the decline in the lapis lazuli trade with Mesopotamia was a factor in the decline of the Harappan civilization. Whether this trade was particularly important for the Harappans is, however, debatable; consequently, this could not have been a factor responsible for the decline. Archaeological evidence does not give direct access to the possible social and political dimensions of the decline of the Harappan civilization. What it does indicate very clearly is that the Harappan culture underwent a gradual process of de-urbanization. The mature Harappan phase was followed by a post-urban phase, known as the late Harappan phase. The Significance of the Late Harappan Phase

There are five geographical zones of the late Harappan phase: Sindh; west Punjab and the GhaggarHakra valley; eastern Punjab and Haryana; the Ganga–Yamuna doab; and Kutch and Saurashtra. In

Sindh, the late Harappan phase is represented by the Jhukar culture at sites such as Jhukar, Chanhudaro, and Amri. The transition from the mature to the late Harappan phase in this region does not show any sudden discontinuity. There were gradual changes in the seals, a decrease in the frequency of cubical weights, and writing came to be confined only to pottery. The evidence of pottery suggests reciprocal contacts between the Jhukar culture of Sindh and the late Harappan culture at Lothal and Rangpur. In the Punjab province of Pakistan and the Ghaggar-Hakra valley, the late Harappan phase is represented by the Cemetery-H culture. There is a decline in the number of settlements from 174 in the mature Harappan phase to 50 in the late Harappan phase. In east Punjab, Haryana, and north Rajasthan, the late Harappan settlements were small compared to the mature Harappan ones. In the Ganga–Yamuna doab, compared to the 31 mature Harappan sites, there are 130 late Harappan sites. The settlements were small, houses were generally made of wattle and daub, but the agricultural base was very diverse. In Kutch and Saurashtra, there is a marked increase in the number of settlements in the earlier part of the late Harappan phase, from 18 in the mature Harappan phase to 120 in the early late Harappan phase. While there was abandonment or severe reduction in population in Sindh and Cholistan, the increase in the number of settlements in Punjab, Haryana, western Uttar Pradesh, northern Rajasthan, and Gujarat shows that this was not the case everywhere (see Chapter 5 for details). In fact, at around the time that people were abandoning Mohenjodaro, the people of Rojdi in Saurashtra were expanding and rebuilding their settlement. The data suggests an eastward and southward shift of settlements and people. The evidence from mature and late Harappan sites shows a complex interplay of elements of continuity and change. Compared to mature Harappan pottery, the slip of late Harappan pottery is less bright. The pots tend to be thicker and sturdier. Some of the classic Harappan shapes—e.g., the beaker, goblet, perforated jar, s-shaped jar, and pyriform (pear-shaped) jar—disappear. Other shapes—e.g., jars of different shapes and the dish-on-stand—continue. Various elements of Harappan urbanism such as the cities, script, seals, specialized crafts, and long-distance trade declined in the late Harappan phase, but did not completely disappear. Some of the late Harappan sites such as Kudwala (38.1 ha) in Cholistan, Bet Dwarka in Gujarat, and Daimabad (20 ha) in the upper Godavari valley can be described as urban, but they are few and far between. Graffiti on pottery occurs in Saurashtra and northern Gujarat as well as in the eastern regions. Four potsherds with Harappan letters were found at late Harappan levels at Daimabad. Some circular seals occur at Daimabad and Jhukar; rectangular seals minus motifs were found at Dholavira. A rectangular conch shell seal with the motif of a three-headed animal, similar to that found on seals of the Persian Gulf, was found at Bet Dwarka. This suggests that contact with the Persian Gulf continued in the late Harappan phase, at least in the Gujarat region. The late Harappan phase at Bhagwanpura shows flourishing specialized craft activity; there are 2 clay tablets and 19 sherds with graffiti, which could represent a script. In Punjab and Haryana, there are faience ornaments, beads of semi-precious stones, terracotta cart frames, kilns, and fire altars. A notable development in the late Harappan phase was the diversification of agriculture. At Pirak in Baluchistan, there was the beginning of double cropping—wheat and barley were being grown as winter crops and rice (with irrigation), millet, and sorghum as summer crops. In the Kachi plain, there were fairly large settlements, growing a variety of crops, supplemented with irrigation. In

Gujarat and Maharashtra, various kinds of millets were being grown as summer crops. Rice and millets were found at late Harappan levels at Harappa. Excavations at Hulas gave evidence of diverse plant remains. Grains included rice, barley, dwarf wheat, bread wheat, club wheat, oats, jowar, and finger millet. Pulses included lentil, field pea, grass pea (khesari), kulthi, green gram (moong), and chickpea. Almond and walnut shells were found, and a single carbonized seed of cotton was identified. The general picture presented by the late Harappan phase is one of a breakdown of urban networks and an expansion of rural ones. There is an overlap between the late Harappan and Painted Grey Ware (PGW) culture at sites such as Bhagwanpura and Dadheri in Haryana, and Katpalon and Nagar in Punjab. Also significant is the overlap between late Harappan and Ochre Coloured Pottery (OCP) levels in western Uttar Pradesh at sites such as Bargaon and Ambakheri. The evidence from this area, Gujarat, and north Maharashtra suggests an eastward and southward migration of the Harappans due to a combination of pressures such as those discussed in the earlier section. CONCLUSIONS

The Harappan civilization was the first urban culture in South Asia. The urban phase of the Harappan culture emerged from the proto-urban early Harappan phase. Archaeological evidence reveals a great deal about this civilization—its varied subsistence base, vibrant craft traditions, and extensive trade networks—but given the non-decipherment of the script, conclusions about many other aspects such as religion, society, and polity remain speculative. There was cultural homogeneity as well as diversity within the vast Harappan culture zone. Some of the neolithic, neolithic–chalcolithic, and chalcolithic sites mentioned in Chapter 3 were roughly contemporaneous with the Harappan civilization and interacted with it. The Harappan civilization did not come to a sudden end. The urban phase was followed by the late Harappan phase, which was marked by the decline of urban features and the diversification of agriculture.

www.pearsoned.co.in/upindersingh Further resources

Chapter Five

Cultural Transitions: Images from Texts and Archaeology, c. 2000– 600 BCE Chapter outline PERSPECTIVES FROM TEXTS ARCHAEOLOGICAL PROFILES OF DIFFERENT REGIONS OF THE SUBCONTINENT, C. 2000– 500 BCE THE PROBLEM OF CO-RELATING LITERARY AND ARCHAEOLOGICAL EVIDENCE CONCLUSIONS

MEGALITHIC BURIAL, HIRE-B ENKAL (KARNATAKA)

Janaka, king of Videha, was performing a great sacrifice, and Brahmanas had come from far and wide to attend. The king announced a prize of 1,000 cows with 10,000 gold pieces fastened to their

horns for the wisest among all the assembled Brahmanas. At this, sage Yajnavalkya asked his pupil Shamashravas to herd the cows home. The other Brahmanas grew furious at his presumption and an intense philosophical contest ensued. One by one, eight interlocutors posed a series of questions to Yajnavalkya on matters related to the sacrifice, the senses, the worlds to which great men departed, the nature of the atman, the making of the universe, and the resting places of the gods and spirits. One of the interlocutors was a woman named Gargi. As her questions built up to a crescendo, Yajnavalkya thundered at her to stop or else her head might fall off. Gargi retreated, but spiritedly subjected the sage to a second round of queries. Vidagdha, the last questioner, had to pay the price of defeat with his head. All had been silenced by Yajnavalkya’s brilliant responses.

This episode is narrated in the Brihadaranyaka Upanishad, a text belonging to the Vedic corpus. Is there a historical basis to this incident? Did a great sage named Yajnavalkya ever exist? Did a woman named Gargi participate in a philosophical quest dominated by men? Was the price of defeat in such contests really death? How many people were actually interested in such esoteric issues? It is difficult to answer such questions with certainty, but the episode does conjure a dramatic scene of philosophical inquiry in which the stakes were very high—of reputation and life itself. The poets who composed the Vedic hymns of praise and supplication to the gods and the priests who explained how the rituals were to be performed were not historians. Vedic texts are religious and ritualistic works, not works of history. However, combined with the available archaeological evidence, they can be used as sources of information on various aspects of the life of people living in the greater Indus valley, the Indo-Gangetic divide, and the upper Ganga valley in the 2nd and 1st millennia BCE. When discussing this period, most accounts of ancient Indian history make a decisive shift from a narrative based on archaeology to one based on Vedic texts. In general, archaeological evidence is cited only when it supports what the texts seem to be suggesting. This approach has resulted in an undue focus on the northern and north-western regions of the subcontinent and a neglect of other areas. It has led to the sidelining of substantial archaeological evidence from neolithic–chalcolithic, chalcolithic, and early iron age cultures that tells us about the lives of ordinary people living in the various regions of the subcontinent during c. 2000–500 BCE. The challenge is to incorporate both literary and archaeological evidence, wherever they are available. However, evidence from these two sources does not always match. When dealing with material culture, priority should be given to archaeological evidence. Vedic literature, on the other hand, is a richer source of information on the development of philosophical concepts and religious ideas and practices. Another challenge is to explore and expand the historical potential of the archaeological evidence from regions for which no texts are available, and where archaeology remains the only window into the past. In order to view the complex historical jigsaw puzzle of the subcontinent in c. 2000–500 BCE, it is necessary to carefully juxtapose the archaeology-based and text-and-archaeology-based profiles of the various regions, recognizing that in some cases, the pieces do not fit together perfectly. Perspectives from Texts USING THE VEDAS AS A HISTORICAL SOURCE

Extracting history from a literature as ancient, vast, and complex as the Vedas is no easy task. Unfortunately, critical editions identifying the original core of the texts are not available. The 19th century translations cannot be relied upon, and recent authoritative translations, whether in the European or Indian languages, are few. A great deal depends on the interpretation of words and phrases, whose meanings may vary from one text and context to another. The Vedic corpus was not a popular literature and, therefore, does not necessarily represent popular ideas or practices. It was composed, preserved, and transmitted by and for a section of the Brahmanas. (Here, the reference is to Brahmanas as a social group. The Brahmanas are also a category of Vedic texts.) The texts were transmitted orally for many centuries and it is not certain when they were first written down. The earliest surviving manuscripts belong to the 11th century CE. Many historians use a rough chronology of c. 1200–1000 BCE or 1500–1000 BCE for the composition of the earliest sections of the Rig Veda. It is possible that parts of the Rig Veda were composed even earlier, perhaps in c. 2000 BCE, but there are limits to how far back its dates can be pushed. The uncertainty of the period of composition of the Rig Veda is a major problem in using this text as a source of history. Books 2–7, the oldest books of the Rig Veda Samhita, are also known as the family books because their composition is attributed to the families of certain seer-poets—Grit-samada, Vishvamitra, Vamadeva, Atri, Bharadvaja, and Vasishtha. Books 1, 8, 9, and 10 seem to be of a later period. The hymns of this Samhita are arranged in a precise pattern. In the family books, they are arranged according to deity, number of stanzas, and metre. The number of hymns increases in each successive book. Within a particular book, the hymns are arranged in groups according to deity—first come the hymns to Agni, then Indra, and then the other gods. And within a group of hymns addressed to a particular deity, the arrangement follows a pattern of a decreasing number of stanzas per hymn (i.e., the preceding hymns have more stanzas than the succeeding ones). In instances where two hymns have the same number of stanzas, the hymn which is in a metre requiring more syllables is placed first. The arrangement of hymns in the other books of the Rig Veda Samhita follows a different, but recognizable order. The pattern of arrangement makes it possible to detect interpolations. Hymns that disrupt the pattern must have been added to the collection later. This does not necessarily mean that they were later in terms of their period of composition. The ‘later’, i.e., less old books of the Rig Veda Samhita may actually contain some very old hymns, and the ‘earlier’ books contain some not-so-old hymns. Sometimes, certain hymns are assigned a later date because their content or ideas seem different. However, such differences could be due to their originating in a different milieu or reflecting different ideas current at the time. The deliberate, careful arrangement of the hymns of the Rig Veda Samhita was the work of its compilers. The language, and possibly also the content, of the hymns may have been modified in the process of compilation, which may have taken place in c. 1000 BCE. The Vedas may have been arranged and compiled because of the desire of priests to create an authoritative text for the sacrifices they performed. We know from other sources that there were various recensions of the Rig Veda, which may have differed from each other in content, arrangement, and traditions of interpretation. Of these recensions, only the Shakala has survived into our own time. Vedic texts can be used as sources of history for the areas in which they were composed. The family books of the Rig Veda Samhita were composed in eastern Afghanistan and the Punjab, the

land of Sapta-Sindhu or the seven rivers. The rivers in question were the Indus, its five tributaries, and the Sarasvati (which can probably be identified with the modern Ghaggar-Hakra). The core geographical area of later Vedic texts was Kuru– Panchala, which comprised the Indo-Gangetic divide and the upper Ganga valley.

SEE CHAPTER 1, PP . 17–18 FOR DETAILS ON THE VEDIC CORPUS

PRIMARY SOURCES

The date of the Rig Veda

The dates suggested for the composition of the Rig Veda range from c. 6000 BCE to 1000 BCE. The chronology of c. 1200–1000 BCE for the family books of the Rig Veda is based on the tentative dates put forward by the German Indologist Max Müller in the 19th century. He worked backwards from dates of later texts to arrive at c. 1200 BCE for the beginnings of Vedic poetry. The reasoning he used is as follows: The 1. Vedanga and Sutra works were roughly contemporary with early Buddhism, so they can be dated c. 600– 200 BCE. As Vedic literature is older than Buddhist literature, it must have been composed before the 6th century BCE. Going 2. by the lists of teachers and other contents of the Vedic Brahmana texts, it can be assumed that the composition of these texts (i.e., the Brahmanas) must have stretched over at least 200 years before 600 BCE. That would mean a time bracket of c. 800–600 BCE for the Brahmanas. The 3. Vedic Samhitas are older than the Brahmanas. Their composition must also have stretched over about 200 years, i.e., c. 1000–800 BCE. The 4. Vedic hymns must have evolved over about 200 years. This suggests c. 1200 BCE as the date for the beginnings of the composition of Vedic poetry.

Max Müller suggested this chain of reasoning only as a way of arriving at a rough date for the Rig Veda. Several Indologists such as H. H. Wilson, G. Bühler, H. Jacobi, and Maurice Winternitz questioned the assigning of 200 years (and not more) for the composition of various categories of texts. Winternitz thought that the Rig Veda was probably older than 1200 BCE. He suggested that the beginning of Vedic literature should be placed closer to 2500 or 2000 BCE, but added that he would prefer not to give any dates at all. Max Müller accepted the criticism provoked by his hypothesis, but reminded his critics that his dates were meant to be hypothetical and provisional. Astronomical references in the Rig Veda have been used to date the text, but have given different results. For instance, Ludwig concluded that the text was composed in the 11th century BCE,

while Jacobi arrived at a 3rd millennium BCE date. Recently, Subhash Kak (2001) has argued that the astronomical references in the Rig Veda can be dated c. 4000–2000 BCE. A 1380 BCE inscription found at Bogaz Koi in north-eastern Syria records a treaty between a Hittite and a Mitanni king. It mentions the gods Indara (Indra), Mitras (Mitra), Nasatia (Nasitya, i.e., the Ashvins), and Uruvanass (Varuna)—deities who are mentioned in the Rig Veda. While a majority of the Mitanni people spoke the local Hurrian language, the inscription indicates that their rulers had Indo-Aryan-sounding names and invoked Indo-Aryan gods. Belonging to about the same period is a Hittite text on horse training and chariotry, written by a Mitannian named Kikkuli. This uses several technical terms which resemble Indo-Aryan ones. While these inscriptions are relevant for the history of the Indo-Aryan languages and gods, they do not give direct or definite information about the date of the Rig Veda. There are close similarities between the language and culture reflected in the Rig Veda and an ancient Iranian text called the Avesta. This could be an important clue to dating the Rig Veda, but unfortunately, the dates of the Avesta are not certain. Its oldest parts may go back to c. 1500 BCE. Very early dates for the Rig Veda that fall within the 7th or 6th millennium BCE are clearly not acceptable. One reason is that we know from archaeology that the north-western part of the subcontinent was at that time still in the stone age, and the Rig Veda clearly belongs to the chalcolithic age. Dates falling within the late 3rd millennium BCE or the early 2nd millennium BCE (calculated on the grounds of philology and/or astronomical references) cannot be ruled out. The date of the Rig Veda remains a problematic issue.

Many different kinds of histories of the Indo-Aryans have been derived from the Vedas. Nationalist historians extracted historical details from the texts but tended to idealize the Vedic age (Altekar [1938], 1991; Majumdar et al. [1951], 1971). A subsequent trend was more dispassionate in approach, but concentrated on fitting data from the texts into long-term unilinear historical and anthropological models (R. S. Sharma, 1983; Thapar, 1990). Recent studies (e.g., Witzel, 1997a, 1997b) offer a more nuanced textual analysis. Nevertheless, when we talk of the ‘Vedic age’ or ‘Vedic culture’, we must be conscious of the problem of dating the Rig Veda, the religious and elite nature of the texts, their specific geographical contexts, and the availability of substantial archaeological data for these and other regions. WHO WERE THE INDO-ARYANS?

The use of Vedic literature as a source of history is linked to a number of questions about the people to whom these texts belonged. Who were the Indo-Aryans? Where did they come from? What was the relationship between the Vedic and Harappan cultures? These issues have not always been treated as purely academic ones. They have political implications, and have been used to serve diverse political agendas, both in colonial and post-colonial times (see Trautmann, 2005). And in spite of vigorous and often volatile debate spanning over two centuries, there are still no definite answers.

During the 19th and early 20th centuries, when large sections of Africa and Asia were colonized by European nations, many scholars thought about history in terms of the movement and interaction of different races. Some scholars used the term ‘race’ loosely in the sense of an ethnic or cultural group. However, another trend was to classify people of the world into different races such as Caucasian, Mongoloid, Negroid, etc. on the basis of physical and other characteristics. These classifications seemed to be objective and scientific on the surface, but most of them were racist. They provided a pseudo-scientific justification for the European subjugation of Asian and African people, whom they presented as inferior races. The theory of a superior white, blond-haired, and blue-eyed Aryan race, which was a part of Nazi propaganda in 20th century Germany, is a myth and is not based on historical facts. This is the case with all theories that claim that a particular group of people are inherently superior to others. Today, most anthropologists have abandoned racial classifications. There is no doubt that people living in different parts of the world look different. But the old, prejudiced category of race, which presented people in different parts of the world as separate, unrelated, and unchanging entities, frozen in time, has been replaced by more meaningful and objective ways of classifying and understanding human cultures. The composers of the Rig Veda described themselves as arya, which can be understood as a cultural or ethnic term. The word literally means kinsman or companion, or it may be etymologically derived from ar (to cultivate). The terms ‘Indo-European’ and ‘Indo-Aryan’, as used by linguists and historians, have nothing to do with racial classifications. They are linguistic terms, referring to families of languages and their speakers. The Indo-Aryans were the speakers of a sub-group of the Indo-Iranian branch of the Indo-European family of languages. The original homeland of the Indo-Europeans and Indo-Aryans is the subject of continuing debate among philologists (scholars who study old languages), linguists, historians, archaeologists, and others. The dominant view is that the Indo-Aryans came to the subcontinent as immigrants. Another view, advocated mainly by some Indian scholars, is that they were indigenous to the subcontinent. Over the years, many original homelands have been proposed for the Indo-Aryans (see Bryant, 2002). These include Tibet, Afghanistan, Iran, the Aral Sea, the Caspian Sea, the Black Sea, Lithuania, the Arctic, the Caucasus, the Urals, the Volga mountains, southern Russia, the central Asian steppes, West Asia, Turkey, Scandinavia, Finland, Sweden, the Baltic region, and India. All these claims are not supported by equally convincing evidence, and none of them is free from problems. One of the more widely accepted views locates the original homeland of the IndoEuropeans in the plains of Eastern Europe, especially the area north of the Black Sea. The Vedas reflect a close connection with Iran. But we do not know when, where, or why the Indo-Iranians and Indo-Aryans parted ways. Today, most historians have discarded the idea of an Aryan invasion of the Indian subcontinent in favour of a theory of several waves of Indo-Aryan migrations. However, there is no consensus on the routes or timing of these migrations. The IndoAryan languages of India include the non-Sanskritic or Dardic languages spoken in the mountains of the north-west, which may represent an earlier wave of Indo-Aryan immigrants. Superior military technology and the use of the horse and chariot may have given the immigrants the crucial initial advantage, enabling them to establish their political dominance in the land of the seven rivers. THE CULTURE REFLECTED IN THE FAMILY BOOKS OF THE RIG VEDA SAMHITA

Historians divide the Vedic corpus into two parts—early and later Vedic texts, although recent studies indicate a more complex internal chronology. Early Vedic literature refers to the family books of the Rig Veda Samhita. Later Vedic literature includes Books 1, 8, 9, and 10 of the Rig Veda Samhita, the Samhitas of the Sama, Yajur, and Atharva Vedas, and the Brahmanas, Aranyakas, and Upanishads attached to all the four Vedas. (Among these later texts, the Mantra portions are the earliest, followed by the Brahmanas, Aranyakas, and Upanishads.) The cultural stages reflected in the two broad strata of early and later Vedic texts have come to be known as the early and the later Vedic cultures. The principal Shrautasutras and some of the early Grihyasutras have been dated c. 800–400 BCE.1 These texts will, however, be discussed in the next chapter. TRIBES AND WARS

The Rig Veda is pervaded with the aura of warring tribes. About 30 tribes and clans are mentioned. Five tribes—the Yadu, Turvasha, Puru, Anu, and Druhyu—are collectively known as the ‘five peoples’ (pancha-jana, pancha-kristhya, or pancha-manusha). The Purus and Bharatas are the two dominant tribes. Initially, they seem to have been allies, but at some point, they fell apart. The Rig Veda mentions a chief of the Purus named Trasadasyu. It also mentions a famous Bharata king named Divodasa and describes his victory over the Dasa ruler Shambara, who had many mountain fortresses. Many Rig Vedic hymns beseech the gods for victory in battle. It is difficult to distinguish between mythical and historical events, between demons and real enemies. There are several references to conflicts with the Dasas and Dasyus. One view is that these were the aboriginal people encountered by the Indo-Aryan tribes. However, they may actually represent earlier (pre-Vedic) waves of IndoAryan immigrants. Prayers to Indra to defeat not only the Dasa but also the Arya enemies indicate that there were conflicts among the Aryas too. There are about 300 clearly non-Indo-European words in the Rig Veda. These ‘loan words’ show that the Rig Vedic people were interacting with people speaking Dravidian and Munda languages. There are many tribes with non-Indo-Aryan names in the Rig Veda, such as the Chumuri, Dhuni, Pipru, and Shambara. The text also refers to Arya chieftains with non-Indo-Aryan names, e.g., Balbutha and Bribu. All this is indicative of processes of cultural interaction. The ‘battle of ten kings’ (dasharajna), recounted in Book 7 of the Rig Veda Samhita may be based on an actual historical incident. In this battle, the Bharata chief Sudas, grandson of Divodasa, fought against a confederacy of 10 tribes. The mention of the Purus, their former allies, as a part of this confederacy indicates that political alliances were fluid and shifting. Vishvamitra, the Bharata purohita, seems to have been replaced by Vasishtha before the battle, reflecting another sort of behind-the-scenes re-alignment. The great battle took place on the banks of the river Parushni (Ravi). The Bharatas won by breaking a natural dam on the river. Marching on to the Yamuna, they defeated a local ruler named Bheda. Sudas eventually settled down along the Sarasvati and celebrated his victory and position of political paramountcy by performing the ashvamedha sacrifice. The word rajan (or raja) occurs many times in the family books of the Rig Veda. Since a fullfledged monarchical state had not yet emerged, this word is best translated as ‘chieftain’ or ‘noble’, rather than as ‘king’. It is not always clear from the hymns whether the rajan was the chief of a tribe, clan, clan segment or several clans. But his main task was to protect his people and to lead them to victory in war. The reference to the chieftain as gopa or gopati (lord of the cattle) indicates that

protecting and increasing the cattle herd was his other major role. The royal priest accompanied the rajan to battle, recited prayers, and supervised the performance of rituals. The importance of royal priests such as Vasishtha and Vishvamitra is reflected in many Vedic hymns. Bali refers to an offering made to a god; it also means tribute periodically offered by the clansmen to the rajan. Tribute was no doubt also extracted from tribes defeated in battle. A regular taxation system had not yet emerged. PRIMARY SOURCES

Hymn to arms (Rig Veda Samhita 6.75)

The following benediction was recited by the purohita (royal priest) either before the chieftain set out on a military expedition or in order to bless the warriors accompanying the consecrated horse in the ashvamedha sacrifice. Note how the various weapons are described and praised, one by one: His face is like a thundercloud, when the armed warrior goes into the lap of battles. Conquer with an unwounded body; let the power of armour keep you safe. With the bow let us win cows, with the bow let us win the contest and violent battles with the bow. The bow ruins the enemy’s pleasure; with the bow let us conquer all the corners of the world. She [the bow] comes all the way up to your ear like a woman who wishes to say something, embracing her dear friend; humming like a woman, the bowstring stretched tight on the bow carries you safely across in battle. These two [the bow tips] who go forward like a woman going to a rendezvous, hold the arrow in their lap as a mother holds a son. Let the two bow-tips, working together, pierce our enemies and scatter our foes. He [the quiver which holds the arrows] is the father of many daughters [arrows], and many are his sons [arrows]. He makes a rattling sound as he goes down into battle. The quiver wins the attacks and all the skirmishes when he is strapped on a back and set to work. Standing in the chariot, the skilful charioteer drives his prize-winning horses forward wherever he wishes to go. Praise the power of the reins: the guides follow the mind that is behind them. Neighing violently, the horses with their showering hoofs outstrip everyone with their chariots. Trampling down the foes with the tips of their hoofs, they destroy their enemies without veering away. Spare us, O weapon flying true to its mark; let our body be stone. Let Soma speak a blessing upon us; let Aditi give us shelter.

He beats them on the back and strikes them on the haunches. O whip for horses, drive forward into battle the horses who sense what is ahead. It wraps itself around the arm like a serpent with coils, warding off the snap of the bowstring. Let the gauntlet [the leather protecting the forearm], knowing all the ways, protect on all sides, a man protecting a man.... Once shot, fly far away, arrow, sharpened with prayer. Go straight to our foes, and do not leave a single one of them there…. I cover with armour those places on you where a wound is mortal. Let Soma the king dress you in ambrosia (or immortality). Let Varuna make wider yet your wide realm. Let the gods rejoice in you as you are victorious. Whoever would harm us, whether it is one of our own people, or a stranger, or someone from far away, let all the gods ruin him. My inner armour is prayer. SOURCE O’Flaherty, 1986: 236–38

The Rig Veda mentions assemblies such as the sabha and samiti. The distinctions between their functions are not entirely clear. The sabha seems to have been a smaller, more elite gathering, whereas the samiti appears to have been a larger assembly presided over by the rajan. Such assemblies may have played an important role in the redistribution of resources. Hymns express the desire for harmony among members (‘Assemble, speak together; let your minds be all of one accord.’). The vidatha has been understood as a tribal assembly with diverse functions. However, it actually seems to refer to a local congregation of people meeting to perform socio-religious rituals and ceremonies for the well-being of the settlement. The family books contain several terms for socio-political units, many of which were based on kinship. These include jana, vish, gana, grama, griha, and kula. Their precise meaning, however, is not always clear. The jana of the Rig Veda can be translated as tribe, vish is often translated as people in general or as clan, and gana as lineage. Grama, which later came to mean village, seems to have originally referred to a mobile group of people who may or may not have been related to each other through kinship. KEY CONCEPTS

Lineage, clan, tribe

Historians use several sociological terms and concepts while describing ancient cultures. Kinship refers to socially and culturally recognized relationships among people, commonly assumed to be based on natural or biological ties. These ties may be based on birth/descent (consanguinal relations), marriage (affinal relations), adoption, or fosterage. There are also other culturally specified kinds of kinship—e.g., in north India, there is the custom of the rakhi

brother–sister relationship and the ‘muh-bola-bhai’ (a man declared to be a brother). Kinship is so important in Indian society that its language has spread far and wide. Younger people routinely address their elders as ‘uncle’ and ‘aunty’ and people who are not even remotely related may address each other as ‘brother’, ‘sister’, ‘mother’, or ‘father’. Kinship systems can be unilineal or multi-lineal. Unilineal kinship systems which recognize descent relationships through the father are known as patrilineal or agnatic. Unilineal kinship systems which recognize descent through the mother are known as matrilineal. Multi-lineal or cognatic systems are those in which descent through both the mother and father is recognized. In both patrilineal and matrilineal systems, relationships through the other parent also receive recognition for different purposes at different times—for instance, at times of marriage, during the performance of rituals, and even in matters of inheritance. For example, in a patrilineal society, a son or daughter may inherit property from their mother’s kin, and the mother’s brother may have a significant role to play in the lifecycle rituals of his sister’s children. A lineage is a group of unilineal kin. In view of the problem of drawing the dividing line between family and lineage, the latter term can be used to refer to relations beyond the three or four generation family. Several unilineal descent groups who trace their descent from a common ancestor, actual or mythical, form a clan. Members of a clan sometimes claim a common place of origin and may have clan property or a clan god. A number of related clans constitute a tribe. ‘Tribe’ is a problematic term. It has often been used by anthropologists to refer to people considered primitive, living in economically less-developed areas, and lacking a script. These days, sociologists are careful to avoid value-laden terms such as ‘primitive’ and are aware of the pitfalls in defining a tribe. André Beteille ([1960], 1977) suggests that a tribe can be defined as a society with a political, linguistic, and somewhat vaguely defined cultural boundary, based on kinship, and lacking in social stratification. Within this very general definition, tribes differ from one another in many ways. In the context of early Indian history, historians often use the term ‘tribal’ to refer to pre-chiefdom and pre-state societies. Others prefer to avoid the use of the term altogether. PASTORALISM, AGRICULTURE, AND OTHER OCCUPATIONS

Animals such as horses, goats, and sheep are mentioned in the family books, but cattle were clearly prized the most. R. S. Sharma (1983: 24) has drawn attention to the many derivations of the word gau (cow) in the Rig Veda. Words for war with the infix gau—such as gavishti, gaveshana, goshu, and gavya—suggest that many battles were in effect cattle raids. Further indications of the importance of cattle come from other words containing the gau infix. The tribal chief was known as janasya gopa. Measures of time included godhuli (dusk) and samgava (morning), measures of area/distance included gavyuti and gocharman. The buffalo was known as gauri or gavala. The daughter was duhitri (she who milks cows). Gojit (winner of cows) was a word for a hero. A wealthy person was known as gomat (owner of cattle). One of the epithets of the god Indra was gopati (lord of cattle).

Some scholars have used the number of references to pastoral versus agricultural activities in the family books as an index of their relative importance, and have concluded that while cattle rearing was of overwhelming importance, agriculture was either a subsidiary activity or one that was practised by non-Indo-Aryans. However, the frequency of usage in religious or ritualistic texts and contexts may not be an accurate indicator of the relative importance of these activities in everyday life. Apart from word frequencies, it is necessary to examine the nature and content of the references. R. N. Nandi (1989–90) has drawn attention to the many references to agricultural activity in the Rig Veda and argues that it was by no means marginal. The verbs vap (to sow) and krish (to cultivate) occur, along with references to various agricultural implements. Phala, langala, and sira are words for the plough, which must have been made of wood. Other implements included the hoe (khanitra), sickle (datra, srini), and axe (parashu, kulisha). The word kshetra has a range of meanings, including a cultivated field. Hymns refer to the levelling of fields for cultivation, the desire for fertile fields (urvara), and furrows (sita) drenched by rain, producing rich harvests. The only terms for cereals are yava (barley or a generic term for cereal) and dhanya (a generic term for cereals). There are references to seed processing, food prepared from cereals, and large jars that were probably used to store grain. Some hymns refer to conflicts among people for the protection of sons, grandsons, cattle, water courses, and fertile fields. Prayers to Indra beseech him to grant or enrich the fields. This god is described as the protector of crops, winner of fertile fields (urvarajit), and one who showers such fields on those who perform sacrifices to him. The later parts of the family books invoke Kshetrapati, who seems to have been a guardian deity of agricultural fields. Wars were fought for cattle, but also for land. Hymns refer to warriors, priests, cattle-rearers, farmers, hunters, barbers, and vintners. The crafts mentioned include chariot-making, cart-making, carpentry, metal working, tanning, the making of bows and bowstrings, sewing, weaving, and making mats out of grass or reeds. Some of these occupations and crafts may have been the jobs of full-time specialists. There are hardly any references to metallurgical activities in the Rig Veda, and very few of these occur in the family books (see Chakrabarti, 1992). The word ayas occurs in several contexts. There are references to Indra’s thunderbolt of ayas; the chariot of Mitra and Varuna having columns of ayas; and the home of Indra and Soma made of ayas. A hymn to Agni compares his splendour to the edge of ayas. Another hymn to Agni beseeches him to be like a fort of ayas to his worshippers. A prayer to Indra asks him to sharpen his worshipper’s thought as if it were a blade of ayas. The family books also refer to the Dasyus’ cities of ayas, forts of ayas, a horse’s jaws of ayas, a vessel of ayas. The few metal objects mentioned in the Rig Veda are kshura (razor), khadi (maybe a bangle), and asi/svadhiti (axe). But it is not clear precisely which metal these objects were made of. A hymn (4.2.17) refers to the doers of good deeds having freed their birth from impurity in the same way as ayas is purified. The medieval commentator Sayana explains this reference as follows: ‘As the smiths heat metal using bellows.’ There are a few references in the Rig Veda to the words dham and karmara, but these occur in the late books 9 and 10, and it is far from certain whether they refer to iron-welding or iron smiths. Some scholars have interpreted the references to ayas, metal objects, and metallurgical activity in the Rig Veda as indicative of iron artefacts and iron working. However, there is no definite evidence that this was so. There is in fact no clear or conclusive reference to iron in the family books. Ayas could have meant copper, copper-bronze, or may have been a generic term for metals.

Anthropological studies have brought out the importance of gift exchanges in simple societies, and some of their observations are useful for understanding the culture reflected in the Rig Veda. In his classic work on the gift, Marcel Mauss [1954], 1980) pointed out that such exchanges may appear on the surface to be voluntary and spontaneous, but are actually strictly obligatory and governed by conventions that have to be observed. It is not the individual but groups (families, clans, tribes) who make the exchanges and are bound by their obligations. Such exchanges—known as prestations—do not only involve material goods of economic value. They also involve the exchange of other things such as courtesies, entertainments, military assistance, ritual, women, children, dances, feasts, and hospitality. The rules of the game in gift exchange are different from the logic that operates in ordinary sorts of economic exchanges. The offering, receiving, and reciprocating of gifts are acts that establish and cement social relationships and social hierarchies. In the Rig Veda, we have noted that gifts (bali) were received by the rajan from members of the clan. Priests received dana (ritual gifts) and dakshina (sacrificial fees) at the conclusion of sacrificial rituals. Gift-giving and receiving do not rule out other kinds of exchange, but trade in the Rig Vedic context was probably minimal. Barter was the mode of exchange and cattle an important unit of value. The word nishka seems to have meant ‘a piece of gold’ or ‘gold necklace’, and there is no indication of the use of coins. There are prayers to the gods to ‘give broad paths to travel’ and ensure a safe journey. Mention is made of chariots and carts drawn by oxen, mules, or horses. The panis (literally, ‘those who possess wealth’) in some instances refer to merchants and in others to stingy people who did not perform sacrifices and hid their wealth. There are references to boats (nau) and the ocean (samudra). Rig Veda 1.116.3 refers to the Ashvins rescuing Bhujya in the ocean with the help of a ship with a hundred oars (shataritra). Book 10 refers to the eastern and western oceans. But both Books 1 and 10 are later books, and historians differ on whether or not the composers of the early sections of the Rig Veda were familiar with sea travel, let alone sea trade. War booty was a major source of wealth (pana, dhana, rayi, etc.). The references to wealthy people and those worthy of attending the assemblies suggest differences in wealth and rank. The rajan and the assemblies must have had a say in the redistribution of war booty, and the rajan and his immediate kinsmen must have got a larger share. Apart from cattle, other items solicited in prayers and sacrifices include houses, horses, gold, fertile fields, friends, plentiful food, wealth, jewels, chariots, fame, and children. The notion of individual private property ownership as we understand it—associated with the right to buy, sell, gift, bequeath, and mortgage—did not exist. The clan as a whole enjoyed rights over major resources such as land and herds. The household was the basic unit of labour, and there is no mention of wage labour. The Rig Veda is, however, familiar with slavery. Slavery, is an extreme form of social subordination. A slave, whether male or female, has no rights, power, autonomy, or honour, is considered the property of the master, and is obliged to perform all kinds of services, no matter how menial. The Rig Veda refers to enslavement in the course of war or as a result of debt. The fact that in later times, dasa and dasi are terms used for male and female slaves, suggests that initially, ethnic differences may have been an important basis of enslavement. Slaves, male and female, generally worked in the household, but were not used to any significant extent in production-related activities. As pointed out by Gerda Lerner (1986), in all cultures, throughout history, there was an important difference in the experience of enslavement for men and women—for women, enslavement generally involved sexual exploitation in addition to exploitation of their labour.

Although the family books reflect differences in rank and some inequalities in wealth, these do not add up to distinct socio-economic classes in the sense of significant differences in access to and control over basic productive resources. However, the absence of a class hierarchy does not mean that Rig Vedic society was egalitarian. The family books reflect inequalities between masters and slaves, and between men and women. The rajan stood at the top of the ladder of political and social power and status, the dasi stood at the very bottom. The Rig Veda mentions food and drink, clothes, and leisure-time pursuits of people. There are references to the consumption of milk and milk products, ghrita (ghee, clarified butter), grains, vegetables, and fruits. Vedic texts refer to meat eating, and to the offering of animals such as sheep, goat, and oxen to the gods in sacrifice (Majumdar et al. [1951], 1971: 396, 461). However, the reference to cows as aghnya (not to be killed) suggests a disapproval of their indiscriminate killing. This issue has sometimes become controversial in view of the sanctity that eventually came to be associated with the cow in Hinduism. However, it should be remembered that religious and dietary practices have always varied considerably over time and space. The drink known as soma consisted of the juice of the soma plant, mixed with milk, sour milk, or yava (cereal). Sura seems to have been an intoxicating drink made out of fermented grain. People wore clothes of cotton, wool, and animal skin, and donned a variety of ornaments. There are references to singing and dancing, and to musical instruments such as the vina (lute), vana (flute), and drums. Dramas may have been a source of entertainment, and chariot racing and gambling with dice were popular pastimes. VARNA IN THE RIG VEDA

The word varna occurs in many places in the family books and usually means light or colour. However, in some passages, it is associated with the Aryas and Dasas. The fact that similar epithets are applied to Dasas and Dasyus, and that both these terms are used to describe certain enemies, indicate an overlap in their connotations. The Rig Veda describes them as a-vrata (people who do not obey the ordinances of the gods) and a-kratu (those who do not perform sacrifices). Another adjective used for them is mridhra-vacha. This can be interpreted in different ways—as referring to their speech being indistinct, unclear, soft, unintelligible, uncouth, hostile, scornful, or abusive. The fact that this epithet is used in one place for the Purus, an Indo-Aryan tribe, makes it unlikely that it meant ‘unintelligible’. In three places in the Rig Veda, the term krishna-tvach or asiknitvach is applied to the Dasyus. This can be interpreted literally as ‘dark skinned’, or as a figurative use of darkness. In one passage, the Dasas are described as anasa. Whether this means noseless (i.e., flatnosed), faceless (in some metaphorical sense) or mouthless (i.e., whose speech is incomprehensible) is uncertain. The old view highlighted the supposed physical differences, and described the Dasas and Dasyus as the dark-skinned, flat-nosed aboriginal people of India who were displaced and pushed southwards by the fair-skinned Aryans. The references cited above should make it clear that the epithets used for the Dasas and Dasyus can be interpreted in different ways. Whether or not there were stark differences in physical appearance can be debated. What is certain is that there were a range of cultural differences, including those of religious practice, and possibly in mode of speech, language, or dialect. Many scholars think that the Dasas and Dasyus were not non-Aryan tribes but earlier waves of Indo-Aryan immigrants who arrived in the subcontinent before the Vedic Aryans. A connection has been suggested between an Iranian tribe called the Dahae and the Dasas of the Rig

Veda, and between the Dahyu tribe and the Dasyus. Although the Rig Veda talks of conflicts between the Aryas and the Dasas and Dasyus, there were also conflicts and military engagements among the Indo-Aryan tribes as well—the conflict between the Bharatas versus the Purus and their allies in the ‘battle of ten kings’ is a case in point. The words ‘Brahmana’ and ‘Kshatriya’ occur frequently in the family books, but the term varna is never associated with them. There is mention of Brahmanas drinking soma and reciting hymns, and although they seem to have been a group who enjoyed respect, there are no indications that membership of this group was based on birth. The words ‘Vaishya’ and ‘Shudra’ are absent. The earliest reference to the division of society into four strata occurs in the Purusha-sukta, a hymn in Book 10 of the Rig Veda Samhita. As this is a later book, the four-fold varna order is seen as a feature of later Vedic texts. The absence of a strict social hierarchy and the existence of an element of social mobility is suggested in Rig Veda 3.44–45. In this hymn, the poet asks Indra: ‘O, Indra, fond of soma, would you make me the protector of people, or would you make me a king, would you make me a sage who has drunk soma, would you impart to me endless wealth?’ This suggests that a man could aspire to different sorts of vocations and goals in life. WOMEN, MEN, AND THE HOUSEHOLD

Nineteenth-century socio-religious reformers and nationalist historians of the early 20th century often presented the Vedic age as a golden age for women. They pointed out that the Vedic people worshipped goddesses; the Rig Veda contains hymns composed by women; there are references to women sages; women participated in rituals along with their husbands; they took part in chariot races and attended the sabha and various social gatherings. Such a presentation of the ‘high’ position of women in Vedic society can be seen as a response to the oppression and humiliation of colonial rule. The idea was to show that in ancient times, Indians were better than the Westerners, at least in the way they treated women. This could also be used as an argument to improve the prevailing condition of women in Indian society (see Chakravarti, 2006). Recent scholarship has shifted the focus from discussing women in isolation to an analysis of gender relations. Gender refers to the culturally defined roles associated with men and women. Earlier, historians tended to focus on the public, political domain, relegating the family, household, and gender relations to the private, domestic domain. Today, the distinction between the private and political domains is recognized as an artificial one. Ideologies and hierarchies of power and authority exist within the family and household, in the form of norms of appropriate conduct based on gender, age, and kinship relations. Further, there is a close connection between relations within the household, marriage and kinship systems, the control of women’s sexuality and reproduction, class and caste relations, and larger political structures. These are all like the interlocking building blocks of a vast and complex social pyramid. For these reasons, gender relations form an important part of social history. The experience of women belonging to different groups in society varied, and it is therefore necessary to break down the category of ‘women’ into more specific subcategories based on rank, class, occupation, and age. Women have to be understood in relation to men, and their relationships are embedded in wider social, economic, and political contexts. For all periods, the vague issue of the ‘status of women’ therefore has to be dissolved into smaller, more meaningful questions, such as:

What were the relations between men and women in the domestic sphere? How was a person’s descent recognized? What were the norms of property and inheritance? What was the role of women in production-related activities? Did they have control over these activities or the fruits of their labour? How was the sexuality and reproductive potential of women controlled and regulated? What was the role of women in the religious and ritual spheres? Did they have access to education and knowledge systems? Did they have direct or indirect access to political power? Further, structures of subordination and control were not total or all-encompassing, and an analysis of gender relations has to move beyond seeing women as passive victims of oppressive social structures. In spite of their subordination, women occupied a variety of social spaces, performed different roles, and were participants and active agents in history. A very small part of their history has, however, been written so far. In the older writings, a great part of the discussion about women of the Vedic age focused on elite women, ignoring the less privileged members of this sex. Although the Rig Veda mentions goddesses, none of them are as important as the major gods. The social implications of the worship of female deities are complex. While such worship does at least mark the ability of a community to visualize the divine in feminine form, it does not automatically mean that real women enjoyed power or privilege. The proportion of hymns attributed to women in the Rig Veda is miniscule (just 12–15 out of over 1,000), as is the number of women sages. This suggests that women had limited access to sacred learning. There are no women priests in the Rig Veda. While women participated as wives in sacrifices performed on behalf of their husbands, they did not perform sacrifices in their own right; nor do they appear as givers or receivers of dana or dakshina. The Vedic household was clearly patriarchal and patrilineal, and women enjoyed relatively little control over material resources. Their sexuality and reproductive resources were controlled through the ingraining of norms of what was considered appropriate behaviour. Early Vedic literature has several words for household units—durona, kshiti, dam/ dama, pastya, gaya, and griha—which may have corresponded to different kinds of households. Considering that this was a patriarchal and patrilineal society, it is not surprising that Rig Vedic prayers are for sons, not daughters, and that the absence of sons is deplored. The Rig Veda attaches importance to the institution of marriage and refers to various types of marriage—monogamy, polygyny, and polyandry. The rituals indicate post-puberty marriages, and there are references to women choosing their husbands. A woman could remarry if her husband died or disappeared. There are also references to unmarried women, such as the Rig Vedic seer Ghosha. Hymn 7.55.5–8 tells of elopement, the man praying that his beloved's entire household—her brothers and other relatives—as well as the dogs, should be lulled into a deep sleep, so that the lovers could creep out stealthily. KEY CONCEPTS

The family and the household

The word ‘family’ means different things to different people. If you ask a person about the members of her family, she might mention herself, her siblings, and her parents. Another person

might include grandparents and great-grandparents, dead or alive. Yet another person might include aunts and uncles, cousins, nephews, nieces, etc. As pointed out by A. M. Shah ([1964], 1998: 15), the word ‘family’ can refer to: the 1. household, i.e., all people living in one house or under one head, including parents, children, and household employees parents 2. and their children, whether living together or separately all 3. those who are held to be close relatives by birth or marriage all 4. those who are either descended or claim to be descended from a common ancestor a5.property-holding unit a6.ceremonial unit, for instance, including all those who have the right to perform the shraddha rites in honour of deceased ancestors.

Definitions of the family that are based on the issue of property holding or the performance of the shraddha do not help in understanding social groups that are property-less or who do not perform the shraddha rituals in the prescribed way. Because the word ‘family’ can mean so many different things, sociologists often qualify it with an adjective that makes it more specific. So, for instance, the terms ‘elementary family’ and ‘nuclear family’ refer to a married couple and their children, who may or may not live together. An extended family means two or more elementary families (or parts of them) joined together. This can take the form of a patrilineal joint family—sons and their families living with their father—in societies based on patrilineal descent, and a matrilineal joint family in societies based on the principle of matrilineal descent. It is not easy to draw the dividing line between the joint or extended family and the lineage. The household is more specific and easier to identify. Members of a household share a common residence. They perform different economic activities, some within, others outside the home. The household is the site of people’s most intimate and profound experiences in life. It is a place where many different kinds of human emotions and experiences are played out every day—those involving love and hatred, conflict and cooperation, oppression and compassion, violence and concern. Households come to be related to other households, families, and lineages through ties of kinship and marriage. The institution of marriage grants social approval to a union of two people assumed to be sexual partners and grants legitimacy to their offspring. Marriage and the household do not necessarily go hand in hand. For instance, among certain matrilineal groups in Kerala and the Lakshadweep islands, the husband does not live with his wife, but visits from time to time. Families can be divided into different types on the basis of descent, residence, membership, and the number of mates. Mention was made earlier of patrilineal and matrilineal social systems. Some societies recognize cognatic descent—i.e., descent in both the mother’s and the father’s line. For example, in American and European societies, although children often still take the

surname of the father, property rights and ideas of closeness and distance with the mother’s or father’s side do not vary. Patriliny and matriliny are not equivalent to patriarchy and matriarchy. Patriarchy means societies in which males (usually the eldest male) exercise dominant power and authority within the family. Matriarchy refers to a system in which such power and authority is vested in women. While there are several instances, including in our own times, of matrilineal societies, no known society of the past or the present can be described as matriarchal. Families in which the wife moves to live in her husband’s father’s house (or his grandfather’s or uncle’s house, if the father is not alive) are known as patrilocal or virilocal. Families in which the husband moves in with his wife’s mother’s family are known as matrilocal or uxorilocal (e.g., the Nayars of Kerala and Khasis of Meghalaya). Another type of arrangement is called duolocal—where the husband and wife continue to live with their respective families even after their marriage (e.g., in the Lakshadweep islands and central Kerala). Family types can also be distinguished from each other on the basis of the number of mates. Monogamy is a system in which a person has only one spouse at a time. In polygamy, one person can have more than one spouse at the same time. There are two types of polygamy— polygyny is a system in which a man can have several wives, while polyandry is a system in which a woman can have several husbands. There is a form of polyandry where the marriage ritual may be between a woman and one man, but the woman may either be considered the wife of all the brothers, or the latter may have access to her sexual and domestic services. Sociological studies reveal a great deal of diversity among families and households in different parts of the subcontinent today. Similar diversity must have prevailed in ancient times as well.

Male dominance and the subordination of women is a feature of all known historical societies. The issue is one of the degree of dominance and subordination, and the structures in which these were embedded. Compared to later Vedic literature, the family books of the Rig Veda Samhita reflect a situation in which social status was not as rigidly defined or polarized as it came to be in later times. However, it was not a society of equals—rank and gender were the two main bases of inequality. RELIGION: SACRIFICES TO THE GODS

The Rig Veda reflects the beliefs and practices of a religious aristocracy and its patrons, and there are several striking similarities with ideas reflected in the Iranian Avesta. The Rig Veda indicates a diversity of religious practice. For instance, there is mention of people who did not worship Indra, and the Dasas and Dasyus are described as not honouring the Vedic gods and not performing sacrifices. The Vedic hymns divide the universe into the sky (dyu), earth (prithvi), and the middle realm (antariksha). The word deva (literally, ‘shining’, ‘luminous’) is frequently used for the gods. The gods are sometimes also called asuras. Initially, this word referred to a powerful being; in later

times it came to be used exclusively in a negative sense for demons. The Rig Veda asserts that there are 33 gods associated with the sky, earth, and the intermediate region, but the actual number of deities mentioned in the text is more. Some gods are mentioned more often than others, but there is no fixed order of importance nor a fixed pantheon. Whichever deity is invoked in a particular hymn is spoken of as a supreme god. Max Müller described this phenomenon as Henotheism or Kathenotheism. Apart from the gods, the Rig Veda mentions gandharvas (celestial beings), apsaras (celestial nymphs, wives of the gandharvas), and malevolent beings such as rakshasas (demons), yatudhanas (sorcerers), and pishachas (spirits of the dead). Different ideas of how the world was created are mentioned in passing—e.g., as a result of a great cosmic battle, the separation of heaven and earth, or the actions of the gods. Deities were worshipped through prayer and sacrificial rituals (yajnas). The sacrifice marked a movement from the everyday, mundane sphere of activity and experience to the sacred sphere. The gods are presented as powerful, mostly benevolent beings, who could be made to intervene in the world of men via the performance of sacrifices. Sacrifices took place in the house of the yajamana (the person for whom the sacrifice was performed and who bore its expenses) or on a specially prepared plot of land nearby. They consisted mostly of oblations of milk, ghee, and grain poured into the fire, accompanied by the recitation of appropriate sacrificial formulae. Some yajnas involved the sacrifice of animals. The gods were supposed to partake of the offerings as they were consumed by the fire. A part of the offerings were eaten by the officiating priests. The goals of Rig Vedic sacrifices included wealth, good health, sons, and a long life for the yajamana. Some sacrifices were simple, domestic affairs, performed by the householder. Others required the participation of ritual specialists. Seven types of sacrificial priests are mentioned in the Rig Veda— the Hotri, Adhvaryu, Agnidh, Maitravaruna, Potri, Neshtri, and Brahmana—each with his particular tasks clearly laid down. Priests were given a fee (dakshina) in return for the important duties they performed. The Rig Veda does not mention temples or the worship of images of deities, which were an important aspect of popular Hinduism of later times. The Rig Veda reflects a naturalistic polytheism—a belief in many gods who personified natural phenomena. The connection is clear in some cases from the very name of the deity, as in the case of Agni (Fire), Surya (the Sun), and Ushas (Dawn). However, the mythology of some deities stretched far beyond their association with a particular natural phenomenon. For instance, although Indra seems to have been originally associated with the thunderstorm, he rapidly outgrew this connection to develop a much more complex personality. The gods were conceived of as anthropomorphic, i.e., as having a physical form similar to that of humans. The level of detail varies, but mention is often made of their head, face, mouth, hair, hands, feet, clothes, and weapons. There is an overlap in some of their physical features, epithets, and exploits. Indra is the most frequently invoked god in the Rig Veda. The hymns vividly describe his appearance and personality. He is vigorous and strong, a great warrior, his weapon is the thunderbolt, and he leads the Aryas to victory in battle. He is bounteous (maghavan) and loves to drink soma. There is reference to his mother and father (Tvashtri is often mentioned as his father). Indrani is his consort and the Maruts his companions. There are many references to Indra defeating hostile forces and demons such as Vala, Arbuda, and Vishvarupa. The most important myth connected with him is his victory over the serpent demon Vritra. In this episode, Indra is fortified by the god Soma and accompanied by the Maruts. He kills Vritra with his thunderbolt and frees the waters that

had been obstructed by the demon. The Rig Veda often mentions Indra as Vritrahan, slayer of Vritra. Many scholars interpret the conflict between Indra and Vritra as a creation myth, in which Vritra symbolizes chaos. PRIMARY SOURCES

Hymn to Indra (Rig Veda 2.12)

This hymn praises Indra, describing various aspects of his personality and referring to various myths connected with him. Note the reference in the fifth verse to people who doubt his existence: The god who had insight the moment he was born, the first who protected the gods with his power of thought, before whose hot breath the two world halves tremble at the greatness of his manly power—he, my people, is Indra. He who made fast the tottering earth, who made still the quaking mountains, who measured out and extended the expanse of the sky, who propped up the sky—he, my people, is Indra. He who killed the serpent and loosed the seven rivers, who drove out the cows who had been pent up by Vala, who gave birth to fire between two stones [this could refer to fire, the sun, or lightning], the winner of booty in combats—he, my people, is Indra. He by whom all these changes were rung, who drove the Dasas down into obscurity, who took away the flourishing wealth of the enemy as a winning gambler takes the stake—he, my people, is Indra. He about whom they ask, ‘Where is he?’ or they say of him, the terrible one, ‘He does not exist,’ he who diminishes the flourishing wealth of the enemy as gambling does—believe in him! He, my people, is Indra. He who encourages the weary and the sick, and the poor priest who is in need, who helps the man who harnesses the stones to press soma, he who has lips fine for drinking—he, my people, is Indra… He who is invoked by both of two armies, enemies locked in combat, on this side and that side, he who is even invoked separately by each of two men standing on the very same chariot—he, my people, is Indra. He without whom people do not conquer, he whom they call on for help when they are fighting, who became the image of everything, who shakes the unshakeable—he, my people, is Indra…. Even the sky and the earth bow low before him, and the mountains are terrified of his hot breath; he who is known as the soma-drinker, with the thunderbolt in his hand, with the thunderbolt in palm, he, my people, is Indra....

[To Indra] You who furiously grasp the prize for the one who presses and the one who cooks [the soma], you are truly real. Let us be dear to you, Indra, all our days, and let us speak as men of power in the sacrificial gathering. SOURCE O’Flaherty, 1986: 160–62

Agni is another important god and is often invoked along with Indra. He represents many aspects of fire—the fire of the cremation pyre, the fire that engulfs forests, the fire that burns enemies, the heat generated by tapas (austerity), and the heat of sexual desire. Most important of all, as the sacrificial fire, he is the intermediary between gods and humans. In this role, he functions as a divine priest. Soma—the personification of the soma plant—is closely associated with Indra and Agni, and is credited with many similar exploits. He is described as a wise god, one who inspires poets to compose hymns, a great god who rules over the earth and all humans. In later hymns, Soma is identified with the moon. Varuna and Mitra are frequently invoked together in the Rig Veda and are members of an eightmember group of gods known as the Adityas. Varuna is associated with kshatra (secular power), sovereignty, and kingship. He restricts and punishes evil-doers with the fetters or bonds that he has at his command. Although the hymns mention his eye and golden mantle, they do not give vivid descriptions of his physical appearance. He is associated with maya, an ability to construct forms. He is an all-seeing god who knows what everybody is up to. Other deities of the Rig Veda include the sun god Surya, son of Dyaus. Surya drives away the darkness by riding in his chariot across the sky, and is sometimes visualized as a white horse or an eagle. Vayu is the wind god. The Ashvins are twin gods associated with war and fertility. Vishnu is mentioned infrequently in the Rig Vedic hymns. He is a benevolent god, and is in places associated with Indra. The Rig Veda mentions his three gigantic strides which encompassed the entire universe. Very few Rig Vedic hymns are addressed to Rudra, a deity associated with great destructive potential. These refer to several attributes similar to those associated with Shiva of later-day Puranic mythology. Rudra is a god who inspires fear. He is not offered the same sacrifices as the other gods—the offering to him consists of a ball of food thrown on the ground, similar to that used to propitiate spirits. The Maruts are Rudra’s sons who drive across the sky in horse-drawn chariots, creating rain and storms. Ushas, goddess of the dawn, is mentioned 300 times in the Rig Veda, and 20 hymns are addressed to her. Representing the victory of light over darkness, she is generous and is invoked by those desiring wealth. Aditi, mother of the Adityas, is another important goddess. Her name means freedom, and she is invoked to bestow freedom from sickness, harm, and evil. Some hymns speak of her as a mother and connect her with the earth and the cow. Raka is a benevolent, bountiful goddess. Sinivali bestows children. Prithvi (Earth) is a minor goddess, most often invoked together with Dyaus. Vach (speech), Ida (literally, ‘the milk and butter offered in the sacrifice’), and Sarasvati (representing the river of this name) are some of the other goddesses mentioned in the Rig Veda. However, except for Ushas, goddesses have a relatively insignificant presence in the text. The hymns of the Rig Veda contain fleeting allusions to myths involving gods, humans, and semidivine beings. Many of these myths are elaborated on in later texts. For instance, Rig Veda 10.95 is a

dialogue hymn consisting of a conversation between king Pururavas and the water nymph, Urvashi. Pururavas implores Urvashi to come back to live with him: ‘My wife, turn your heart and mind to me.’ Urvashi refuses: ‘… What use to me are these words of yours? I have left you, like the first of the dawns. Go home again, Pururavas. I am hard to catch and hold, like the wind.’ The details of the Urvashi–Pururavas myth are given in later texts. Such dialogue hymns may have been part of ritual performances. FURTHER DISCUSSION

The soma plant and its juice

In the Rig Veda, soma is a plant, the juice extracted from a plant and the name of a god. Soma can be identified with the haoma of the Avesta. The Rig Veda describes soma as a divine drink that confers immortality and many hymns describe its exhilarating effect. It is the drink of the gods and Indra is particularly fond of it. For humans, soma seems to have had the ability to alter physiological functions, alter states of mind, and sharpen creativity. It is described as endowing men with strength in battle, keeping them awake and alert at night, and inspiring poets to compose their hymns. The descriptions suggest that the juice of this plant had hallucinogenic, intoxicating, or sympathomimetic (stimulating the sympathetic nervous system or producing similar results) properties. At some point of time, the soma plant seems to have become difficult to obtain and substitutes had to be used. The pressing, straining, and drinking of soma juice was an important part of Vedic rituals. The juice seems to have been extracted by laying the plant on a skin and pressing it with stones. It was filtered through sheep’s wool and then offered to the gods. The juice was sometimes mixed with water and milk. Over 100 different identifications have been suggested for the soma plant. It has been identified with plants such as Cannabis sativa L. (hemp, bhanga), Panax ginseng C.A.M. (ginseng), Peganum harmala L. (Syrian rue), Papaver somniferum L. (opium poppy), and Amanita muscarita (fly agaric, a mushroom with hallucinogenic properties). The plants of the Ephedra genus are strong candidates. Varieties of these leafless plants grow in many parts of Asia and Europe, but they are not common in India. They have been used in folk medicine for a long time, and are identified as the original haoma by members of the Parsi community even today. The Ephedras contain ephedrine or pseudo-ephedrine, both of which have sympathomimetic effects. Studies have shown that their effects on human physiology can include the following: a rise in blood pressure, increase in heart muscle contraction, decrease in pulse rate, stimulation of metabolism, increased perspiration, hyperglycaemia followed by hypoglycaemia, stimulation of insomnia, tremor, nausea, and dilation of eye pupils. However, it is possible that the soma juice consisted of the extract of not one, but more than one, type of plant. SOURCE Nyberg, 1997

The Rig Vedic concept of rita corresponds to the ancient Iranian concept of asha. It refers to the order of the universe, the order of the sacrifice, and the moral order that human beings should adhere to. Some hymns refer to Varuna and Mitra as the guardians or furtherers of rita. In the later Book 10, there is a dialogue hymn in which Yami appeals to her brother Yama (in later mythology, the first son of the sun, the first mortal man, and king of the dead) to commit incest with her in order to procreate. Yama rejects her overtures, stating that to do so would be contrary to rita and to the ordinances of Mitra and Varuna. As far as funerary practices are concerned, the Rig Veda refers to both cremation and burial. The ideas of a vital force (asu) or spirit (manas) that survive death occur in the text. There are references to a heavenly paradise as well as a terrible hell. These issues are discussed in greater detail in later Vedic texts.

www.pearsoned.co.in/upindersingh VEDIC HYMNS THE HISTORICAL MILIEU OF LATER VEDIC AGE TEXTS ASPECTS OF EVERYDAY LIFE

Compared to the Rig Veda Samhita, later Vedic literature reveals greater complexity in political organization, social life, and economic activities. Agriculture increases in importance. Cereals such as barley (yava), wheat (godhuma), and rice (vrihi) are mentioned, and there are several references to agricultural operations such as sowing, ploughing, reaping, and threshing. The Atharva Veda has charms to ward off pests and to avert drought, reflecting the anxieties that farmers must have had. Land was occupied by extended families, and the clan seems to have exercised general rights over land. The institution of private property in land had not yet emerged. The household was the basic unit of labour. Slaves were not used for productive purposes to any significant degree, and there are no words for hired labour. Hymns in praise of gifts (dana-stutis) in the later books of the Rig Veda refer to generous presents of cows, horses, chariots, gold, clothes, and female slaves made by kings to priests. This indicates the items valued in society, the concentration of wealth in the hands of rulers, and the relationship and exchanges between kings and priests. The earliest references to the gift of land occur in later Vedic texts, but the attitude towards this practice was still ambivalent. The Aitareya Brahmana suggests that the king should gift 1,000 pieces of gold, a field, and cattle to the Brahmana who anoints him. Yet the same text tells us that when king Vishvakarman Bhauvana wanted to make a gift of land as dakshina to his Brahmana priest Kashyapa, the earth goddess herself appeared before him and said that no mortal should give her away. A similar story occurs in the Shatapatha Brahmana in the context of the performance of the sarvamedha sacrifice. The earliest literary references to iron in the Indian subcontinent are found in later Vedic literature. The terms krishna-ayas, shyama, and shyama-ayas (the black or dark metal) in the Yajur

Veda and Atharva Veda clearly refer to this metal. There are indications of the use of iron in agriculture. The Taittiriya Samhita (5.2.5) of the Black Yajur Veda mentions ploughs driven by 6 or even 12 oxen. These must have been heavy and may have been made of iron. The Atharva Veda (10.6.2–3) mentions an amulet born of a ploughshare, smitten away with a knife by a skilful smith. The reference to the smith and the fact that iron is definitely known in the Atharva Veda suggest that the ploughshare in question was made of iron. In the context of implements used in the ashvamedha sacrifice, the Shatapatha Brahmana (13–2.2.16–19) connects iron with the peasantry. Elsewhere, the same text (13–3.4.5) connects this metal with the subjects or people (praja). Early Buddhist texts belonging to c. 600–200 BCE contain several references to iron. The Suttanipata refers to many objects (a goad, stake, ball, and hammer) made of ayas. Especially important is a simile that mentions a ploughshare that has got hot during the day, and which ‘splashes, hisses, and smokes in volumes’ when thrown into water. This seems to be a reference to the process of quenching iron objects. The term ayovikara kushi in Panini’s Ashtadhyayi has been translated as ‘iron ploughshare’. All these references suggest that between c. 1000 BCE and 500 BCE, the use of iron in agriculture had become prevalent in the Indo-Gangetic divide and the upper and middle Ganga valley. Later Vedic texts mention various kinds of artisans, such as carpenters, chariot makers, bow-andarrow makers, metal workers, leather workers, tanners, and potters. There is a long list of crafts and occupations in the list of victims in the purushamedha sacrifice, described in the Vajasaneyi Samhita (30) and the Taittiriya Brahmana (3.4). These include the following: doorkeeper, charioteer, attendant, drummer, mat maker, smith, ploughman, astrologer, herdsman, maker of bowstrings, carpenter, wood-gatherer, basket maker, jeweller, vintner, elephant keeper, and goldsmith. Vocations mentioned in other later Vedic texts include those of the physician, washerman, hunter, fowler, ferryman, servant, barber, cook, boatman, and messenger. Wagons drawn by oxen were probably the most frequent mode of transport. Chariots (rathas) were used for war and sport, and people rode on horses and elephants. Boats are mentioned, but it is not clear whether they were for riverine or sea travel. The extent of trade is not certain. Exchange was still via barter, as there is no clear reference to coinage. The general milieu as can be gathered from the texts is a rural one, although towards the end of the period, there are traces of the beginnings of urbanism—the Taittiriya Aranyaka uses the word nagara in the sense of a town. Although only philosophical and religious texts of the time have survived, these allude to other branches of learning. The Chandogya Upanishad (7.1.2) gives a list of subjects of study including the Veda, itihasa, purana, spiritual knowledge (brahma-vidya), grammar, mathematics (rashi), chronology (nidhi), dialectics (vakovakya), ethics (ekayana), astronomy, military science, the science of snakes, and knowledge of portents (daiva). Later Vedic texts only indicate how sacred knowledge was imparted. Great importance was attached to the relationship between teacher and pupil and to oral instruction. The Shatapatha Brahmana refers to the upanayana ceremony, which initiated the young boy into brahmacharya—the stage of celibate studenthood. Education—of whatever kind—seems to have been largely restricted to elite males. The leisure pastimes mentioned in later Vedic texts are similar to those referred to in the family books of the Rig Veda. Chariot racing and dicing were popular, as were music and dancing. Lute players, flute players, conch blowers, and drummers are mentioned. So are musical instruments such as the cymbals (aghati), drums, flutes, lutes, and a harp or lyre with 100 strings (vana). The term

shailusha, mentioned among the victims in the purushamedha in the Vajasaneyi Samhita, may mean an actor or dancer. The Yajur Veda mentions a vansha-nartin (pole-dancer or acrobat). As for the food people ate, apupa was a cake mixed with ghee, or made out of rice or barley. Odana was made by mixing grain variously with milk, water, curds, or ghee; beans, sesame or meat were sometimes added. Karambha was a porridge made of grain, barley or sesame. Rice was sometimes fried, or else cooked with milk and beans. Yavagu was a gruel made out of barley. Milk products such as curds, sour milk, and butter were consumed. Meat was eaten on special occasions, such as when honouring guests. There are references to an intoxicating beverage called sura. The soma plant had become difficult to obtain, so substitutes were allowed. People wore woven cotton clothes. Clothes made of woollen thread (urna-sutra) are also mentioned often, and were probably made of sheep’s wool or goat’s hair. There is mention of turbans and leather sandals. Ornaments such as nishka were worn around the neck, and jewels or conch shells were worn as amulets to ward off evil. The Brahmana texts frequently mention the prakasha —either an ornament of metal or a metal mirror. THE EMERGENCE OF MONARCHY

Warfare is a striking aspect of the milieu of both early and later Vedic literature. Book 1 of the Rig Veda Samhita refers to a battle of 20 kings, involving 60,099 warriors (the numbers need not be taken literally). But the nature of political units was changing. The 6th century BCE political map of north India showed the existence of different kinds of political systems—monarchical states (rajyas), oligarchic states (ganas or sanghas), and tribal principalities. The roots of these developments lie in the period c. 1000–600 BCE. While some communities retained their tribal character, others were making the transition towards statehood. Larger political units were formed through the coalescing of tribes. The Purus and Bharatas came together to form the mighty Kurus, the Turvashas and Krivis formed the Panchalas, and the Kurus and Panchalas seem to have been allies or confederates. Later Vedic texts reflect a transition from a tribal polity based on lineage to a territorial state. Some historians argue that this transition was not yet complete. On the other hand, since the end of the period of composition of later Vedic texts falls within the 6th century BCE, when territorial states did evidently exist according to the testimony of other sources, it makes little sense to insist that the state emerged in the post-Vedic and not in the later part of the later Vedic age. Witzel (1995) has argued that the Kurus represent the first state in India. He suggests that it was the Kurus under their king Parikshit (and their Brahmana priests) who initiated the collection and codification of the Vedic corpus into a canon. This included the re-arrangement of old and new poetic and ritual material, and was necessary to fulfil the needs of the newly developed shrauta, ritual presided over by various ritual specialists. As explained in Chapter 4, the transition to a state polity is always the culmination of a number of complex political, social, and economic processes. The emergence of a monarchical state would have involved multiple processes of conflict, accommodation, and alliances. Monarchy involves the concentration of political power in the hands of a king. The supremacy of the rajan was achieved by sidelining rival claimants to power, establishing coercive mechanisms, and control over productive resources. Apart from the monarchies, there were polities that maintained their tribal moorings and where political power was in the hands of assemblies, not kings.

The rajan of later Vedic texts is, like his Rig Vedic counterpart, a leader in battle. But he is also a protector of settlements and of people, especially Brahmanas. He is a custodian of the social order and sustainer of the rashtra (this term does not necessarily refer to a well-defined territory). Hereditary kingship was emerging. The Shatapatha and Aitareya Brahmanas refer to a kingdom of 10 generations (dasha-purusham rajyam). There are a few references (e.g., Atharva Veda 1.9; 3.4) to the election of the king, but these probably amounted to a ratification of hereditary succession. There is an interesting reference to the Srinjayas expelling their king Dushtaritu Paumsayana from the kingdom, in spite of his 10 generations of royal descent. This was no doubt an exception to the rule. Later Vedic rituals exalted the supremacy of the king, both over his kinsmen and over his people. Terms such as samrajya and samrat reflect the imperial aspirations and ambitions of certain kings. The emergence of monarchy was accompanied by speculations on the origins of the institution and attempts to provide a legitimizing ideology. Some of these speculations refer to the divine realm, others to the human sphere. The Aitareya Brahmana (1.1.14) states that on being defeated in battle by the demons, the gods realized that the reason for their defeat was that they had no king. So they elected a king, who led them to victory against the demons. Elsewhere in the same text (8.4.12), it is said that the gods, led by Prajapati, decided to install Indra as their king on the grounds that he was the most vigorous, strong, valiant, and perfect among them all, and the one who best carried out tasks that needed to be done. Later Vedic texts emphasize the close connection between the king and the gods. The Shatapatha Brahmana asserts that the king gains identity with Prajapati through the performance of the vajapeya and rajasuya sacrifices. As the visible representative of Prajapati, although one, he rules over many. Such statements should be understood as attempts to exalt the status of the king, not as a theory of the divinity of kings, nor as indicative of their worship. The emergence of the rajan as wielder of supreme political power involved his distancing himself from those closest to him—his kinsmen. This distancing was emphasized in ritualized contests such as the chariot race in the vajapeya sacrifice, and the cattle raid and game of dicing in the rajasuya sacrifice. In earlier times, such contests may have decided who was worthy of becoming king, but now they were ritual enactments in which the outcome—the victory of the rajan —was already decided and known. Another aspect of the rajan’s increasing power was his acquiring greater control over productive resources. Bali, which was initially a voluntary offering, probably consisting of agricultural produce and cattle, gradually became obligatory. The Shatapatha Brahmana (1.3.2.15) states that the Vaishya offers bali because he is under the vasha (control) of the Kshatriya, and has to give up what he has stored when he is told to do so. The rajan is referred to as vishamatta—eater of the vish (people), indicating that he lived off what the people produced. The rajan’s appropriation of bali from the people does not, however, quite amount to a clearly defined and organized system of taxation. References to the sabha and samiti continue in later Vedic texts. For instance, in the Shatapatha Brahmana (4.1.4.1–6), the king prays: ‘May the samiti and the sabha, the two daughters of Prajapati, concurrently aid me.’ But with the increase in royal power, the power of the assemblies must have correspondingly declined. Later Vedic texts indicate a close relationship between the king and his purohita (his Brahmana priest and counsellor). Purohita literally means ‘one who is put in front’ (by the king). The relationship between king and purohita is likened to that between earth and heaven. The king is

considered the feminine, subordinate party in this relationship (Coomaraswamy [1942], 1993). The importance of the purohita is graphically illustrated in the rajasuya ceremony, where he introduces the king to the assembled people and announces: ‘This man is your king. Soma is the king of us Brahmanas’ (Shatapatha Brahmana 5.3.12, 4.2.3). The system of administration seems to have been fairly rudimentary. Kumkum Roy (1994b) has underlined the close connection between the emergence of the monarchical system, the varna hierarchy, the organization of kinship relations, and the structure of households. The grand shrauta sacrifices performed by the king legitimized the king's control over the productive and reproductive resources of his realm, while the domestic sacrifices performed by the grihapati legitimized his control over the productive and reproductive resources of his household. Brahmanical texts implicitly recognize the connections between the political and domestic spheres in their description of the rajan as a custodian of the social order. FURTHER DISCUSSION

The ceremony of the jewel offering

The ratnahavimshi (ceremony of the jewel offering) was a part of the rajasuya sacrifice. It involved the rajan going on successive days to the homes of certain people—the ratnins (literally, ‘jewels’)— and offering oblations to certain gods. There is some variation in the names and order of the list of ratnins in different texts. They included the following: the 1. Brahmana or purohita (he usually heads the list) the 2. rajanya (nobles) mahishi 3. (chief queen) parvrikti 4. (the discarded queen; it is necessary to visit her to ward off evil) senani 5. (commander of the army) suta 6. (charioteer or bard) gramani 7. (village headman) kshattri 8. (royal chamberlain) sangrahitri 9. (charioteer, master of treasury, or collector of tribute?) 10. bhagadugha (literally, ‘milker of shares’, distributor of food or perhaps collector of the king’s share of the produce) 11. akshavapa (literally, ‘thrower of dice’, a functionary connected with dicing or perhaps with the maintenance of accounts) 12. govikartana (chief huntsman) 13. takshan (carpenter) 14. rathakara (chariot maker) 15. palagala (courier) 16. sthapati (probably a judge or a local chief)

The ratnahavimshi ceremony indicates the status of the ratnins and the king’s dependence on them. Some ratnins were related to the king through kinship, whereas others were functionaries with whom he had no kinship relations. This illustrates the transitional nature of the later Vedic

polity—it was in between a polity in which kinship was still an important factor and one marked by an elaborate military and administrative machinery. Curiously, the Brahmana texts state that some of the ratnins were inferior both to the Brahmanas and to the Kshatriyas. So, immediately after the ceremony, the rajan was supposed to perform two rites to atone for the sin of associating these unworthy persons with the sacrifice. SOURCE Sharma [1959], 1996: 143–58 THE VARNA HIERARCHY

Although kinship ties were still very important, later Vedic texts indicate the beginnings of a class structure in which social groups had different degrees of access to productive resources. Varna was partly an ideology that reflected the increasing social differentiation of the times. It was even more an ideology that justified this differentiation from the point of view of the elite groups. In dividing society into four hereditary strata, this ideology defined social boundaries, roles, status, and ritual purity. Members of the four varnas were supposed to have different innate characteristics, which made them naturally suited to certain occupations and social rank. The varna hierarchy was to remain an important part of the social discourse of the Brahmanical tradition for many centuries, and the duties and functions of the four varnas are elaborated on in the Dharmashastra literature of later times. The Purusha-sukta (Purusha hymn) in Book 10 of the Rig Veda Samhita refers to four social groups—Brahmana, Rajanya (instead of Kshatriya), Vaishya, and Shudra, though the word varna is not mentioned. It describes the four groups, and a whole lot of other things as well, as originating from different parts of the body of a primeval giant named Purusha, in the course of a sacrifice supposed to have been held long, long ago, in which Purusha was the sacrificial offering. The body symbolism in the Purusha hymn indicates that the four varnas were visualized as inter-related parts of an organic whole. At the same time, it clearly indicates a hierarchy of ranks, with the Brahmana at the top and the Shudra at the bottom. The fact that the varnas are described as being created at the same time as the earth, sky, sun, and moon indicates that they were supposed to be considered a part of the natural, eternal, and unchangeable order of the world. In fact, as pointed out by Brian K. Smith (1994), the varna scheme was extended beyond society to the classification of other aspects of the world, the gods, and nature. Initially, there seems to have been some ambiguity about the relative positions of the higher varnas. In the Panchavimsha Brahmana (13, 4, 17), where Indra is associated with the creation of the varnas, the Rajanya are placed first, followed by the Brahmana and Vaishya. The Shatapatha Brahmana (13.8.3.11) also places the Kshatriya first in the list. Elsewhere, in the same text (Shatapatha Brahmana 1.1.4.12) the order is as follows: Brahmana, Vaishya, Rajanya, and Shudra. However, the order of the four varnas in the Brahmanical tradition became fixed from the time of the Dharmasutras onwards. The relationship between the Brahmana and Kshatriya varnas was close but complex. Later Vedic texts emphasize the importance of the purohita for the king, and the close relationship between the Rajanya and at least a section of the Brahmana community. On the other hand, the conflict between

the gods Mitra and Varuna has been seen as symbolic of a conflict between the two varnas. Mitra represented the principle of brahma (sacred power) and Varuna the principle of kshatra (secular power). There are several statements about the relationship between brahma and kshatra, describing them variously as antagonistic, complementary, or dependent on each other. Upanishadic philosophy has also been viewed, at least in part, as a reflection of the Kshatriya challenge to Brahmanical supremacy in the field of ultimate knowledge. The first three varnas were known as dvija, literally ‘twice-born’, i.e., those entitled to the performance of the upanayana ceremony, which was considered a second birth. They were eligible to perform the agnyadheya or the first installation of the sacred sacrificial fire, which marked the beginning of ritual activities prescribed for the householder. On the other hand, the texts also emphasize differences between the three varnas. The Aitareya Brahmana (8.36.4) states that the rajasuya sacrifice endowed each of the four varnas with certain qualities—the Brahmana with tejas or lustre, the Kshatriya with virya or valour, the Vaishya with prajati or procreative powers, and the Shudra with pratishtha or stability. Later texts such as the Shrautasutras laid down the different details of the performance of sacrifices such as the soma sacrifice and the agnyadheya, depending on the varna of the sacrificer. PRIMARY SOURCES

The Purusha-sukta (Rig Veda 10.90)

Purusha has a thousand heads, a thousand eyes, a thousand feet. He pervaded the earth on all sides and extended beyond it as far as ten fingers. It is Purusha who is all this, whatever has been, and whatever is to be. He is the ruler of immortality, when he grows beyond everything through food. Such is his greatness, and Purusha is yet more than this. All creatures are a quarter of him; three quarters are what is immortal in heaven. With three quarters Purusha rose upwards, and one quarter of him still remains here. From this he spread out in all directions, into that which eats and that which does not eat. From him Viraj [the active female creative principle] was born, and from Viraj came the Purusha. When he was born, he ranged beyond the earth behind and before. When the gods spread the sacrifice with Purusha as the offering, spring was the ghee, summer the fuel, autumn the oblation. They anointed Purusha, the sacrifice born at the beginning, upon the sacred kusha grass. With him the gods, sadhyas [demigods] and sages sacrificed. From that sacrifice in which everything was offered, the melted fat was collected, and he [Purusha?] made it into those beasts who live in the air, in the forest, and in villages.

From that sacrifice in which everything was offered, the verses and chants were born, the metres were born from it, and from it the formulae were born. Horses were born from it, and those other animals that have two rows of teeth; cows were born from it, and from it goats and sheep were born. When they divided Purusha, into how many parts did they apportion him? What do they call his mouth, his two arms, and thighs and feet? His mouth became the Brahmana; his arms were made into the Rajanya; his thighs the Vaishya, and from his feet the Shudras were born. The moon was born from his mind; from his eye the sun was born. Indra and Agni came from his mouth, and from his vital breath the Wind was born. From his navel the middle realm of space arose; from his head the sky evolved. From his two feet came the earth, and the quarters of the sky from his ear. Thus they set the worlds in order.... SOURCE O’Flaherty, 1986: 29–32

The Brahmanas had an exalted status in the varna hierarchy, associated as they were with the performance of sacrifices and with knowledge, specifically the study and teaching of the Vedas. In the Aitareya Brahmana (33.4), when Varuna is told that a Brahmana boy was going to be sacrificed to him instead of the son of king Harishchandra, he remarks, ‘A Brahmana is indeed preferable to a Kshatriya’. The Shatapatha Brahmana (11.5.7.1) associates the Brahmana with four special attributes: purity of parentage, good conduct, glory, and teaching or protecting people. He is also associated with receiving four privileges from the people—honour, gifts, freedom from being harassed, and freedom from being beaten. The Kshatriyas or Rajanya were connected with strength, fame, ruling, and warfare. The Vaishyas were associated with material prosperity, animals, food, and production-related activities such as cattle rearing and agriculture. In the soma sacrifice, prayers were offered for the protection of the brahma, kshatra, and vish. The goals varied, depending on the varna to which the yajamana belonged. For the Brahmana, the goal was priestly lustre (brahmavarchas), for the Rajanya it was prowess (indriya), and for the Vaishya, it was animals and food (pashu and anna). The position of the Shudra at the bottom of the varna ladder was fixed from the very beginning. He was associated with serving the higher varnas and performing menial tasks. He could not perform Vedic sacrifices. A dikshita (one who had undergone initiation for a Vedic sacrifice) was not supposed to speak to a Shudra. According to Aitareya Brahmana 35.3, the Shudra is at the beck and call of others, can be made to rise at will, and can be beaten at will (yatha-kama-vadhya). There were groups in society who were considered even lower than the Shudras. Slaves (dasas and dasis) are mentioned among gift items in the dana-stutis. However, on occasion, children born of slave women could aspire to high status. For instance, in Book 1 of the Rig Veda, there is a reference to Kakshivan, son of the sage Dirghatamas by a woman slave of the queen of Anga.

Kavasha Ailusha, author of a Vedic hymn in Book 10, is also described as the son of a woman slave. These were probably exceptional instances. Although there are no clear indications of the practice of untouchability in later Vedic texts, groups such as the Chandalas were clearly looked on with contempt by the elites. The Chhandogya Upanishad and Taittiriya and Shatapatha Brahmanas mention the Chandala in a list of victims to be offered in the presumably symbolic purushamedha (human sacrifice), and describe him as dedicated to the deity Vayu (wind). The dedication to Vayu has been interpreted as indicating that the Chandala lived in the open air or near a cemetery, but this is far from certain. The Chhandogya Upanishad (5.10.7) states that those who perform praiseworthy deeds in this world swiftly acquire rebirth in a good condition—as a Brahmana, Kshatriya, or Vaishya, while those who perform low actions acquire birth in a correspondingly low condition—as a dog, boar or Chandala. The Shatapatha Brahmana (1.4.1.10) gives the story of a king named Videgha Mathava who originally lived on the banks of the Sarasvati and crossed the Sadanira (Gandak) river with his priest Gotama Raghugana, preceded by Agni Vaishvanara. Historians have often interpreted this story as reflecting the eastward movement of the Indo-Aryans and the first agricultural ‘colonization’ of the eastern lands through burning down the forests. On the other hand, giving an early Videhan king a respectable north-western origin may have been a way of legitimizing his power, and the reference to Agni may allude to the extension of Brahmanical sacrificial ritual to these areas. Later Vedic texts reflect processes of social interaction, conflict, and assimilation. According to the Aitareya Brahmana (33.6), when his 50 sons did not accept Shunahashepa (Devarata) as his son, Vishvamitra cursed them to become the Andhras, Pundras, Shabaras, Pulindas, and Mutibas. This story reflects the attempt of the Brahmanical tradition to extend some amount of recognition to ‘outsiders’. Some non-Indo-Aryan groups were assimilated into the varna hierarchy, usually at the lower rungs. In fact, the Shudras may have been a non-Indo-Aryan tribe living in the north-west, who later lent their name to the fourth varna (Sharma [1958], 1980: 34–35). However, not all tribal groups were assimilated. Some were simply acknowledged. Later Vedic texts mention forest people such as the Kiratas and Nishadas. They also show the emergence of the concept of mlechchha, a category that included various tribal groups and foreign people considered to be ‘outsiders’ by the Brahmanical tradition (see Parasher, 1991). While later Vedic texts suggest that society in the upper Ganga valley was becoming increasingly stratified, there was still a certain amount of fluidity in occupations. This is suggested in Rig Veda 9.112.3, where the poet says: ‘I am a reciter of hymns, my father is a physician, and my mother grinds (corn) with stones. We desire to obtain wealth in various actions.’ GENDER AND THE HOUSEHOLD

The household was an important institution, not only for its members, but also for the larger social and political units of which it was a part. A series of household rituals legitimized the householder’s control over the productive and reproductive resources of the household (Roy, 1994b). In later Vedic literature, the variety of household forms of earlier times made way for an idealized griha unit headed by the grihapati. Only a married man, accompanied by his legitimate wife, could become the yajamana in a sacrifice. Marriage (vivaha) was important for the continuation of the patrilineage. Relations between husband and wife (pati and patni) and father and son were hierarchically organized. Women came to be increasingly identified in terms of their relations with men. Words

such as stri, yosha, and jaya were closely associated with wifehood and motherhood, actual or potential. The grihapati had control over the productive resources of the household unit and the reproductive potential of his wife. This control was maintained by a domestic ideology that clearly laid down the structures of dominance and subordination within the family. The productive resources of the household were transferred from father to son, and rituals such as the agnyadheya emphasized the importance of ties with the patrilineal ancestors (pitris). The Grihyasutras, the earliest of which go back to this period, give lists of six or eight types of marriage (discussed in the next chapter). Later Vedic texts refer to marriage by capture, and to a woman choosing her spouse. Polygyny was more prevalent than polyandry. Kings could have any number of wives and concubines. The Aitareya Brahmana (3.5.3.47) states that even though a man may have several wives, one husband is enough for one woman. The Maitrayani Samhita refers to the 10 wives of Manu. A woman was married not only to a man but into a family. There are references in a later Rig Vedic hymn and in the Atharva Veda to the practice of a widow marrying her younger brother-in-law. The later Vedic ideas and ceremonies of marriage are reflected in a complex hymn in the tenth Mandala, often referred to as the Surya-sukta (Surya hymn) (Rig Veda 10.85). This hymn suggests that the bride was simultaneously considered a precious asset and a stranger with destructive potential. The marriage ceremonies seem to have been largely confined to the bride, groom, and their immediate families. In the marriage hymn in the Atharva Veda (14.1–2), the priest is assigned a more prominent role in neutralizing the dangerous potential of the bride and in ensuring her incorporation into her new home. Women are praised and exalted in some places in later Vedic texts. For instance, the Shatapatha Brahmana (5.2.1.10) states that the wife is half her husband and completes him. The Brihadaranyaka Upanishad (6.4.17) mentions a ritual for obtaining a learned daughter. On the other hand, women were generally excluded from the study of the Vedas. Although their presence as wives was required in the shrauta sacrifices, they could not perform such sacrifices independently in their own right. Later texts even introduce the possibility of an effigy of gold or grass in place of the wife. Most of the samskaras (except, of course, marriage) did not apply to them. In such crucial respects, the position of a woman—no matter what her varna—was indeed similar to that of a Shudra. In fact, the later Dharmashastra equation between women and Shudras goes back to the Vedic texts (see Shatapatha Brahmana 14.1.1.31). Later Vedic texts reflect the idea that the menstrual blood of women is dangerous and polluting (Smith, 1991). A menstruating wife is not supposed to participate in sacrifices. The sacrifice has to be postponed or it has to be performed without her. The Taittiriya Samhita reflects other taboos as well—it was inappropriate to talk to, sit near, or eat food cooked by a menstruating woman. According to this text, when Indra killed Vishvarupa, son of the god Tvashtri, he transferred onethird of the stain of killing a Brahmana to women. This ‘stain’ is said to have taken the form of women’s menstrual periods (Taittiriya Samhita 2.5.1). Women were clearly expected to conform to a docile role. Shatapatha Brahmana (10.5.2.9) states: ‘A good woman is one who pleases her husband, delivers male children, and never talks back to her husband.’ According to the same text (4.4.2.3), women own neither themselves nor an inheritance. The Atharva Veda (1.14.3) describes a life of spinsterhood as the greatest curse for women, and deplores the birth of daughters (6.11.3). The Aitareya Brahmana (7.15) describes a

daughter as a source of misery, and states that only a son can be the saviour of the family. The desire for sons is borne out in many hymns. A gestation rite called the pumsavana was prescribed to ensure the birth of a male child. The Atharva Veda contains charms for changing a female foetus into a male one. The Maitrayani Samhita (4.7.4) says: ‘Men go to the assembly, not women.’ Women appear as gifts and commodities of exchange, for instance in the references to rajas gifting their daughters to win over sages. The only form of ritual gift giving or exchange that women could be part of was giving the first alms to the brahmachari, who was supposed to begin his stint by begging from his mother or his teacher’s wife. The increasing social differentiation and emergence of a state was accompanied by an increasing subordination of women. References to women’s work in later Vedic texts include tending cattle, milking cows, and fetching water. There are also the vayitri and siri (female weaver), peshaskari (female embroiderer), bidalakari (female splitter of bamboo), rajayitri (female dyer), and upalaprakshini (woman corn grinder). The Shatapatha Brahmana mentions women carding wool. Apala is described in the Rig Veda (8. 80) as having taken care of her father’s fields. Vishpala (Rig Veda 1.112.10 and 1.116.5) was a woman warrior who lost a leg in battle, and there are references to other women warriors such as Mudgalini and Vadhrimati. A few women—Gargi and Maitreyi—participated in philosophical debate with Upanishadic sages. RELIGION, RITUAL, AND PHILOSOPHY

Later Vedic literature contains a variety of ideas on creation. The Purusha-sukta describes creation as the result of a primordial sacrifice, while other hymns describe creation as emanating from the sun or from Hiranyagarbha (the golden embryo). A hymn to the god Vishvakarman (10.81) imagines the creator god as an artisan—as a sculptor, smith, woodcutter, or carpenter—and as the first sacrificer and the sacrificial offering. The Nasadiya hymn, in Book 10 of the Rig Veda Samhita, has one of the most abstract and profound explorations of the mysteries of creation. In the family books of the Rig Veda, certain gods were brought together by invoking them in the same sacrificial rituals. In the later parts of the text, some hymns emphasized the connections among them. There are 40 hymns in the Rig Veda addressed to Vishvadevas—all the gods. Some hymns speak of the various gods as manifestations of the same divine being. Thus, Rig Veda 1.164 points out the differences in the names Agni, Indra, and Vayu, and goes on to assert that there is one being, whom the poets speak of as many (ekam sad vipra bahudha vadanti). The sacrificial ritual of the Brahmana texts

The Brahmana texts reflect a situation where sacrifices had become longer, more elaborate, and expensive. The sacrifice is presented as the act that created the world, and the correct performance of sacrifice was seen as necessary to regulate life and the world. While some sacrifices involved the participation of just one priest, others involved many more, and the ritual specialists were extremely important. The god Prajapati, who is most closely identified with sacrifice, is the most important deity in the Brahmanas. PRIMARY SOURCES

The Nasadiya hymn (Rig Veda 10.129)

There was neither non-existence nor existence then; there was neither the realm of space nor the sky which is beyond. What stirred? Where? In whose protection? Was there water, bottomlessly deep? There was neither death nor immortality then. There was no distinguishing sign of night nor of day. That one breathed, windless, by its own impulse. Other than that there was nothing beyond. Darkness was hidden by darkness in the beginning; with no distinguishing sign, all this was water. The life force that was covered by emptiness, that one arose through the power of heat. Desire came upon that one in the beginning; that was the first seed of mind. Poets seeking in their heart with wisdom found the bond of existence in non-existence. Their cord was extended across. Was there below? Was there above? There were seed-placers; there were powers. There was impulse beneath; there was giving-forth above. Who really knows? Who will here proclaim it? Whence was it produced? Whence is this creation? The gods came afterwards, with the creation of this universe. Who then knows whence it has arisen? Whence this creation has arisen— perhaps it formed itself, or perhaps it did not—the one who looks down on it, in the highest heaven, only he knows—or perhaps he does not know. SOURCE O’Flaherty, 1986: 25–26

FURTHER DISCUSSION

The sacrificial arena

The elaborate shrauta (Vedic) sacrifices involved the use of three fires—the garhapatya (householder’s fire), ahavaniya (offeratorial fire), and dakshinagni (southern fire). These fires were supposed to be placed in pits of different shapes. The pit for the garhapatya was supposed to be round, that of the ahavaniya square, and that of the dakshinagni crescent shaped. The position of the fires—and everything else—was fixed. The garhapatya was located in the west, the dakshie for the various ritual actions), the Udgatri (prinagni in the south, and the ahavaniya to the east. The garhapatya was lit first of all, and the other two fires were then lit from its coals. The vedi was a rectangular area with concave sides, situated between the garhapatya and ahavaniya fires. It was covered with sacred grass, and the equipment required for the sacrifice was placed here. The positions of the priests such as the Hotri (the priest of the Rig Veda, responsible for recitation), the Adhvaryu (the priest of the Yajur Veda, responsible for the various ritual actions),

the Udgatri (priest of the Sama Veda, responsible for the singing), and the Brahmana were specified. The yajamana and his wife also had their assigned places.

F IGURE 5.1 DIAGRAM OF SACRIFICIAL ARENA

The agnihotra was a simple domestic sacrifice, to be performed daily by the head of a dvija household, morning and evening. It involved the pouring of oblations of milk, and sometimes vegetal substances, into the fire, to the god Agni. There were also the periodic new-moon and full-moon sacrifices, and those performed at the beginning of the three seasons. There were even grander, longer, more elaborate ones which involved the participation of many different ritual specialists along with their assistants, which must have been performed by wealthy people and kings. The yajamana underwent a diksha (consecration) before the sacrifice, and had to follow a number of rules until its completion. The dakshina was an important part of the sacrifice, and as the sacrifices became longer and more complicated, it became larger and larger. A number of complex sacrificial rituals were associated with kingship. The vajapeya sacrifice was connected with the attainment of power and prosperity, and also contained a number of fertility rites. It included a ritual chariot race in which the rajan raced against his kinsmen and defeated them. The ashvamedha was a sacrifice associated with claims to political paramountcy and incorporated several fertility rites as well. The rajasuya was the royal consecration ceremony. Apart from a number of agrarian fertility rites, it included a ritual cattle raid, in which the rajan raided the cattle of his kinsmen, and also a game of dice, which the king won. At a larger, symbolic level, in the rajasuya, the king was presented as standing in the centre of the cyclical processes of regeneration of the universe (Heesterman, 1957). The Upanishads

The word ‘Upanishad’ (literally, ‘to sit near someone’) is usually understood as referring to pupils sitting near or around their teacher. Alternatively, it could mean connection or equivalence; the Upanishads were constantly suggesting connections and equivalences between things. The knowledge that was to be imparted and absorbed was no ordinary knowledge. It was allencompassing, the key to liberation from the cycle of birth, death, and rebirth, something that could only be taught to select, deserving pupils. It was difficult to explain and even more difficult to comprehend. It was revealed through discussion, debate, and contest among seekers, using a variety of devices—stories, images, analogies, and paradoxes. The oldest Upanishads are in prose, the later ones in metre. The Brihadaranyaka and Chhandogya are among the earliest. The Upanishads and Aranyakas deal with similar things, and the distinction between the two categories of texts is not always clear. For instance, the Brihadaranyaka Upanishad is considered both an Aranyaka and Upanishad. While the early Upanishads belong to the period c. 1000–500 BCE, many others are of a later period. These texts mark the first clear expression of certain key ideas and practices that are associated with Hindu and certain other Indian philosophical and religious traditions. These include the concepts of karma, rebirth, and the idea that there is a single, unseen, eternal reality that underlies everything. The Upanishads also deal with the practices of meditation and yoga. Considering the fact that the Upanishads were the work of many different people living in various parts of north India over many centuries, it is not surprising that they do not contain a single, cohesive, uniform system of ideas. They deal with many issues, but are especially concerned with the two fundamental concepts of atman and brahman (not to be confused with the god Brahma). A major concern of Upanishadic thought is to explore and explain their meaning and mutual relationship. The word brahman comes from the root brih, which means to be strong or firm. (The word occurs in the Rig Veda, atman does not.) It means something that grants prosperity, a vital force that strengthens and animates. In the Upanishads, there are many efforts to describe brahman. The fact that the texts have difficulty in explaining it is not surprising. The Kena Upanishad (2.1) asserts that the gods themselves were unable to understand brahman, and even those who think they have understood it do not. The Taittiriya Upanishad (3.1.1) states that brahman is that from which all beings are born, that by which they are sustained, and that into which they enter on death. Brahman is the eternal, imperishable reality in the universe. In the Brihadaranyaka Upanishad (3.8.11), the sage Yajnavalkya tells Gargi that the imperishable brahman sees but can’t be seen, thinks but can’t be thought of, perceives but can’t be perceived. The Mundaka Upanishad (1.1.7) explains that just as a spider spins and gathers its web, just as plants grow upon this earth, and just as head and body hair grow from a living person, even so does everything in this world arise from the imperishable brahman. Later Upanishads speak of brahman as of a god. If brahman is the ultimate reality pervading the universe, the atman is the ultimate reality within the self of an individual, i.e., the imperishable essential self. There are many explanations of the atman in the Upanishads. The Brihadaranyaka Upanishad (3.7.23) describes it as the knowing subject within us, which sees but is not seen, hears but is not heard, comprehends but is not comprehended, knows but is not known. In the Chhandogya Upanishad (3.14.2–3), the atman is described as lying deep within the heart, smaller than a grain of rice, barley, or mustard seed,

smaller even than a millet grain or millet kernel. Paradoxically, it is also described as larger than the earth, the intermediate region, and the sky, larger than even all the worlds put together. PRIMARY SOURCES

The atman, according to Uddalaka Aruni

The Chhandogya Upanishad tells the following story: One day, Uddalaka Aruni told his son Shvetaketu to go forth and take up the celibate life of a student, as their family was Brahmana only in name and none had so far devoted themselves to study. So Shvetaketu went off to become a student when he was 12 years old. He learnt all the Vedas and came back swollen headed when he was 24, thinking that he knew everything. His father Uddalaka Aruni saw this. He went on to instruct Shvetaketu on a number of issues about which the son knew nothing, and soon made him realize just how little he knew. In the following conversation between father and son in the Chhandogya Upanishad (6.13.3), Uddalaka uses graphic analogy to explain the nature of the atman to Shvetaketu. The first speaker is Uddalaka, and the father and son speak alternately: ‘Bring a banyan fruit.’ ‘Here it is, sir.’ ‘Cut it up.’ ‘I’ve cut it up, sir.’ ‘What do you see there?’ ‘These quite tiny seeds, sir.’ ‘Now, take one of them and cut it up.’ ‘I’ve cut one up, sir.’ ‘What do you see there?’ ‘Nothing, sir.’ Then he told him: ‘This is the finest essence here, son, that you can’t even see—look, on account of that finest essence, this huge banyan tree stands here. Believe, my son: the finest essence here—that constitutes the self of this whole world; that is the truth; that is the self (atman). And that’s how you are, Shvetaketu.’ SOURCE Olivelle, 1998: 255

The word maya occurs in the Shvetashvatara Upanishad. Scholars disagree on whether the idea or something similar is present in earlier Upanishads as well. Maya, often translated as ‘illusion’, can be interpreted in other, different ways. It can mean ignorance (avidya), the inability to realize oneness with brahman, or the creative power of ishvara (god) from the human point of view. The idea of a cycle of death and rebirth is present in the Brahmanas and Upanishads. The Shatapatha Brahmana states that those who do not perform the sacrificial rites correctly will be

born again and suffer death again. It also talks of a world where material pleasures are enjoyed by those who perform the sacrifices, and of a hell where evil-doers are punished. The same text refers to the dead as having to face two fires—good people pass through, while evil-doers perish in the flames. A person is born again after death and is punished or rewarded for his/her deeds. Some of the Upanishads explain the doctrine of transmigration. Death and rebirth are connected with ignorance and desire, and deliverance can be attained through knowledge. The Upanishads refer to three worlds—the worlds of humans, ancestors (pitris), and gods. Those who will be reborn go after death to the world of the fathers, while those who are destined for immortality go to the world of the gods. The goal of Upanishadic thought is the realization of brahman. Liberation (moksha, mukti) from the cycle of samsara could only be achieved through such knowledge. This knowledge (jnana) could not be obtained through mere intellectual exertion. This was knowledge of an inner, intuitive, experiential kind, that could only come upon the seeker as a sort of revelation that would transform him instantaneously. Later Upanishads such as the Shvetashvatara point towards yogic meditation as a means of realizing brahman. Performing of sacrifices and following an ethical code of conduct were of no use towards this end. In the Chhandogya Upanishad (3.8.11), Yajnavalkya tells Gargi that even if a man were to make offerings, perform sacrifices, and indulge in austerities for thousands of years, it wouldn’t amount to anything. The same text (2.23.1) states that people who performed sacrifices, recited the Veda, and gave gifts (dana), those who devoted themselves to the performance of austerities (tapa), and those who led a celibate life of studenthood in their teacher’s house studying the Veda—all these people gain worlds earned by merit. A person steadfast in the knowledge of brahman, on the other hand, attains immortality. In later times, there were many different interpretations of Upanishadic thought, which came to be known as Vedanta (literally, ‘end of the Veda’; also known as Uttara Mimamsa). Upanishadic thought reflects different ideas about atman, brahman, and the world, and statements such as tat tvam asi (you are that), aham Brahm-asmi (I am brahman), and brahma-atma-aikyam (unity of brahman and atman) can be interpreted in different ways. The Bhagavad Gita combined certain aspects of Upanishadic philosophy with a doctrine advocating righteous action. One of the most influential interpretations of the Upanishads was that of the 9th century thinker Shankara. According to Shankara’s monistic Advaita Vedanta (non-dualist Vedanta), the Upanishads tell us that there is only one single, unified reality—brahman—and everything else is not fully real. However, there is also a pantheistic strand in Upanishadic thought which identifies the universe with brahman. There is also a theistic strand of thought, which visualizes brahman as a god who controls the world. Given the diversity and complexity of Upanishadic ideas, it is not surprising that later thinkers interpreted them in very different ways. The Upanishads are often seen as anti-sacrifice and anti-Brahmana. The Brihadaranyaka Upanishad states that the performance of sacrifice leads to the world of the fathers (pitriyana), but knowledge leads to the world of the gods. Upanishadic knowledge is in several places associated with kings or Kshatriyas. There are references to Brahmanas being instructed in the knowledge of brahman by kings such as Ajatashatru, Ashvapati, and Pravahana. In the Chhandogya Upanishad (1.8–9), Pravahana tells Uddalaka Aruni that this knowledge has never till the present been in the possession of a Brahmana. In the Brihadaranyaka Upanishad (3–4), Yajnavalkya’s ideas are contradicted by Brahmanas, but are received with enthusiasm by king Janaka.

However, the fact that the Upanishads were included in the Vedic corpus as part of shruti should caution us against stretching this argument too far. For one thing, there are connections between the ideas of the Upanishads and early Vedic texts. Furthermore, the Upanishads do not reject sacrifice; rather, they employ the vocabulary of sacrifice to new ends. Ritual is re-described symbolically and allegorically. The link between humans and the cosmos is not the ritual itself but knowledge of the forces symbolically represented in the ritual. Knowledge of this symbolic meaning becomes more important than the performance of the ritual. An example of this is the re-description of the ashvamedha yajna in the Brihadaranyaka Upanishad. In this re-description, the various parts of the horse’s body are identified with different parts of the cosmos—his head is dawn, his eye is the sun, his breath is the wind, and his mouth is fire. The horse and the horse sacrifice take on new, symbolic meaning. Nevertheless, although ritual was not rejected, the emphasis had certainly shifted to the attainment of a new kind of knowledge. POPULAR BELIEFS AND PRACTICES

The Brahmanas were manuals for sacrificial priests, while the Upanishads reflect an esoteric quest for a special kind of self-knowledge. Although some of the ideas in these texts may have had a wider circulation, the Brahmanas, Upanishads, and Aranyakas cannot be described as texts reflecting popular beliefs and practices. The Atharva Veda, on the other hand, contains a number of charms and spells—for wealth, children, prosperity, health, etc.—reflecting the concerns of ordinary people. It also has hymns dealing with marriage and death. Although considered the latest Veda from the point of view of language and form, some of the ideas and practices reflected in this text are clearly very old. PRIMARY SOURCES

Atharva Veda spells

To win the love of a woman (Atharva Veda 6.9): As the creeper holds the tree in a tight embrace, so embrace me: be my lover and do not depart from me! As the eagle which seizes its prey beats its wings at the sun, so I beat at your heart: be my lover and do not depart from me! As the sun during the same day encircles the sky and the earth, so I encircle your heart: be my lover and do not depart from me! Against fever (Atharva Veda 5.22): May Agni drive the fever away from here— and so also may Soma, and the stone of the press, and Varuna of pure will, and the altar and the strewing and the flaming logs of wood! May enmities disappear!

You who make yellow all those whom you burn as in the fire, whom you consume— well, fever, may you be without strength: flee away there, flee away below! That wrinkled fever, daughter of wrinkles, red like a powder, throw it down, drive it away, O herb possessed of all powers! You are not comfortable in a strange land. Although you are powerful, have pity on us! Fever has found its proper occupation, it will go back among the Bahlika (people of the north-west). So cold, then burning, you make us shake with coughing, terrible are your characteristics, O fever; spare us from them! Do not take as allies the lingering sickness, nor the cough, nor shortness of breath; never come back again from where you have gone, O fever, I implore you! O fever, with your brother the lingering sickness, with your sister the fit of coughing, with your cousin the itch, go away and stay with other people! The fever which returns on the third day and that which dies down on the third day, the persistent fever and the autumn fever, the cold, the burning, the summer fever, and that of the rainy season, make them all disappear! To the people of Gandhara and of Mujavant, to those of Anga and of Magadha, we send the fever, like a messenger, like a treasure! SOURCE Renou, 1971: 23–24

Archaeological Profiles of Different Regions of the Subcontinent, c. 2000–500 BCE

We now move from literature to archaeology. The following sections of this chapter give a summary of the cultural sequences in different parts of the subcontinent as reflected in archaeological evidence. The discussion takes off from where Chapters 3 and 4 ended, and is organized into two parts—the first deals with neolithic–chalcolithic and chalcolithic cultures, and the second with early iron age cultures. The reason why more space has been given to certain regions and sites is not necessarily because they were more important, but because they have been more intensively studied. Full published reports are available for comparatively few sites, and there are some regions for which properly worked-out archaeological sequences and secure dates are unavailable. We can assume the continued existence throughout these centuries of hunter-gatherer communities, who must have interacted with agricultural–pastoral groups. NEOLITHIC–CHALCOLITHIC AND CHALCOLITHIC CULTURES THE NORTH-WEST AND NORTH

As mentioned in Chapter 4, in the north-west, the mature Harappan culture was followed by the late Harappan phase, represented by the Jhukar culture in Sindh and the Cemetery-H culture in Punjab. In

both cases, there are elements of continuity and change; the most clear change is the virtual disappearance of urban features. The Jhukar culture is known from excavations at Jhukar, Chanhudaro, and Amri. The distinctive pottery is a buff ware with a red or cream slip, with paintings in black, showing some continuity with mature Harappan pottery traditions. The cubical stone weights and female figurines of the Harappan type became rare. The typical rectangular Harappan seals were replaced by circular stamp seals, and writing was confined to potsherds. The Cemetery-H culture is represented, among other sites, at Harappa. Here, at the lower Cemetery-H levels, the graves consisted mostly of extended burials. The pottery showed some continuity with earlier levels, but there were also new forms and designs. In the upper levels, there were urn burials with disarticulated bones. M. R. Mughal’s study of the Bahawalpur area indicates changes in the number, frequency, and nature of settlements in the Cemetery-H phase. Although there were some fairly large settlements (e.g., Kudwala, 31.1 ha, and four sites—Lurewala, Lundewali II, Gamuwala Ther, and Shahiwala—between 15 and 20 ha), most of the sites were small, under 5 ha. Several of the mature Harappan settlements were abandoned, and late Harappan settlements were established in new locations. The number of sites dropped from 174 (mature Harap-pan) to 50 (late Harappan). There was a decline in the number of industrial sites, and an increase in multi-functional sites combining habitation with craft production. There was also a notable increase in short-duration camp sites. The decline in settlements and population in this area was the result of the drying up of the Hakra river. In the area between Peshawar and Chitral, on both sides of the Hindu Kush mountains, there are a number of cemeteries belonging to the Gandhara Grave culture. The C-14 dates for this culture range between c. 1710 and 200 BCE. The sites include Loebanr, Aligrama, Birkot Ghundai, Kherari, Timargarha, Lalbatai, Kalako-deray, Balambat, and Zarif Karuna. The graves generally consist of an oblong pit, sometimes with stone-lined walls, usually closed in with a stone slab. This pit was often dug into the base of a larger upper pit, which was filled with soil and charcoal, and often surrounded by a circle of stones. There were three types of burials—flexed burials, post-cremation burials including those in urns, and fractional burial. Both single and multiple burials occur. The site of Katelai yielded two burials of horses along with their masters. The grave goods included lots of plain, buff-red, or grey pottery in a range of shapes such as tall goblets, pedestal cups, beakers with fl ared mouths, and bottles with long and slender necks. Some graves yielded fl at, female fi gurines with appliqué breasts, occasionally with incised eyes and necklaces. There were copper/bronze objects such as pins with decorated tops, and a bronze model of a horse was found at Katelai. Iron objects were rare.

Mythological motifs on Cemetery-H pottery

The Cemetery-H urns bear naturalistic designs (leaves, trees, stars), but also an interesting series of what seem to be mythological motifs. The latter include peacocks with a human form drawn in the middle, and bulls/cows with plantlike attachments to their horns. In one scene, there are two long-horned animals facing each other, held by a man with long wavy hair; a dog seems to be skipping menacingly behind one of the animals.

These scenes have been interpreted in various ways. Some scholars have tried to connect them with the ideas of death and afterlife in the Vedas. However, all these interpretations remain speculative.

F IGURE 5.2 DESIGNS ON CEMETERY-H POTS

The Ghaligai cave sequence is an important reference point in this area. In this cave, Phases V, VI, and VII correspond to the early, middle, and late phase of the Gandhara Grave culture. Phase V was associated with a number of graves located on the hill-sides. There were cist graves made of vertical and horizontal stone slabs. Post-cremation burials outnumbered inhumations. Remains of rectangular stone houses were identified, and many different types of wheel-made pots and copper and bone artefacts were found. In Phase VI, there were more inhumations than post-cremation burials. Copper artefacts continued, and there was a fine wheel-made pottery in many different shapes, including chalices and cup-on-pedestal. Phase VII represented a late phase of the Gandhara Grave culture and yielded wheel-made red pottery and human terracotta figurines. Iron made its appearance. There is a similarity between some of the pottery types of Periods V–VII and those found in parts of central Asia.

F IGURE 5.3 G ANDHARA G RAVE CULTURE BURIAL, LOEBANR

In Kashmir, at sites such as Burzahom and Gufkral, the neolithic phase was followed by a megalithic phase. Megaliths are monuments made of large, roughly dressed slabs of stone. At Burzahom, there are massive menhirs (single, tall stones) and a large megalithic stone circle. Grey or black burnished ware made way for a coarse red ware. Bone and stone tools typical of the earlier period continued, but in fewer numbers. There were a few metal objects. At Gufkral, the megalithic phase (Period II) is marked by fallen menhirs, and was represented by a 50–60 cm thick habitational deposit. There was a nearly 10 cm thick floor, running throughout, with a few breaks marked by pits. Many of the latter were refuse pits, going down to natural soil levels, and contained lots of broken pottery and animal bones. The pottery of Period II showed continuity with neolithic Period I and included a burnished grey ware, gritty red ware, and thick dull-red ware, but the proportion of thick dull-red ware and wheel-made pottery had increased. There were lots of large finished and unfinished ring stones. Other arte-facts included a copper point, a wooden bead, pestles, spindle whorls, a fine awl, and a miniature pot. The number of bone tools decreased, but there were innovations such as handles for tools, mostly made from the tibia of sheep/goat and bone marrow sockets. All the grains of the preceding neolithic period continued. Rice and millet made their appearance towards the end of Period II. Faunal remains included the bones of cattle, sheep, goat, dog, pig, and fowl. The bones of sheep and goats outnumbered those of cattle. Hunting seems to have declined, because the only wild animal bones found were those of ibex. Iron has been reported at megalithic Gufkral.

MAP 5.1 MAJOR NEOLITHIC–CHALCOLITHIC SITES IN THE INDIAN SUBCONTINENT

Kiari in Ladakh yielded handmade pottery, similar to that found in Period II at Burzahom. Structural remains consisted of hearths. There were stone artefacts such as saddle querns, pestles, and burnishers. Bones of domesticated cattle, sheep, and goats were found. There is a calibrated date of 1000+ BCE from Kiari (Chakrabarti, 1999: 207). In the Almora area of the Uttarakhand Himalayas, there are various kinds of megalithic burials— dolmens, cairns, menhirs, and cists (see the later section on the megaliths for explanations of these terms). The cist burials of this area are associated with red, grey, and black pots, including pedestalled bowls and spouted pots, and also with horse burials. THE INDO-GANGETIC DIVIDE, THE UPPER GANGA VALLEY, AND THE DOAB

The late Harappan phase

As many as 563 late Harappan sites have been identified in the area between the Yamuna and Sutlej. Most of the settlements are small, under 5 ha. Evidence of mud floors with post holes and hearths, mud-brick structures, storage pits, kilns, and a fire altar were found from Sanghol in Ludhiana district of Punjab. The late Harappan settlement at Dadheri consisted of mud houses built on a mud platform. Artefacts found here included copper and terracotta objects and beads of carnelian and lapis lazuli. At Banawali in Haryana, there is evidence of mud houses and a rich range of artefacts including faience ornaments, beads of semi-precious stones, and objects made of copper, clay, and terracotta. Sanghol yielded evidence of a wide range of plant remains in a late Harappan context that was dated c. 1900–1400 BCE. An analysis of the palaeo-botanical remains (Saraswat, 1996–97) identified hulled barley, naked barley, dwarf wheat, bread wheat, jowar millet, Italian millet, khesari, field pea, lentil, chickpea (gram), horse gram, Egyptian clover (barseem), linseed, and sesame (til). The remains of hyacinth bean (sem), fruits (grape, lemon, karaunda, anwala), and opium poppy seeds were also found. The plant remains identified at Mohrana comprised hulled and naked six-row barley, dwarf wheat, club wheat, lentil, and grape pips. The chalcolithic cultural sequence in the doab includes the late Harappan phase, the Ochre Coloured Pottery (OCP) culture, the copper hoards, and the Black and Red Ware (BRW) phase. Some of these phases spilt out into adjoining areas as well. There are almost 70 late Harappan sites in the doab region, mostly along the higher banks of the tributaries of the Yamuna—the Hindon, Krishni, Kathanala, and Maskara. Most of the settlements are small (the largest measures 200 × 200 m), and the average distance between settlements is 8–12 km. The thickness of the deposits is 1–2 m. Three sites have been excavated—Alamgirpur in Meerut district and Hulas and Bargaon in Saharanpur district. The late Harappan occupation at Hulas may go back to before 2000 BCE and it seems to have continued till about 1000 BCE. There is very little structural evidence from late Harappan sites in the doab. Houses were generally made of wattle and daub. At Hulas, however, rectangular mud-brick structures with rammed floors, post-holes, and hearths were identified in the earliest phase. In the middle phase, clusters of two or three circular wattle-and-daub structures—perhaps storage bins—were found inside some of the rectangular mud houses. In the final stage, five round furnaces were found in some of the structures. A few burnt bricks were found at this site and at Alamgirpur.

P OTTERY FROM LATE HARAPPAN LEVELS, B HORGARH, DELHI

The late Harappan pottery of this area is made of well-levigated clay and includes both handmade and wheel-made types, with both coarse and fine fabrics. It has a thin cream wash or a bright red slip, over which geometric and naturalistic designs were painted in black. A small proportion of the pots also have incised designs. The other artefacts found at these sites included chert blades, stone querns and pestles, and bone points. There were a few copper objects—a broken blade from Alamgirpur, and a fragmentary chisel and some rings from Bargaon. Ornaments included bangles of terracotta, carnelian, and steatite, and beads of terracotta, steatite, agate, carnelian, and faience. Circular and triangular terracotta cakes and terracotta animals, carts, and wheels were also found. It can be inferred that people living at these sites continued to grow crops such as wheat and barley, which were known in the area in the mature Harappan phase. Rice husk was found embedded in the cores of potsherds at Hulas and Un. The list of plant remains from late Harappan Hulas is impressive—rice, barley, dwarf wheat, bread wheat, club wheat, oats, jowar (sorghum), ragi (finger millet), lentil, field pea, grass pea, kulthi, moong (green gram), chickpea (gram), a broken cowpea, cotton, castor, almond, walnut, fruits, and wild grasses. This was clearly an agricultural community with a diverse and well-established agricultural base. RECENT DISCOVERIES

The Sanauli cemetery

Excavations at Sanauli (Baghpat district, UP) by D. V. Sharma and his team have revealed what seems to be a vast late Harappan cemetery, although the excavators prefer to label it mature Harappan. Due to the standing crop of sugarcane around the site, it was not possible to estimate its size. The Yamuna river flows about 6 km west of the site today, but it may have been closer in protohistoric times. The discoveries at Sanauli, tentatively dated c. 2200–1800 BCE, are similar in some respects to those found at other mature or late Harappan sites, but they also have certain unique features. So far, 116 graves have been excavated from different depths. All of them were laid in a northwest–southeast orientation; 52 were extended burials, 35 were secondary burials, and 29, which did not contain any human remains, seem to be symbolic burials. A double burial (Burial 27) at middle levels contained the skeletons of two males, aged 30–35 years. The grave goods included four flask-shaped vessels and a small rimless bowl near the head. A dish-on-stand with a splayed outer rim was placed in the middle of the grave, below the hip portion of the skeletons. A beautifully decorated long steatite bead and another bead of white-banded agate were also found. Only one skull was found among the skeletal material. There was also a triple burial (Burial 69) along with two urn burials. This was a secondary burial. Only one skull was found, placed upside down. The absence of skulls may have been due to the peculiar circumstances of death of the individuals concerned. This grave contained 21 pieces of pottery of different types, including three dish-on-stand and two pitchers with lids in the shape of bull heads. One of the symbolic burials (Burial 28) in the upper levels contained two mushroom-shaped dish-on-stands. It also had a violin-shaped copper container with 28 tiny, paper-thin, stylized copper objects arranged in six rows. A burnt brick wall with a finished inner surface ran parallel to the burial in the east. Another symbolic burial (Burial 106) contained patterns of steatite inlays, the outer lines of which resembled a human effigy. A completely burnt clay trough found at middle levels of the site may have been used for cremating the dead. The grave goods in this cemetery comprised copper objects, gold ornaments including heartshaped bracelets, beads of semi-precious stones, steatite, faience, and glass. One burial

contained a copper antennae sword, along with a sheath. There is evidence suggestive of animal sacrifice in some middle and upper level burials. The dish-on-stand was clearly an important part of the grave goods. Its form evolved over time and the mushroom shape found in the upper levels has not been found elsewhere. In most burials, it occurred either below the hip or near/ below the head of the buried person; in a few instances, it was close to the feet. It was also used as an offering stand, in one case, holding the head of a goat.

A preliminary study of about 40 skeletons by S. R. Walimbe identified the skeletons of 10 males and 7 females. The sex of 17 skeletons could not be determined. The bones of five child burials were analysed; one of them was 1–2 years old, two were 3–5 years, and two were about 10 years. There were also remains of six sub-adults.

A spectacular cemetery of this kind must have been associated with a large habitation site. This has not yet been located. SOURCE D. V. Sharma et al., 2005–06

The Ochre Coloured Pottery culture

The Ochre Coloured Pottery (OCP) was discovered in 1950–51 in western Uttar Pradesh at the sites of Bisauli (Badaun district) and Rajpur Parsu (Bijnaur district). It is an ill-fired, wheel-made ware with a fine to medium fabric, and a thick red slip, sometimes decorated with black bands. Some potsherds have incised designs and post-firing graffiti. The pottery was given its name because when it was rubbed, it left an ochre colour on the fingers. This could be because of water-logging, wind action, poor firing, or a combination of such factors. Subsequently, OCP was found to be widely distributed in the doab, with a concentration in the Saharanpur, Muzaffarnagar, Meerut, and Bulandshahr districts of western Uttar Pradesh. Over 80 sites have been identified in Saharanpur district alone. This pottery has been found outside this area as well, and its distribution extends north–south from Bahadarabad near Hardwar in Uttar Pradesh to Noh and Jodhpura in Rajasthan, and east–west from Katpalon near Jullundar in Punjab to Ahichchhatra near Bareilly. The OCP phase in Rajasthan seems to be earlier than that in the doab. OCP occurs in two sorts of stratigraphic contexts. At Hastinapura, Ahichchhatra, and Jhinjhana, the OCP level was followed by a break in occupation and a Painted Grey Ware (PGW) level. At Atranjikhera and Noh, the OCP level was followed by a Black and Red Ware (BRW) level, and then a PGW level. Certain sites such as Bargaon and Ambakheri show an overlap between the late Harappan and OCP phase. Some scholars maintain that OCP is just a degenerate form of late Harappan pottery. According to others, it was an independent ceramic tradition that was influenced in some areas by the Harappan pottery tradition. At least two broad categories of OCP can be identified—a western zone (represented at sites such as Jodhpura, Siswal, Mitathal, Bara, Ambakheri, and Bargaon) that shows links with the Harappan tradition, and an eastern zone (represented at sites such as Lal Qila, Atranjikhera, and Saipai) that does not display any such links.

MAP 5.2 OCHRE COLOURED P OTTERY SITES

F IGURE 5.4 OCHRE COLOURED P OTTERY FROM AMBAKHERI

Major excavated sites include Lal Qila (Bulandshahr district), Bahadrabad and Ambakheri (both in Saharanpur district), Atranjikhera (Etah district), Ahichchhatra (Bareilly district), and Saipai (Etawah district). The OCP deposits are generally shallow, ranging from 0.5 to 1.5 m in thickness. The settlements are usually small (upto 200–300 sq m), although there are a few larger settlements such as Lal Qila (632 sq m). The average distance between two sites is 4–6 km in Saharanpur district, and 5–8 km in other parts of the upper Ganga valley. Due to the disturbed nature of the deposits and the small area covered by excavations, very few structural remains were found at most OCP sites. There were some remains of wattle-and-daub houses at Lal Qila, Atranjikhera, Daulatpur, and Jodhpura. Very few mud- or burnt bricks were found. At Atranjikhera, people lived in mud houses with frames made of posts of babul, sissoo, sal, and chir pine trees. An unlined well was also found. At Jodhpura, a mud-brick structure with the bricks joined together with mud mortar was discovered. Lal Qila must have been an important settlement, going by its size, structures, and range of artefacts. There were remains of oblong wattleand-daub structures with mud floors and post-holes, and a few sun-dried bricks with mud mortar. An unlined pit may have functioned as a well. Apart from pottery, very few artefacts have been found at OCP sites. Stone objects included querns and beads. Bone tools were found at Lal Qila. A few copper artefacts also occur at various sites. A piece of copper and fragments of a terracotta crucible containing copper granules were discovered at Atranjikhera. A hooked spearhead and harpoon made of copper were found at Saipai. Lal Qila yielded five copper objects—two pendants, one bead, an arrowhead, and a broken celt. The terracotta objects found at this site included anthropomorphic and animal figurines, wheels, bangles, balls, tablets, gamesmen, crucibles, discs, beads, grinders, and querns. Terracotta figurines of

humped bulls were found at Ambakheri. The Lal Qila pottery included a vase painted with a seminaturalistic humped bull with long, curved horns. The people who lived at OCP sites obtained their food from agriculture, animal husbandry, and hunting. Plant remains at Lal Qila included wheat, barley, and rice. Atranjikhera yielded rice, barley, gram, and khesari. This suggests that people grew two crops a year—rice in summer and barley and legumes in winter. At Saipai, sandstone pounders, querns, and pestles were found, and there were bones of domesticated Bos indicus. Lal Qila yielded complete animal skeletons on floors, and there were circular fire pits with charred bones of domesticated cattle, buffalo, goat, sheep, pig, horse, dog, and wild deer. Many of the bones had cut marks, indicating that the animals were killed for their meat. Thermoluminescence dates from Atranjikhera, Lal Qila, Nasirpur, and Jhinjhina range between 2650 and 1180 BCE. The OCP culture can be seen as a late contemporary of the mature Harappan and late Harappan cultures, with certain sites showing contact between them.

The copper hoards

In 1822, a copper harpoon was discovered at Bithur in Kanpur district. Since then, over 1300 copper objects of a similar range have been found in various parts of India, mostly in hoards. Archaeologists refer to them as copper hoards. Copper hoards have been found at about 90 sites across an area stretching from the upper Ganga valley to Bengal and Orissa. There have also been several discoveries in Haryana, Rajasthan, and Madhya Pradesh, and a few in Gujarat, Karnataka, Kerala, and Tamil Nadu. However, the largest concentration of sites is in the doab region of Uttar Pradesh. The number of objects found together varies from 1 to 47, except in the case of Gungeria in Madhya Pradesh, where 424 objects weighing over 200 kg were found in a single hoard, along with 102 silver objects. Since most of the copper hoard discoveries were accidental and the objects were not found in a stratified context, it is very difficult to date them. The hoards found in Bihar and West Bengal may in fact belong to the historical period. In view of this, the site of Saipai (in Etawah district), where the copper objects were found in the course of an excavation in an OCP level, is especially important.

COPPER HARPOONS FROM SHISHUPALGARH AND HASTINAPURA

F IGURE 5.5 COPPER HOARD OBJECTS

MAP 5.3 COPPER HOARD SITES

The copper hoards include many different kinds of objects such as flat celts, shouldered celts, bar celts, harpoons, antennae swords, and anthropomorphic figures. Most of them seem to be part of hunting equipment. Typological differences can, to some extent, be associated with geographical

areas. For instance, in the eastern zone of Bihar, West Bengal, and Orissa, there is a predominance of flat celts, shouldered celts, bar celts, and double axes. In the Uttar Pradesh and Haryana areas, these types occur along with anthropomorphs, antennae swords, hooked swords, and harpoons. Sites in Rajasthan have yielded mainly flat celts and bar celts. A comparison of the Harappan copper artefacts and the copper hoard objects shows striking differences in typology and alloying techniques. About 46 per cent of the copper hoard objects had up to 7 per cent arsenic alloying. On the other hand, only 8 per cent of analysed Harappan artefacts show arsenic alloying. The site of Sanauli has recently yielded two antennae swords of the copper hoard types in a late Harappan context. One of these was found in situ in a grave, and has a copper sheath. The evidence of the copper hoards suggests that between the mid-3rd and 2nd millennium BCE the upper Ganga valley had emerged as a distinct copper-manufacturing area, with interactions extending into the regions of Haryana, Gujarat, Madhya Pradesh, the Deccan, Kerala, and Tamil Nadu. What is not clear is whether it was an independent centre of copper working or whether it represents an extension of the older and better documented centre of copper metallurgy in north-eastern Rajasthan.

SEE P . 215 FOR DETAILS AN PHOTOGRAPHS OF SANAULI

FURTHER DISCUSSION

The copper anthropomorph

The most enigmatic artefact among the copper hoard objects is the anthropomorph. This is a large object, between 25 and 45 cm in length, 30 and 43 cm in breadth, and weighing up to 5 kg. The length is in almost all cases greater than its breadth (the Bisauli piece is an exception to this rule). The object usually has in-curved arms, sharpened on the outer edge, and plain outstretched legs. The arms are thinner than the head, which was thickened by beating the top.

In 2001, a hoard of 31 copper anthropomorphs was found at Madarpur in the Moradabad district of Uttar Pradesh (D. V. Sharma et al., 2001–02). They were discovered by workers digging the soil for the preparation of mud-bricks. The artefacts were in situ, stacked one on top of the other. Such a large number of anthropomorphic figures have not been found elsewhere. What is also intriguing is that the shapes are not identical, and there are some that do not occur anywhere else. The artefacts were found in a deposit which also yielded OCP. Madarpur seems to have been a place which specialized in the production of copper anthropomorphs. What was the anthropomorph used for? One suggestion is that it was a weapon. D. P. Agrawal suggests that when thrown, it has a sort of whirling boomerang effect, and that it may have been a missile used to kill birds. However, why such elaborate artefacts should have been made for this purpose is not clear. The different shapes of the objects also go against this theory. Another possibility is that they had a religious or ritualistic function. It may be noted that tiny anthropomorphic figures, similar to the copper hoard types, are worshipped in parts of northern India as the god Shani.

The Black and Red Ware phase in the doab

Black and Red Ware (BRW) was long known to occur at several archaeological sites in association with various other pottery types, in many different cultural contexts. However, the existence of a distinct and independent Black and Red Ware phase in the doab was first recognized during excavations at Atranjikhera in the 1960s. Here, a BRW level was found sandwiched between OCP and PGW levels. A similar stratigraphic sequence was later identified at Noh and Jodhpura in Rajasthan. Some archaeologists maintain that there are links between the BRW of the doab and Rajasthan, while others disagree. BRW levels at Atranjikhera did not yield any stone or metal artefacts. There were only fragments

of stone, waste flakes, chips, and cores of quartz, chalcedony, agate, and carnelian. Three beads (of carnelian, shell, and copper) and a fragment of a comb made of bone were also found. BRW levels at Noh yielded a shapeless piece of iron, terracotta bead, and bone spike. As for evidence of subsistence patterns, rice, barley, gram, and khesari were found from OCP levels at Atranjikhera, and it is likely that the cultivation of these crops continued into the BRW phase. Grains of rice and moong were found at BRW levels at this site.

BRW POT, MEGALITHIC LEVEL, MASKI (KARNATAKA)

PRIMARY SOURCES

Black and Red Ware

As its name indicates, Black and Red Ware (BRW) refers to a pottery that is both red and black. The two colours may appear on the same surface of the pot, or one surface may be black, the other red. BRW should not be confused with black-on-red ware (e.g., the typical Harappan pottery), in which both the inner and outer surfaces of the pot are red, and designs are painted in black. Many of the BRW pots are black inside and red outside. This could be a result of the inverted firing technique: In this, the pots are positioned upside down in the kiln with some vegetal material placed inside them. When the pot is fired, its outer part is exposed to oxidizing conditions and turns red, while its inner part is subjected to reducing conditions and turns black. Another possibility is that the pots went through two rounds of firing (double firing)—i.e., they were first fired red and then re-fired, so that one of the surfaces became black, or vice versa. Black-and-red pottery occurs in many parts of the subcontinent in several different cultural contexts. For example, it occurs at neolithic sites (Chirand, Piklihal, etc.), pre-Harappan Lothal, many Harappan sites in Gujarat (e.g., Lothal, Surkotada, Rojdi, Rangpur, and Desalpur), chalcolithic sites in the middle and lower Ganga valley (Chi-rand, Pandu Rajar Dhibi, etc.), sites of the Ahar/Banas culture (Ahar, Gilund), Malwa culture (Navdatoli, Inamgaon), Kayatha culture (Kayatha), and Jorwe culture (Chandoli), iron age PGW sites (Atranjikhera, Hastinapur, etc.), South Indian megalithic sites (Brahmagiri, Nagarjunakonda, etc.), and at early historical levels

all over the subcontinent. At certain sites in the doab (e.g., Atran-jikhera) and Rajasthan (e.g., Noh), there is a distinct BRW level between the OCP and PGW levels. Not all of this black-and-red pottery are identical. There is in fact a great deal of variation in technique, fabric, and shape among the black-and-red pottery that occurs in different geographical and chronological contexts. In view of all this, it is clear that all black-and-red pottery cannot be treated as representing a single ceramic culture, a single community of artisans, or a single community of consumers. The existence of different varieties of black-and-red pottery at various sites does not necessarily show the existence of cultural uniformity or cultural contact. When we talk about black-and-red pottery in In-dian archaeology, its specific geographical and cultural context must always be indicated. The case of Black and Red Ware shows that we must always be very careful in making historical inferences on the basis of superficial similarities in pottery. SOURCE H. N. Singh, 1979 WESTERN INDIA

In Chapter 3, there was a discussion of the early phase of the Ganeshwar–Jodhpura culture of northeastern Rajasthan, with special reference to the sites of Jodhpura and Ganeshwar. The early phase of the Ahar/Banas culture of south-east Rajasthan, represented at sites such as Ahar, Gilund, and Balathal was also discussed. Rajasthan continued to be a major region for copper metallurgy during the succeeding centuries as well. At Ganeshwar, Period III is dated from c. 2000 BCE onwards. There was a wide range of pottery in this phase. Hundreds of copper artefacts, e.g., arrowheads, rings, bangles, spearheads, chisels, balls, and celts were found. This shows that Ganeshwar was a major centre for the manufacture of copper artefacts. Compared to Period II, there was a decline in the number of microliths and animal bones, suggesting a decline in hunting. Period I at Ahar is divided into three periods—Ia (dated from 2500 BCE), Ib (dated from 2100 BCE), and Ic (dated from 1900 BCE). Period Ia was discussed in Chapter 3. Here we will look at Periods Ib and Ic. As far as pottery is concerned, there is continuity in BRW throughout Periods Ia, Ib, and Ic, but there are some changes in the types and proportions of the associated wares. For instance, in Period Ia, there were mostly convex-sided BRW bowls; buff and imitation buff-slipped wares, red wares, and some grey ware. In Period Ib, the BRW continues, and there is a lot of grey ware and red ware, but no buff and buff-slipped ware. Period Ic was marked by deeply carinated BRW bowls and lustrous red ware. The artefacts discovered in Period Ib at Ahar included microlithic fluted cores and a blunt-backed blade of quartz; beads of agate, calcite, carnelian, faience, jasper, schist, shell, steatite, bone, and terracotta; terracotta objects such as ear studs, skin rubbers, head scratchers (?), votive tanks, crucibles, dice, bangles, finials, pipes, pendants, and human and animal (bull, horse, and maybe elephant) figurines. Copper objects included rings, bangles, kohl sticks, celts, and a knife blade. In Period Ic, there were microlithic scrapers and borers; beads of carnelian, crystal, glass, jasper, lapis, schist, shell, and terracotta; terracotta skin rubbers, ear studs, a votive tank, crucible, bull and

elephant figurines, stoppers, pendants, bangles, balls, and pipes. The copper items comprised rings and kohl sticks. Evidence from Ahar indicates that the people who lived here cultivated rice, and possibly millet. The evidence of structures and pottery suggest that in Period Ib, the site was more thickly populated than in the preceding and succeeding periods. It is also likely that there was interaction between the chalcolithic agricultural people of Ahar and the ‘mesolithic’ hunter-gatherers who lived at sites such as Bagor. Investigations at the late Ahar culture settlement at Purani Marmi (Chittorgarh district, Rajasthan) near Balathal yielded important information on the subsistence base of the people who lived here. A total of 545 animal bones from 4 habitational layers were analysed and identified as those of cattle, sheep, goat, buffalo, blackbuck, spotted deer, and domestic fowl. Two types of freshwater molluscs were also found. The relative quantities of the bones indicate that the people were mainly pastoralists engaged in cattle and buffalo rearing. They supplemented this with some amount of sheep and goat herding and a limited amount of hunting. Mention should also be made of a number of megalithic sites in the Aravalli stretches of Rajasthan, for instance, Khera, Satmas, and Daosa. Very few details or dates are available for them. The most common type of megalith found in this region is the cairn. In Gujarat, the mature Harappan phase was followed by a late Harappan phase. As mentioned in Chapter 4, Kutch and Saurashtra show a marked increase in the number of settlements, from 18 in the mature Harappan phase to 120 in the early late Harappan phase. The late Harappan settlements in Gujarat can be divided into two phases—the pre-lustrous red ware sites (Lothal B, Rojdi, Babar Kot, Padri) and the lustrous red ware sites. Lothal (in Period B, also known as Phase V) revealed remains of houses made of mud and reed. Short blades of jasper and chalcedony replaced the long chert blades of the mature Harappan phase. Jasper and carnelian beads made way for biconal terracotta beads, and cubical weights of chert and agate were gradually replaced by larger, truncated ones made of schist and sandstone. There was a decrease in the use of copper. Rectangular steatite seals with the Harappan script continued, but without animal motifs. Rojdi Ia, Rangpur IIB and IIC also represent the late Harappan phase. The settlement at Rojdi was about 7 ha in size. The main settlement area was surrounded by a stone rubble wall on three sides (the Bhadar river lay on the east), with a double-bastioned gateway in the western wall. There were other structures of stone masonry as well. Various types of metal artefacts were found, e.g., an axe, bar celt, bangles, rings, a fishhook, pieces of wire, and a pin. The plant remains included millets, barley, mustard, khesari, lentil, linseed, pea, vetches/beans, various kinds of gram, jujube, and a number of weeds, medicinal plants, and grasses (which may have been used for animal fodder). The late Harappan site of Babar Kot measured about 2.7 ha and had a stone fortification wall. The plant remains included millets and gram. Prabhas Patan II (Somnath Patan in Junagadh district) on the banks of the river Hiran is divided into two sub-phases—the earlier one has late Harappan pottery but no lustrous red ware, and the later one has late Harappan pottery associated with lustrous red ware. A structural complex made of stone blocks set in mud mortar and divided into smaller compartments was interpreted as a warehouse. Artefacts included a steatite seal amulet, segmented beads made of faience, and cubical chert blades. There were copper objects and beads of chalcedony, carnelian, and agate, and a gold ear ornament.

At Dwarka (Jamnagar district, Gujarat), marine archaeologists found the remains of a submerged settlement and identified its inner and outer walls, bastions, and a large stone jetty. Stone anchors and lustrous red ware were found at the site. The island of Bet Dwarka also revealed a submerged site. The settlement seems originally to have been 4 × 0.5 km, and there are remains of fortifications. A Harappan seal carved with a three-headed animal, lustrous red ware, BRW, and a jar inscribed with Harappan writing were found. Other discoveries included a coppersmith’s stone mould and some shell bangles. There is a thermoluminescence date of 1570 BCE from Bet Dwarka, which is considered to be a late Harappan site.

DWARKA: MARINE ARCHAEOLOGISTS DIVING NEAR SAMUDRA NARAYAN TEMPLE

There are many late Harappan sites in the Rupen valley in north Gujarat, with and without lustrous red ware. The settlements tend to be located on old sand dunes, close to sources of water. Most of them are small and have a thin occupational deposit. The late Harappan sites in this area seem to mostly represent seasonal camp sites of pastoralists. There is more information from the site of Kanewal in Kheda district at the mouth of the Gulf of Cambay. Here, there were circular wattle-and-daub huts with rammed floors. The artefacts included oblong terracotta cakes, beads of carnelian, faience, shell, and terracotta; terracotta spindle whorls and net sinkers; copper objects; and various types of pottery including lustrous red ware. Some of the pottery found at sites in this area has graffiti in the Harappan script, indicating some level of literacy among the people who lived here.

DIVER MEASURING SUBMERGED STRUCTURE

CIRCULAR STONE STRUCTURE IN THE INTER-TIDAL ZONE THE MIDDLE GANGA VALLEY

There are a large number of protohistoric sites in this region, especially in the trans-Sarayu area. Narhan in Gorakhpur district (UP) is on the northern bank of the Sarayu (Ghaghara), about 30 km east of Imlidih (Purushottam Singh, 1994). Excavations at this site revealed a cultural sequence stretching from the second half of the 2nd millennium BCE to the 7th century, CE. Period I at Narhan (labelled the Narhan culture) was dated c. 1300–700 BCE. Remains of wattle-and-daub houses with post-holes and hearths were found. The pottery was marked by a white-painted BRW, along with some white-painted black-slipped ware, red-slipped ware, and plain red ware. Other artefacts included bone points; pottery discs; terracotta beads, dabbers, and balls; and one polished stone axe. The copper objects included a ring and fishhook. Recent chemical analysis of these objects, as well as of those found in later periods indicates that they were made of a low-tin bronze. The metal workers were familiar with techniques such as alloying, cold working, annealing, and casting. The source of the copper ores seems to have been the Rakha mines of Bihar. An exceptionally wide range of plant remains were found at Narhan. Period I remains included cultivated rice (Oryza sativa), hulled and six-row barley (Hordeum vulgare), three kinds of wheat (Triticum compactum, T. aestivum, and T. compactum), pea, green gram, gram or chickpea, and khesari. Oilseeds—mustard and flax (alsi) were found, as were seeds of jackfruit (katahal). Fragments of mahua, sal, tamarind, teak, siris, babul, mulberry, ganiyari, Nux vomica, tulsi (holy basil), mango, katahal, and bamboo were identified. Animal bone remains included those of humped cattle and sheep/goat, wild deer or antelope, horse, and fish. An interesting find was the impression of a fishhook and thread on a mud clod. Iron rust showed that the hook was made of iron, and the analysis of a tiny surviving fibre revealed that the thread was made of ramie (Boehmeria nivea), a

strong, water-resistant fibre. Two iron pieces (a 13 cm long bar and another fragmentary piece), were found in the upper deposits of Period I. Iron objects increased in the subsequent period. The Narhan sequence is repeated at many places in the middle Ganga plains, including in the area south of the Sarayu. At Khairadih, Period I was marked by BRW and associated pottery. Calibrated dates gave a range of 1395–848 BCE. At Rajghat near the Ganga, the early occupation was marked mainly by a blackslipped ware. The many BRW sites in the area to the south of Mirzapur include Raja Karna Ka Tila on the Karamnasa river. Period I at this site yielded BRW, microlith chips, a clay sling ball, shells, terracotta beads and discs, and bone points and arrowheads. Rice, barley, ragi, foxtail millet, lentil, field-pea, khesari, and moong were identified. Period II began in about 1300 BCE and gave evidence of iron. Imlidih Khurd is a site on the banks of the Kuwana river. Period I represents the pre-Narhan culture and goes back to c. 1300 BCE. It yielded a crude, handmade cord-or-mat-impressed red ware, including spherical bowls, pedestalled bowls, vases with flaring rims, and handi-like and spouted vessels. There were remains of wattle-and-daub houses, a storage pit (1.95 m in diameter), a circular bin-like structure (about 85 cm in diameter), and ovens. Artefacts included beads of agate, faience, and terracotta, a few steatite micro-beads, bone points, and pottery discs. The faunal remains included bones of domesticated cattle, sheep/goat, and pig. Cattle bones were the most numerous, and had cut marks. Bones of freshwater turtle, fish, and freshwater molluscs were also found. The plant remains were extremely varied and included rice, barley, bread wheat, dwarf wheat, jowar, bajra (pearl millet), lentil, moong, field pea, grass pea, mustard, and sesame. The seeds of fruits—wild jujube, anwala, and grape—were also found. The evidence indicates that agriculture based on two crops a year was already established in the trans-Sarayu plain in the first half of the 2nd millennium BCE. Period II at Imlidih Khurd belongs to the Narhan culture and is dated c. 1300–800 BCE. It was marked by intense structural activity in the form of at least two successive mud floors with several post-holes and ovens. The typical pottery was a white-painted BRW, similar to that found at Narhan. Other artefacts included bone points, pottery discs, terracotta beads, a copper arrowhead, two copper beads, and some curious terracotta pieces that may have been legs or pedestals of some indeterminate object, possibly for ritualistic use. The plant remains comprised rice, barley, wheat, kondon-millet, lentil, chickpea, moong, and anwala, along with various weeds and wild plant species. The faunal remains included the bones of domesticated cattle, goat/sheep, horse, and dog. The bones of wild animals comprised those of boar, hog deer, chital (spotted deer), and barasingha (swamp deer). Except for the molluscs, the aquatic fauna of Period I continued into Period II. Recent excavations at the 40 ha site of Agiabir in Mirzapur district revealed a long cultural sequence extending from the Narhan phase to the early medieval period. In Period I (the Narhan culture phase), the main pottery types were BRW, black-slipped ware, and red ware. The pottery showed some differences with the typical range of Narhan ware. People lived in wattle-and-daub huts, and two silos used for storing grain were found. There were lots of beads, especially those made of agate. A bead-making workshop was identified. Faience objects, microliths, terracotta beads, bone points, terracotta discs, one copper fishhook, and a clay lamp or incense burner were found. Fireplaces associated with charred animal bones gave evidence of peoples’ food habits.

Period II at Agiabir has been described as pre-NBP with iron. Iron and copper objects were the noteworthy finds of this phase. There are a number of sites marked by megaliths in and near the northern fringes of the Vindhyas in Allahabad, Banda, Varanasi, and Mirzapur districts of south-eastern Uttar Pradesh. These include Kakoria, Jang Mahal, and Kotia. The main types of megaliths that occur are cairns and stone circles. Some of the graves gave evidence of fractional burial. Others were associated with animal burials. At Kotia, the graves yielded few human skeletal remains, but three contained the remains of domesticated sheep, pig, and cattle. Cut marks suggest that the animals were killed at the time of burial. Many of the megaliths in this area are devoid of skeletal remains of any kind, and may represent memorials for the dead. The habitation site of Kakoria lies on both sides of the Chandraprabha river, immediately to the north-west of the megalithic cemetery at the base of a hillock. The pottery from the habitation and burial sites included BRW, black-slipped ware, and red ware. Most of it was wheel made, and the main forms included dishes, bowls, perforated vessels, lids, pedestalled cups, and elliptical and globular jars. Lots of microliths made of agate, chalcedony, and chert were found. Beads of terracotta and semi-precious stones, sling balls, grinding stones, and a few copper objects also occurred. Unlike other parts of India, most of the megaliths of southern Uttar Pradesh belong to a pre-iron age. Kotia in the Belan valley is an exception. Here there were many iron tools, including a spearhead, two sickles, an arrowhead, and an adze, all indicating advanced metallurgical techniques. The Kotia pottery included BRW, red ware, black-slipped ware, and a dull, coarse black or grey ware, all with a thick fabric. There were many bone fragments of domesticated animals such as ox, sheep, and pig, some with cut marks. A date ranging from the 2nd millennium BCE (or earlier) to the 7th century CE has been suggested for pre-iron Kakoria. The megaliths of Jang Mahal have been estimated as belonging to the beginning of the 1st millennium BCE. Kotia is placed later, between c. 800 BCE and 300 BCE. EASTERN INDIA

The early phase of occupation at sites such as Chirand and Senuar in eastern India was discussed in Chapter 3. These sites continued to be occupied into the 2nd millennium BCE. At Chirand, chalcolithic Period II is in many respects a continuation of neolithic Period I. There were microliths, polished celts, beads of terracotta, steatite, and semiprecious stones. The pottery was dominated by BRW along with grey/buff, black- and red-slipped wares. Copper made its appearance in Period II, and the upper levels yielded evidence of many iron objects. The earliest calibrated dates for Period II give a range of 1936–1683 BCE. At Senuar, Period II is neolithic–chalcolithic. The 2.02 m thick deposit showed a basic continuity with the preceding period. The new elements were some copper objects—a fishhook, piece of wire, needle, and an indeterminate object. A fragmentary rod of lead was also found. The plant remains showed the introduction of bread wheat, kondon millet (Pasupalum scrobiculatum), chickpea, green pea, and horse gram (Dolichos bilorus). There was an increase in the number of faunal remains compared to the earlier period. Barudih in the Singhbhum district, in the Chhotanagpur plateau of Jharkhand, yielded interesting evidence of microliths, neolithic celts, iron slag, and wheel-made pottery in the same ‘neolithic’

level. The iron objects included a sickle. The earliest radiocarbon dates give a range of 1401–837 BCE for this site. There seem to be close connections between the cultural patterns in Bihar and West Bengal. Over 65 BRW sites have been found in West Bengal. On the basis of size, the settlements can be divided into three categories—0.5–2 acres, 4–5 acres, and 8–9 acres. The BRW phase began in this region in about the middle of the 2nd millennium BCE. The problem is that ‘Black and Red Ware settlements’ are found upto about 400 BCE, stretching across a period of over a thousand years. Clearly, they belong to different periods, and there is a need to identify their chronological moorings. There is overall similarity in the range of artefacts found at the Bengal BRW sites—in the pottery, stone tools, beads of semi-precious stones, and fairly limited copper objects. Rice must have been the most important crop. The abundance of deer bones and antlers suggest the presence of large tracts of forests and grassy land. The agriculturists of the plains must have been interacting with communities, including hunter-gatherers, living in the Chotanagpur plateau, an area rich in stone and metal (especially copper and tin). Many BRW sites show some familiarity with iron, but the iron industry in this area really emerged in a major way only towards the end of the BRW phase. Pandu Rajar Dhibi in the Ajay valley is an important site in West Bengal. Period I, with calibrated dates from c. 1500 BCE onwards, revealed microliths, ground stone tools, bone tools, and pottery. No metal was found, but this may be due to the limited area covered by the excavations. In chalcolithic Period II, there were a few copper artefacts, beads of semi-precious stones, terracotta figurines, iron spearheads and points, slag, and ovens. The pottery included a BRW with designs painted on in white, along with other associated wares such as a red-slipped black-painted pottery, black-slipped pottery, and a buff/grey plain ware. The faunal remains included the bones of domesticated cattle, buffalo, goat, and deer, along with those of hog deer, sambar, fish, turtles, and fowl. At Bharatpur in the Damodar valley, Period I yielded microliths, small neolithic celts, bone tools, steatite beads, copper objects, and pottery dominated by BRW. The earliest calibrated date range for this period is 1735–1417 BCE. Period I at Mahisdal in the Kopai valley gave evidence of house floors rammed with terracotta nodules, lots of mi-croliths and bone tools, beads of steatite and semiprecious stones, terracotta bangles, a terracotta phallus, and one flat copper celt. The pottery consisted of BRW and associated wares. A storage pit with lots of charred rice grains was found. The earliest calibrated range of dates for Period I at Mahisdal is 1619–1415 BCE. In Orissa, neolithic stone tools occur in many places as surface finds, but there is a lack of stratified finds and dates. The neolithic sites in Orissa include Kuchai in Mayurbhanj district. Domesticated rice was found at the neolithic site of Baidipur. Sankarganj in Dhenkanal district gave a calibrated date of c. 800 BCE for a level yielding neolithic celts and copper artefacts. Recently, a neolithic celt manufacturing site was discovered at Sulabhdihi in the Sundargarh district of Orissa (Behera, 1991–92). At the recently excavated site of Golbai Sasan on the Mandakini river in Orissa, neolithic Period I showed traces of floors and post-holes. There was red and grey handmade pottery with cord or tortoiseshell impressions, and a few pieces of worked bone. Period IIA was chalcolithic. The outlines of circular huts (3.9–7.9 m in diameter), with hearths and post holes along the circumference, were identified. Both handmade and wheel-made pottery was found, including BRW, dull red ware and burnished black, chocolate brown, and red wares. Copper artefacts included a

chisel, bangle, fishhook, and ring. The polished stone tools included axes, adzes, and shouldered celts. Bone artefacts included weapons and ornaments (such as earstuds and pendants). Spindle whorls, sling balls, and a crude human figurine were among the other artefacts. The features of Period IIA continued into Period IIB, with the addition of an iron tool shaped like a stone celt. The plant remains of Periods IIA and IIB included rice, moong and kulthi. Faunal remains comprised bones of cattle, goat, deer, and elephant. The occupation of Golbai Sasan seems to fall within the 2nd millennium BCE, if not earlier. THE NORTH-EAST

The North-eastern states, comprising Assam, Arunachal Pradesh, Meghalaya, Tripura, Manipur, Nagaland, and Mizoram are rich in archaeological finds and potential, but have not been explored adequately. There are very few dates. Lots of neolithic tools have been found in the Garo, Cachar, and Naga hills, but mostly as surface finds. The evidence from the few excavated sites, meagre as it is, is therefore very important. At Sarutaru, 25 km south-east of Guwahati, excavations yielded shouldered celts and round-butted axes. The pottery included handmade brown, buff, and grey wares, some with cord impressions. However, the ‘neolithic’ phase at Sarutaru may be as recent as the early centuries CE. Excavations at nearby Marakdola revealed a 1 m thick deposit which yielded wheel-made pottery of fine kaolin clay. Similar pottery was found at Ambari near Guwahati, in contexts dated between the 7th and 12th centuries CE. In the north Garo hills, at Daojali Hading, a 45 cm thick neolithic deposit yielded stone and fossil wood axes, adzes, chisels, hoes, grinding slabs, querns, and mullers. Handmade grey and dull-red pottery with cord marks, dull-red stamped pottery, and plain red pottery were found. Excavations at Selbalgiri, on a terrace of the Rongram river, yielded a microlithic level, followed by a 60 cm deposit containing stone celts and pottery. Neolithic tools and handmade grey ware have been found at several places in Na-galand, but the sites have not been excavated. The site of Napchik in Manipur has given an early thermoluminescence date of 1650 ± 350 BCE for handmade cordmarked ware. Other artefacts found at this site included stone choppers, scrapers, flakes, an edged knife, grinding stone, and polished celts. Neolithic tools have been found at various places in Meghalaya. A small-scale excavation was conducted at Selbalgiri. An excavation at Pynthorlangthen revealed a 1 m thick neolithic deposit containing neolithic adzes, axes, chisels, points, blades, and scrapers. This seems to have been a factory site. Not all the sites in the North-east that have yielded stone tools and handmade pottery are necessarily early, and some are positively late. For example, the ‘neolithic’ level at the Kanai Gaon Reserve in Dibrugarh district has given a date in the 6th century CE. More excavations and a better idea of the chronology of the sites are required for a clearer picture of the neolithic and neolithic– chalcolithic horizons in this part of the subcontinent. Similarities between some of the artefact types of this region and those from East and South-east Asia have been noted, but nothing definite can be said about the connections or interactions. THE CULTURAL SEQUENCE IN CENTRAL INDIA

There is some evidence of late Harappan pottery in the north-western part of the Malwa plateau at sites such as Sihoniya, Khudai, and Bassaiya (on the banks of the Asan river, a tributary of the Chambal). Late Harappan pottery has also been reported at Manoti in Mandasore district. However, at present, little detail is available. There is, however, quite a bit of data on the well-established cultural sequence of the Kayatha, Ahar, and Malwa cultures (Dhavalikar, 1979a). The first of these was discussed in Chapter 3. Here, we will discuss the Ahar and Malwa phases.

The Ahar culture

As mentioned in Chapter 3, the Ahar culture that flourished in south-eastern Rajasthan also spread to the Malwa region of central India. Ahar culture levels have been identified at Kayatha and at several sites in the Chambal valley. The typical Ahar pottery is a coarse, wheel-made Black and Red Ware, with designs painted on in white (usually on the outer surface, but sometimes also on the inner one). There are bowls and dishes of various kinds; the bowls usually have thin incised grooves on the neck. Another associated pottery type is the red-slipped ware, which includes variants such as tan-, orange-, chocolate-, and brown-slipped pottery, all highly burnished. Coarse handmade red and grey wares are also found. The other artefacts include necklaces made of short, cylindrical beads. Unlike the meagre remains of stone artefacts at Ahar culture sites in Rajasthan, there was a prominent blade tool industry at Ahar levels at Kayatha. One of the unique features of the Ahar culture in central India are the terracottas. The animal terracottas mostly consist of naturalistic or stylized bull figurines, made of very fine clay with few impurities, baked at a uniformly high temperature. Many of the figurines have prominent humps and long, pointed horns. There is no decoration on their surface, only nail marks. An interesting find was a pair of short horns on a pedestal. It is possible that such figurines may have had a cultic significance. People lived in small mud houses with walls made of reed screens, thickly plastered with mud. Sometimes house floors were made of gravel and cobble, rammed in hard, compact clay. At Kayatha, there is evidence of a large-scale fire towards the end of the Ahar phase.

The Malwa culture

The Ahar culture phase was followed by the Malwa culture. Navdatoli (west Nimar district), on the southern banks of the Narmada, is the largest settlement of this culture. Calibrated dates for the beginning of the settlement are in the range of 2000–1750 BCE. Other important sites are Maheshwar (Nimar district), Nagda (Ujjain district), and Eran (Sagar district). The cultural sequence at the recently excavated site of Chichali (Khargaon district, MP) consists of Ahar, Malwa, Jorwe, and early historical levels. The typical Malwa pottery has a somewhat coarse core and a thick buff or orange slip. Designs were painted on in black or dark brown, usually just on the upper part of the pots. The pottery includes lotas, concave-sided bowls, channel-spouted bowls, and pedestalled goblets. Malwa ware is exceptionally rich in form and designs. Over 600 different kinds of motifs occur on the pots, mostly geometric, but some naturalistic. Plants, animals, and even humans are painted on the pots. There are representations of the blackbuck, bull, deer, peacock, pig, tiger, panther, fox, tortoise, crocodile, and insects.

Navdatoli does not show any signs of planning; the houses were built haphazardly, with lanes in between. People lived in circular or oblong wattle-and-daub houses with floors plastered with lime. Houses had wooden posts all around to support a roof that was probably conical. The walls were low; sometimes there was no wall, the sloping sides of the roof coming down to ground level. Mud was often plastered over screens of split bamboo. The diameter of the circular houses ranged from 1 to 4.5 m. The rectangular houses were 5–6 m long. Chulhas and storage jars were found in houses. At Nagda, on the banks of the Chambal, there was evidence of the use of mud-brick. At Eran, there was a massive mud fortification wall and a moat. More artefacts of stone than copper are found at Malwa culture sites, probably because of the scarcity of copper. There are lots of stone blades. Over 23,000 microliths were found in the Navdatoli excavations (Sankalia et al., 1958), most of them made of chalcedony, but a few also made of carnelian, agate, jasper, and quartz. The fact that all the tool types were evenly distributed in all layers and areas of the site suggests that every household in Navdatoli made its own tools. Some tools were hafted, others hand-held. Stone artefacts included saddle querns, rubbers, hammer stones, and mace heads or weights. Copper artefacts included flat axes, wire rings, beads, bangles, fishhooks, chisels, nail parers, thick pins, and a broken mid-ribbed sword. The axes had round indentation marks, similar to those found at Ganesh-war. An analysis of some of the copper objects revealed tin and lead alloying. Navdatoli also yielded beads of steatite, terracotta, faience, agate, amazonite, carnelian, chalcedony, glass, jasper, lapis lazuli, and shell. There were terracotta animal figurines and spindle whorls. Plant remains included wheat, barley, linseed, black gram, moong, lentil, anwala, ber, and khesari. Rice was found in the later levels. The faunal remains comprised bones of wild deer and domesticated cattle, sheep, goat, and pig.

MAP 5.4 MAJOR CHALCOLITHIC SITES IN MALWA AND THE DECCAN

Excavations at Malwa culture sites yielded some remains of religious or ritualistic activity. At Navdatoli, a 2.3 × 1.92 × 1.35 m pit was dug into the middle of the floor of a house of the earliest occupational phase. The sides and base of the pit were plastered with mud. Wood was found inside, and there were charred wooden posts at its four corners. This pit can be identified as a fire altar where sacrifices were performed. Another interesting discovery at Navdatoli was a huge storage jar decorated with a female figure (a goddess? a worshipper?) on the right, a lizard or alligator on the left, and what looks like a shrine in between. The shrine seems to have been associated with the lizard. There were four such shrines on the four sides, and the shoulder of the jar was ornamented with appliqué patterns. On the other side of the jar was a shrine with a tortoise to its left (the figure on the right cannot be made out). It can be noted that a shell amulet in the shape of a tortoise was found at Malwa culture levels at Prakash (in Maharashtra). A standing human figure with dishevelled hair on a fragmentary channel-spouted bowl found at Navdatoli is identified by some scholars as a proto-Rudra.

Mention may also be made of the bull figurines found at some Malwa culture sites. The evidence from Dangwada suggests the worship of bulls, trees, snakes, and female deities, and there are fire altars where sacrifices were probably performed. Malwa culture sites have given evidence of burials within houses. At Azadpur near Indore, there was a child burial under a house floor. The body was laid in a north–west orientation, with the feet cut off after death. A serrated blade and a small terracotta tablet were placed below the head, and a stone to its right. THE CHALCOLITHIC FARMERS OF THE DECCAN

The late Harappan and Malwa cultures

Discoveries at Daimabad suggest that the late Harappan culture extended into the Dec-can. Elsewhere in this region, the general chalcolithic cultural sequence consists of the Savalda culture, followed by the Malwa and Jorwe cultures (Dhavalikar, 1979b). The Savalda culture was discussed in Chapter 3. Here, we will look at the late Harappan phase, and more so at the Malwa and Jorwe cultures, with special reference to the sites of Daimabad and Inamgaon. The detailed excavation reports for both these sites provide an exceptionally detailed range of information about the lives of the early chalcolithic farmers of the Deccan. As mentioned earlier, the Malwa culture spread from central India to the Deccan. The main concentration of sites in the Deccan was in the Tapi valley, with fewer settlements in the Pravara– Godavari and Bhima valleys. The Malwa ware of the Deccan is a little different from that of central India. The fabric is fine, not gritty and unevenly baked, and the pots were uniformly fired at high temperatures. The typical forms are deep bowls and spouted vessels with flaring mouths (the latter are not found in central India). A coarse handmade red or grey ware, similar to that of the southern neolithic, was also used. Important Malwa culture sites include Daimabad, Inamgaon, and Prakash. Daimabad (Ahmednagar district, Maharashtra) is a deserted village on the banks of the Pravara, a tributary of the Godavari. It was excavated during 1976–79 by an Archaeological Survey of India team under the direction of S. A. Sali. This important site has a long, well-documented chalcolithic sequence. Period I (before 2300/2200 BCE) belongs to the Savalda culture, Period II (2300/2200– 1800 BCE) is late Harappan, Period III (1800–1600 BCE) has been labelled the ‘Daimabad culture’, Period IV (1600–1400 BCE) represents the Malwa culture, and Period V (1400–1000 BCE) the Jorwe culture (Sali, 1986).

B ONE KNIFE, DAIMABAD

In Daimabad Period II (late Harappan), the size of the settlement increased to about 20 ha. The houses were arranged on either side of a 30–50 cm thick wall made of black clay. The largest house measured 6.3 × 6 m. There was a grave lined with mud-bricks containing a skeleton laid out in an extended position. The body seems to have been originally covered with reeds of fibrous plants. The main pottery type was a fine red ware with linear and geometric designs painted on in black; the shapes included the dish-on-stand, bowl-on-stand, dishes, and vases. There was also a burnished grey ware, a thick, coarse hand-made ware, and a few specimens of ribbed bichrome and deep red wares. Two button-shaped seals with Harappan writing and four inscribed potsherds were among the singular discoveries. Other artefacts included stone tools such as microlithic blades, stone and terracotta beads, shell bangles, gold beads, and a terracotta measuring scale. The presence of copper slag indicated that copper was smelted locally. The plant remains included millets, gram, and moong —all of which were present in Period I—with horse gram making its appearance for the first time.

P OTTERY FROM DIFFERENT PHASES, DAIMABAD

There was a break in occupation for about half a century between the end of Period II at Daimabad and the beginning of Period III, which has been called the ‘Daimabad culture’. The typical pottery of Period III was a black-on-buff/cream ware. Other artefacts included microlithic blades, bone tools, beads, and a single piece of worked elephant tusk. Part of a copper-smelting furnace was found, as were three different types of burials—a pit burial, post-cremation urn burial, and symbolic burial. Hyacinth bean was the new addition to the plant remains.

www.pearsoned.co.in/upindersingh PLAN OF DAIMABAD COMPLEX WITH APSIDAL ‘TEMPLE’

Period IV at Daimabad belonged to the Malwa culture. Many structural remains of this period were identified. People lived in fairly spacious, usually rectangular mud houses, with mud-plastered floors, wooden posts embedded in the thick mud walls, and steps leading up to the doorway from outside. A house with two furnaces, one with a copper razor, was identified as a coppersmith’s

workshop. On the basis of the occurrence of fire altars, certain structures were tentatively identified as religious structures. An elaborate structural complex including a mud platform with fire altars of different shapes, and an apsidal temple associated with sacrificial activity were identified. There were 16 burials, either pit or urn burials. Twigs of a fibrous plant were laid out at the bottom of the pits. The artefacts of Period IV included microlithic blades, copper objects, faience beads, and terracotta and bone objects. The plant remains included barley, three kinds of wheat, ragi, lentils, pulses, and ber. Sugandha bela (Pavonia odorata) may have been used to make a perfume. FURTHER DISCUSSION

The Daimabad bronzes

In 1974, a farmer named Chhabu Laxman Bhil discovered a hoard of metal objects while digging at the base of a shrub in Daimabad village. The headman of nearby Ladgaon village reported the discovery to the police. The objects were subsequently acquired by the Archaeological Survey of India from the district authorities. The hoard consisted of the following four objects: A 1.man (16 cm high) standing on and driving a simple two-wheeled chariot (45 cm long and 16 cm wide) attached by a long pole to two yoked oxen standing on two cast copper strips. There is a small figure of a dog standing on the central pole of the guard of the chariot. The man holds the upper horizontal bar of the guard with his left hand and a long stick curved at both ends in his right. His chest and belly are somewhat elongated. His upper chin and lower lip are protruding. He has a short nose, wide open eyes, and curved eyebrows. His curly hair is parted in the middle and rolled into a bun at the nape of his neck. His knees are slightly bent and his penis is surmounted by four hoods of a cobra; a2.water buffalo (31 cm high and 25 cm long) on a four-legged platform attached to four solid wheels; an 3. elephant (25 cm long) on a similar platform (27 cm long), but with the axles and wheels missing; and a4.rhinoceros (25 cm long and 19 cm high) standing on the axles of four solid wheels.

The objects were solid cast and heavy, weighing 60 kg altogether. They reveal considerable casting skill and aesthetic finesse. Chemical analysis showed that they were made of bronze with varying, but low, tin content.

Although the hoard was not found in the course of the initial excavation, later excavations near the find-spot correlated its find-spot to the late Harappan phase. These artefacts do not seem to have been utilitarian objects. They may have had a religious or ritualistic significance, and the fact that they are on wheels suggests that they were part of a procession. S. A. Sali was tempted to identify the human figure as the god Shiva, lord of the beasts, but this is very conjectural. Metal figures of this kind have not been found elsewhere in India, and the Daimabad hoard remains an enigma. SOURCE Sali, 1986: 477–79

Inamgaon (in Pune district) is located on a terrace of the Ghod, a tributary of the Bhima. It is one of the largest, most intensively and extensively excavated chalcolithic sites in Maharashtra. The excavations, undertaken by a team from Deccan College, Pune, under the direction of M. K. Dhavalikar, H. D. Sankalia, and Z. D. Ansari, lasted for 12 seasons between 1968 and 1983, and provided a lot of information about the lives of the farmers who lived in this place hundreds of years ago. Period I (c. 1600–1400 BCE) belonged to the Malwa culture, Period II (c. 1400–1000 BCE) to the early Jorwe culture, and Period III (c. 1000–700 BCE) to the late Jorwe culture. Here, we will focus on Period I (Dhavalikar et al., 1988). The floors of as many as 134 houses were exposed in the course of the Inamgaon excavations. Out of the 32 houses of Period I, 28 were rectangular, 1 circular, and 3 were pit dwellings. The rectangular houses had rounded corners with very low mud walls, over which must have been a wattle-and-daub construction and a thatched, conical roof. These are the kinds of houses that villagers of this area live in even today. The houses were spacious, 8 × 5 m on the average, and were often divided into two by a wattle-and-daub screen. Oval-shaped hearths for cooking were found inside. Sometimes, there was an additional hearth in the courtyard; this may have been used for roasting meat. There were two kinds of storage structures—overground bins made of wickerwork and silos dug into the ground, inside or outside the houses.

INAMGAON: P ERIOD I (MALWA PHASE) POT

The early chalcolithic farmers of the Deccan obtained their food by farming, hunting, and fishing. The fact that barley was the main crop is not surprising, considering this area does not get the amount of rainfall required for wheat cultivation. The faunal remains at Inamgaon included the bones of domesticated animals such as humped cattle, buffalo, goat, sheep, dog, and pig. The bones of wild animals included those of sambar, chital, blackbuck, hare, and mongoose, as well as birds, reptiles, fish, and molluscs. Tools of stone and copper have been found at various Malwa culture sites. Siliceous stone such as chalcedony and agate were mostly used, and the tools were usually made on blades or flakes. Polished stone axes occur rarely. Microwear analysis has identified tools used for different purposes —plant working, meat cutting, antler or bone working, and hide scraping. Copper artefacts included knives, chisels, fishhooks, axes, and ornaments such as bangles and beads. At Inamgaon, there were lots of beads and pendants, mostly of terracotta, jasper, ivory, and carnelian; also of shell, steatite, faience, paste, amazonite, serpentine, cipper, gold, and calcite. Among the semi-precious stones, jasper and carnelian, which were locally available, were used more than those obtained from distant sources. The fact that shell beads were found at Inamgaon is interesting, as this is an inland site, with the sea about 200 km away. In Period I at Inamgaon, the only burials discovered were child burials. In all three periods, children were buried in pits in two urns placed mouth to mouth horizontally. Human and animal terracotta figurines were found at all levels. The features and contexts of some of the female figurines suggest a possible cultic significance. The large number of bull figurines suggests that this animal may have been venerated.

The Jorwe culture of the Deccan

The Jorwe culture was first discovered at the site of Jorwe, and was later found to have extended over a large area, covering practically the whole of modern Maharashtra, except the coastal Konkan district. The Pravara–Godavari valleys seem to have been the nuclear zone of this culture. The peripheral zone extended up to the Tapi river in the north and the Krishna in the south. The main excavated sites are Daimabad, Inamgaon, Theur, Songaon, Chandoli, Bahal, Prakash, Jorwe, and

Nevasa. Prakash is the largest Jorwe site in the Tapi valley, Daimabad in the Godavari valley, and Inamgaon in the Bhima valley. All three settlements were 20 ha or more in size. These large sites represented permanent agricultural villages. Jorwe, Bahal, and Nevasa are medium-sized settlements. The average Jorwe culture settlements were, however, much smaller—usually 1–2 ha. This category includes Walki and Gotkhil, which seem to have been sites of predominantly seasonal agricultural-cum-pastoral occupation, while Garmals appears to have been a temporary camp site located close to a source of chalcedony. These facts point to the existence of a settlement hierarchy. Radiocarbon dates from Nevasa, Chandoli, and Songaon suggest a time frame of c. 1300–1000 BCE. At Inamgaon, on the other hand, the dates for the early Jorwe culture are c. 1400–1000 BCE, while the late Jorwe phase is dated c. 1000– 700 BCE.

P ERIOD II (EARLY JORWE) TERRACOTTA LAMP

Jorwe pottery is fine, well baked, and rich in form and design. The pots have a red or brightorange matt surface on which designs—usually geometric—were painted in black. The shapes include a concave-sided bowl with sharp carination, spouted jar with flaring mouth, and high necked jar with globular profile. There is also a coarse, handmade red and grey pottery. Oval lamps of red and grey ware are also found. A pottery kiln has been identified at Inamgaon.

P OTTERY FROM DIFFERENT PERIODS, P RAKASH

At Daimabad, Period V represents the Jorwe culture. The settlement grew to about 30 ha in this period. There were traces of a mud fortification wall with bastions. The excavators identified the houses of a butcher, lime maker, potter, bead maker, and merchant. There was an elliptical structure with approach paths plastered with cow dung; clusters of pots seem to have contained offerings including copper objects, shaped stones, and tool hafts made of cattle bones. The artefacts included microliths, copper objects, beads, and terracotta figurines. There was also a terracotta cylinder seal depicting a horse-drawn cart or chariot. The crop list of this phase is more or less the same as that of the preceding period, with the addition of three new types of millets (kodon millet, foxtail millet, and jowar). Out of the 48 burials, 44 were urn burials, three were extended pit burials, and one was an extended burial in an urn. One of the curious things about the Daimabad burials belonging to all phases is that except for one burial belonging to the late Harappan phase, all of them were of infants and young people. An analysis of teeth remains of the skeletons showed the presence of dental caries, gross enamel hypoplasia, tartar accumulation, and calculus deposits. There was one instance of infantile scurvy. At Inamgaon, Periods II (early Jorwe) and III (late Jorwe) revealed rectangular houses, similar to those of Period I (Malwa culture). The fact that the houses were laid out almost in rows, with an

open space (perhaps a lane or road) in between, suggests an element of planning. The houses had fire pits, usually with a flat stone at the bottom daubed with mud, serving as a stand for the cooking vessel. The nitrogen in the soil in the courtyards shows that animals were tied here. On the basis of the discoveries in various houses, it was possible to reconstruct who lived where. The houses of artisans such as potters, goldsmiths, lime makers, bead makers, and ivory carvers were on the western periphery of the settlement, while the farmers and other well-to-do people lived in the middle. A large, five-roomed Period II structure in the centre of the settlement was identified as the house of the ruling chief. This had a granary next to it. In Period III, the chief seems to have lived in the eastern part of the settlement, on the river front. One of the structures was identified as a granary or a temple for fire worship. Other public works that must have involved community effort included a stone embankment wall, geared towards protecting the settlement from floods and for storing water. Irrigation channels were also identified. Inferences about the social and political organization of the people were made on the basis of the details of material evidence. The settlement layout and the burials suggest a ranked society. The subsistence base at Inamgaon included farming, hunting, and fishing. Grains/ seeds of barley, wheat, lentil, kulthi, grass pea, ber, and a very few grains of rice were found. Barley was the main crop, followed by wheat. The domesticated animals included cattle, buffalo, goat, sheep, pig, and horse (the horse is rare, occurring towards the end of Period II). Cattle were the most important domesticated animal throughout. People hunted animals such as deer. The horse, ass, and four-horned antelope were the new animals added to the list of faunal remains known from Period I. The evidence of fishhooks indicates fishing. Period II (early Jorwe) was the most prosperous period at Inamgaon and reflected an intensification of farming and animal domestication. Irrigation was probably used to grow winter crops such as wheat, peas, and lentils. The population of the settlement also seems to have increased. In Period III (late Jorwe), on the other hand, there was a gradual but drastic change in productivity. The cultivation of winter crops such as wheat and pea declined, and the reliance on hardier crops such as barley, lentil, and horse gram increased. There was also a greater dependence on hunting and collecting wild plants. A rich assemblage of artefacts has been found at Jorwe culture sites. Blade flakes were made of siliceous stones such as chalcedony and agate. Polished stone axes and chisels of dolerite occurred rarely. Ornaments included beads of chalcedony, agate, carnelian, and jasper. Gold occurred occasionally in the form of beads at Daimabad and spiral ear ornaments at Inamgaon. At Inamgaon, pottery kilns and many lime kilns were identified. Copper was scarce, and was used sparingly for axes, chisels, knives, and fishhooks, and also for bangles and beads. A furnace for extracting copper from ore was found at this site.

P ERIOD III (LATE JORWE) POTTERY, INAMGAON

FURTHER DISCUSSION

Food, nutrition, and health among the people of Inamgaon

Scientists conducted trace element analysis on 165 human bone samples found in the Inamgaon burials. The aim was to investigate the relationship between subsistence, age, status, and changes in diet over time. They reached the following conclusions: The 1. people of the early Jorwe phase consumed a diet containing relatively more agriculturally produced plant food, animal food, and dairy food. The 2. late Jorwe phase people had a diet rich in animal food, fish, and locally gathered plants. Burials 3. were generally under house floors, sometimes in the courtyard. The bones found in the rectangular houses in the central part of the mound reflected a more nutritious diet than those found in the round huts. The diet-related elements suggest some sort of status difference within the community. There 4. does not seem to have been any difference in the diet of males and females in any phase. The 5. rise in weaning age in the late Jorwe period may be associated with a gradual shift from an agricultural to a semi-nomadic lifestyle. The 6. microscopic analysis of the skeletons showed evidence of infantile scurvy, other types of degenerative joint diseases, and fractures. The 7. dental health of the people was good—the incidence of dental caries and gross enamel hypoplasia was low, but people seem to have lost their teeth somewhat early in life.

SOURCE V. D. Gogte and Anupama Kshirsagar in Dhavalikar et al., 1988, Vol. 1, Part 2: 991– 98

FURTHER DISCUSSION

Goddesses with and without heads

Female figurines of clay—both baked and unbaked—have been found at In-amgaon and Nevasa. Some of them are headless. It is likely that these figurines represented goddesses connected with fertility. At Inamgaon, an interesting discovery was made under a house floor belonging to Period II (the early Jorwe phase). There was a female figurine in a clay receptacle. Over this was a headless female figurine and a bull. All the figurines were unbaked, showing that they were meant for temporary use. The headless figurine had a hole in its abdomen, and the bull had a hole in its back. When a stick was inserted through both the holes, the headless female figurine was found to sit perfectly on the bull’s back! The fact that the figurines were buried under a house floor suggests they were part of an important household ritual. It is possible that the headless figurine represented a goddess connected with fertility, childbirth, or the welfare of children. SOURCE Dhavalikar et al., 1988, Vol. 1, Part 1: 571–79

F IGURE 5.6 INAMGAON FIGURINES

Inferences can be made about networks of exchange on the basis of the evidence from Jorwe levels at Inamgaon. Gold and ivory were probably obtained from Karnataka, conch shell from the Saurashtra coast, and amazonite from Rajpipla in Gujarat. Apart from tapping the locally available copper and nearby chalcopyrites, this metal may also have been obtained from Rajasthan and from

the Amreli district in Gujarat. Haematite, marine fish, and marine shell must have come from the Konkan coast, and hyacinth bean from the upper Ghod valley. Both these regions were occupied by hunter-gatherers, to whom the chalcolithic farmers may have offered beads and pottery in exchange. Within the Jorwe culture zone, Inamgaon and Daimabad may have been major suppliers of pottery to other settlements. The occurrence of Jorwe pottery at Navdatoli in central India and T. Narsipur in Karnataka suggests that the Jorwe people had contact with neolithic farmers of north Karnataka and chalcolithic communities of central India. There were also connections with the late Harappans and lustrous red ware users of Gujarat. The precise nature of these contacts is, however, not clear.

P ERIOD III (LATE JORWE) TERRACOTTA FIGURINE, INAMGAON

At Jorwe culture sites, adults were usually buried in an extended position, children in urns placed horizontally mouth to mouth. Burial pits were dug into house floors, occasionally in the courtyard. An unusual feature was that in the case of adults, the feet were deliberately cut off, perhaps to keep the spirit of the deceased within the house. At Inamgaon, there was a curious urn burial in the courtyard of the large five-room house. The burial belonged to the transitional phase between Periods II and III, and is dated c. 1000 BCE. The urn was made of unbaked clay and had four stumpy legs. The jar was 80 cm high and 50 cm wide, and had a painting of a boat with long oars. One of its sides was modelled to resemble a woman’s abdomen. Inside was the skeleton of a male, about 40 years old, seated in foetal position with the knees flexed up to his knees, his chin pressed down to his chest. Unlike the skeletons found in other burials, his feet were intact, not cut off. Close to this burial, but belonging to an earlier phase, was a burial consisting of a four-legged jar along with a similar jar cut into half and placed by its side. It contained no skeletal remains, only a painted globular jar with a lid. This might have been the symbolic burial of a person whose body could not be found, perhaps someone who had died in battle. Going by their location and nature, these two

burials seem to have been those of important people; perhaps they represent two generations of ruling chiefs. Recent excavations at Walki (Pune district, Maharashtra) on the Bhima river have brought to light a new Jorwe culture site. A total of 106 structural features were identified here. The houses, most of them circular, were arranged in clusters of five or six huts. The high nitrogen content in some floors points to animal dung, indicating that animals used to be tethered here. Some of the floors may have been used as threshing floors. In each hut cluster, there was a circular silo with lime-plastered sides and base, probably used to store grain. The fact that these huts did not have walls suggests they were not occupied in the rainy season. There were some other large, squarish or rectangular huts with low mud walls in the central part of the habitation. These seem to have been occupied all year. X-ray diffraction analysis of the pottery suggests that Inamgaon, which is located 27 km away, provided earthen pots to Walki. Two unique agricultural implements were found here—a bone ploughshare and a seed drill made of antler. Walki seems to have been a pastoral-cum-agricultural satellite farmstead of Inamgaon (Shinde, 1994: 171). By c. 1000 BCE, practically all Jorwe settlements in the northern Deccan were suddenly deserted, although the one at Inamgaon continued till c. 700 BCE. One theory is that the settlements were abandoned because of increasing aridity, which may have led to food scarcity. On the other hand, the evidence of burnt structures points to some other sort of disaster. At Inamgaon, the small huts and coarse pottery at late Jorwe levels contrast sharply with the spacious homes and fine pottery of the early Jorwe phase. They suggest increasing poverty, a time of trouble. Recent studies of the late Jorwe phase, especially at sites such as Sheriwadi, Pimpalsuti, and Talegaon in the Bhima basin have brought out the connections (e.g., in pottery) between the late Jorwe culture and the succeeding megalithic and early historic phases in the Deccan, but the relationship between these various phases is not at present very clear. NEOLITHIC–CHALCOLITHIC SITES OF SOUTH INDIA

Early neolithic sites in Karnataka, Tamil Nadu, and Andhra Pradesh were discussed in Chapter 3. Reference was also made there to the beginning of the chalcolithic phase in the Kurnool district of Andhra Pradesh. We take the story on from there. The early occupation of sites such as Utnur, Watgal, and Budihal comprised the first stage of the neolithic in South India. The second stage is represented at some of the older sites, as well as in a number of new sites that came to be occupied in this period. Watgal is one of the older sites which continued to be occupied in the 2nd millennium BCE. Period III at this site is dated post-2000 BCE. This level revealed three burials and many large storage pits. Artefacts included BRW sherds, agate beads, carved steatite earrings, human and animal figurines, six copper/bronze artefacts, and three iron objects that may have belonged originally to a later period. Horse gram and ragi were the new grains in this period. Period IV was post-1500 BCE. Artefacts included terracotta figurines (fewer than in Period III) and beads of lapis lazuli, dolerite, copper/bronze, and marine shell. There were megalithic chamber graves. One of these contained an iron knife, a small piece of gold-wrapped silver wire, and various kinds of pottery spread out over four large stones. The infant burials were both of the urn and extended types. The second stage of the southern neolithic–chalcolithic is also represented at the earlier levels of sites such as Sanganakallu, Brahmagiri, Piklihal, Maski, Tekkalakota, and Hallur (all roughly falling

within the time bracket of c. 2100–1700 BCE). Settlements were established on top of granite hills, on levelled terraces on hillsides or on plateaux between the hills. People lived in round wattle-anddaub huts. Stone tools such as celts and blades were made and used, but there is also evidence of many copper and bronze artefacts. The Karnataka region is well-known for its gold mines, so it is not surprising that gold objects have been found at Tekkalakota. The pottery range of this stage is similar to that found at the earlier neolithic sites in the area, with some new features such as perforated and spouted vessels and the roughening of the outer surface of pots. Extended burials were located within the habitation area and usually contained grave goods such as stone tools and pottery. Children were buried in urns.

NEOLITHIC CELT, B RAHMAGIRI

MAP 5.5 SOME NEOLITHIC–CHALCOLITHIC SETTLEMENTS IN SOUTH INDIA

The third phase followed the second at these sites. Stone tools continued, but there was an increase in the number of copper and bronze tools such as chisels and flat axes. The new elements in pottery were a grey and buff ware with a harder surface, and there was also a wheel-made unburnished ware with purple paint. There are few radiocarbon dates for this phase, but it may be roughly dated c. 1500–1050 BCE. The upper levels of most of the sites merge into a megalithic phase. At Sanganakallu (Bellary district), the earlier neolithic phase was a-ceramic and devoid of copper; this was followed by a phase with copper tools and wheel-made pottery. In both phases, there were ground and polished stone tools, microliths and bone points, and chisels. The pottery of the neolithic–chalcolithic phase included black-on-red ware (some painted with designs in red ochre) and pale grey, burnished grey, and brown wares. There was also a coarse brown and black pottery. Terracotta figurines mostly comprised bulls and birds. Bones of cattle, sheep, goat, and dog were identified. The neolithic phase at Sanganakallu seems to have begun in about 2000 BCE. At Brahmagiri (in the Chitradurg area), neolithic Period IA was marked by remains of wattle-anddaub huts with wooden or bamboo posts, supported by stone. The artefacts included ground and polished stone tools, microlithic blades, and grey pottery (mostly handmade). Copper–bronze objects made their appearance in Period IB. Extended burials of adults and urn burials of children were found at the site. At Piklihal, the lower levels yielded floors of circular huts, neolithic tools, and microlithic blades. The pottery was handmade and consisted mostly of grey and burnished grey wares. There were also some specimens of black, buff, red/brown wares, some with paintings in red ochre and

purple. Terracotta figurines of humans, animals, and birds were discovered. Bones of domesticated cattle, goat, and sheep were found. The upper neolithic levels gave evidence of rectangular wattleand-daub huts, one of them with a hearth inside and a saddle quern outside. The artefacts included fragments of a copper bowl and pottery made on a slow wheel. New pottery types included painted black-on-red ware and a green ware with mottled surface. Beads of carnelian, shell, and magnesite were found. NEW DIRECTIONS IN RESEARCH

Pictures on stone

Pictures made on granite rocks can be seen in many places in Karnataka and Andhra at sites such as Kupgal, Piklihal, and Maski. They are difficult to date, but a rough chronology can be worked out on the basis of style, content, and weathering. Some of the images may belong to the mesolithic, some to the neolithic– chalcolithic stage, and others are much later. Many of the pictures were made by crayoning (rubbing dry colours on to the stone surface) rather than painting. There are also many rock bruisings, made by hammering or pecking the motifs onto the rock surface. Cattle are the dominating theme. At Kupgal (Bellary district, Karnataka), there is an outcrop of granite hills, locally known as Hiregudda (literally, ‘big hill’). These hills have hundreds of rock pictures, mostly bruisings, ranging from neolithic to modern times. Humped cattle with long horns are the most common theme. They are usually depicted singly, occasionally in pairs; they sometimes have anthropomorphic figures riding on them or surrounding them, with bows and arrows in hand. Individual anthropomorphic figures are the next frequently occurring theme. Many of them are ithyphallic. There are also several scenes depicting heterosexual intercourse. There are also people standing in a chain-like formation, usually interpreted as dancers. Other less frequently occurring motifs include the elephant, tiger, deer, buffalo, birds, footprints, and abstract designs. In general, the scenes tend to be small and simple; large, complex scenes are absent. N. Boivin’s study of the Kupgal rock art points out that some of the locations where the images were made must have been difficult to access for artists as well as viewers. Boivin suggests that certain images seem to celebrate male prowess and sexuality, as well as links between men and cattle. Perhaps they were made by young men associated with cattle herding, maybe even cattle raiding. Kupgal was evidently a major stone quarrying and tool production centre. Another possibility is that the pictures were made by men who came here to quarry stone. The making and viewing of these pictures may have been part of ritualized activity, involving ‘rock music’ as well. This is suggested by the fact that some of the dolerite boulders at the site seem to have been used for percussion purposes—they have grooves which emit a deep sound like a bell or gong when hit with a granite stone. The point emphasized by Boivin is that it is necessary to look beyond the images in isolation and to take into account the wider physical and social landscape. It can be noted that neolithic ash

mounds once stood at the base of the Kupgal hill. It is also important to note the fact that this rock art site continues to hold a special meaning for groups living in the area today. Rock bruisings are still made; cattle continue to be the main subject, although the style has changed. SOURCE Boivin, 2004

At Maski, Period I is neolithic–chalcolithic. This yielded ground and polished stone tools, microlithic blades, and a fragment of a copper rod. There were beads of carnelian, agate, amethyst, chalcedony, shell, coral, glass, and paste. The pottery included a dull-red ware and a pinkish buff ware. There were also a few potsherds of painted black-on-red ware and a dull grey ware with incised designs. Animal bones included those of short-horned humpless cattle, buffalo, sheep, and goat. Rock bruisings and paintings have been found in the area. At Tekkalakota (Bellary district), the early neolithic phase was marked by a handmade grey pottery, both plain and burnished, in some cases with designs painted on in black, purple, or violet. The second phase had black-and-red and dull brown pottery. Apart from the typical neolithic stone tools, both phases yielded microliths, bone tools, beads of steatite and semi-precious stones, and copper and gold artefacts. The structural remains suggested that people lived in round huts with conical roofs, sometimes sup-ported by stone s at the base. Extended and fractional burials in urns were found. Animal bones comprised those of cattle, sheep, and tortoise. Charred grains of kulthi and hyacinth bean were identified. Calibrated dates for the site are c. 2100–1800 BCE.

P OTTERY FROM DIFFERENT PERIODS, MASKI

Hallur is located on the banks of the river Tungabhadra in the Dharwar region. Period I is neolithic and is divided into an earlier and later phase. The floors of the round wattle-and-daub huts were made of stone chips and river sand. The first phase mainly had handmade plain and burnished grey wares, as well as some reddish-brown ware with purple paintings. In the second phase, a painted Black and Red Ware made its appearance. The stone tools comprised ground and polished tools and microliths. Other artefacts included copper fishhooks and double axes, as well as beads of steatite, quartz, bone, and shell. A double urn burial was discovered. Animal bones comprised those of cattle, sheep, and goat, with the addition of horse bones in the second phase. Calibrated dates for Hallur Period I range between c. 2000 and 1400 BCE. The subsistence base of the southern neolithic–chalcolithic communities included agriculture, animal domestication, and hunting. Horse gram and ragi were found at Tekkalakota and Hallur. Paiyampalli yielded horse gram and green gram. These are the staple food crops of the area even today. The neolithic–chalcolithic farmers probably made terraces on the hillsides for cultivation. The numerous cattle bones, many with cut marks, found at all sites reflect the importance of cattle rearing. There are numerous figurines of humped cattle, and these animals also occur in rock paintings at sites such as Maski. Recently, mesolithic and neolithic paintings of humped bulls in a distinctive style have been reported in the rock shelters at Budagavi (Anantapur district, AP). A recent re-investigation of the plant and animal remains of seven neolithic sites in Karnataka and Andhra Pradesh (Korisettar et al., 2001) has provided detailed information regarding patterns of subsistence across the southern neolithic sites. Evidence from seven sites was examined—Hallur,

Sanganakallu, Tekkalakota, Hiregudda, Kurugodu, Hattibelagallu, and Velpumadugu. Cattle were the most important domesticated animals; goats and sheep were less important. Chickens may also have been domesticated. There were some water buffalo bones, but it is not clear whether they belonged to domesticates. Wild animals that were hunted included the antelope, deer, and pig. There was occasional use of freshwater resources such as fish and molluscs, even at sites that were a bit far from rivers. Measurement of cattle bones indicated that the cattle herded by southern neolithic people were of medium to medium–heavy build. The cropping pattern consisted of an emphasis on kharif (summer) crops such as small millets, pulses (moong bean), and horse gram. Additional crops such as wheat, barley, pigeon pea, pearl millet, and hyacinth bean were selectively grown. Wheat and barley must have been winter crops. Fruits and tubers must have been gathered during the dry season. The evidence of plants cultivated in different seasons of the year matches the evidence of the thickness of the occupational deposit at these sites, indicating that they were occupied all year round. FROM COPPER TO IRON: EARLY IRON AGE CULTURES OF THE SUBCONTINENT

All over the world, the iron age comes after the copper-bronze age. The transition from copper to iron raises a number of questions: Was iron smelting an accidental by-product of copper smelting? Were the smelting and working of iron well within the range of the technical expertise of coppersmiths, or did they involve a gigantic technological leap? After using metals such as copper and bronze for so many centuries, why did some communities start making and using iron tools? There are certain important technological aspects to these issues. Copper melts at 1083°C, while iron melts at the much higher temperature of 1534°C. Therefore, the smelting of iron requires furnaces that can maintain very high temperatures. Iron ore is associated with many more impurities than copper ores and requires the maintenance of a number of conditions for successful smelting. A temperature of 1250°C has to be maintained in the furnace for the separation of unwanted gangue materials from smelted material. A good blast of air has to be supplied to the furnace, along with constant supplies of fuel. Another important prerequisite is the efficient use of fluxes. A flux is a smelting aid, a substance added to molten ore, which combines with impurities to form slag that can be extracted. The technology of carburization—heating the iron in association with carbon to make steel—was another important step that had to be mastered before iron came into widespread use. The evidence of iron lumps, pieces, or artefacts from chalcolithic levels at sites such as Lothal, Mohenjodaro, Pirak, Allahdino, Ahar, and Gufkral indicates that certain chalcolithic communities were familiar with iron and were able to smelt it from the ores. Iron may have initially been extracted accidentally in the copper-smelting furnace when sufficiently high temperatures were attained, if there was iron oxide in the copper ore, or if a haematite flux was used to smelt these ores. But this represented an initial, experimental stage. The large-scale use of iron and the achievement of technical finesse in iron working was something that happened gradually and at a later stage. Copper ores are not as widely available as iron ores, and it is possible that a shrinking of trade networks may have given an impetus towards the increasing replacement of copper with iron. This was especially so once the requisite technological knowledge of iron smelting and working had been achieved, and people realized the superiority of iron over copper and bronze in terms of hardness and durability.

The beginning of iron technology is not the same thing as the beginning of the iron age. A distinction has to be made between the presence of a few iron objects at a site and a significant use of iron. But how is ‘significant use’ to be assessed? This has to be done on the basis of the total volume of iron artefacts in themselves and in relation to those of other metals and materials, and by their nature and purpose. It is necessary to try to identify when people started using iron for everyday activities, especially for production purposes. In the case of the agricultural societies, it is necessary to try to identify when iron implements started being used in agricultural operations for making tools such as ploughs, hoes, and sickles. This marks the beginning of the iron age.

MAP 5.6 EARLY FINDS OF IRON IN THE SUBCONTINENT

As pointed out by Chakrabarti (1992: 33), iron ores suitable for pre-industrial smelting are found in all parts of the subcontinent, leaving aside the alluvial river valleys. Evidence from later Vedic texts (cited in earlier sections in this chapter) suggests familiarity with iron and the use of iron in

agriculture in the Indo-Gangetic divide and upper Ganga valley in c. 1000–500 BCE. The evidence from archaeology gives more detailed and specific evidence for the beginning of iron technology and the beginning of the iron age in various parts of the subcontinent. Although lists of artefact types are available from several sites, more information on iron-smelting and iron-working sites is required. At least six early iron-using centres can be identified in the subcontinent: Baluchistan and the north-west; the Indo-Gangetic divide and the upper Ganga valley; Rajasthan; eastern India; Malwa and central India; Vidarbha and the Deccan; and South India. All these centres are located in or near iron ore resources and all of them have given evidence of pre-industrial smelting. There is a widely prevalent but misplaced belief that iron technology was introduced into the subcontinent by the IndoAryans. Chakrabarti’s analysis indicates that there is no evidence that iron technology diffused into the Indian subcontinent from West Asia or anywhere else. The use of iron in central and South India seems to have started earlier than in the north-west or the Ganga valley, and this metal seems to have entered the productive system in most parts of the subcontinent by c. 800 BCE. However, recent evidence from certain Uttar Pradesh sites has altered part of this picture dramatically. The following section summarizes the evidence of early iron age zones in the subcontinent. Certain regions do not find mention, either because they have not been explored properly or because they are areas where iron made its appearance at a somewhat later date. For instance, in Assam, Orissa, and Gujarat, there is no evidence of iron before the historical period. The picture in the Punjab plains and Sindh is unclear.

SEE PP . 248–49 FOR INFORMATION ON RECENT IRON FINDS IN UTTAR PRADESH A CLARIFICATION ABOUT THE INDIAN MEGALITHS

Megaliths have been mentioned in earlier sections of this chapter, and they will be mentioned even more frequently in connection with the beginning of iron technology in peninsular India. The word ‘megalith’ comes from two Greek words, megas meaning great or big and lithos meaning stone. Megaliths include different kinds of monuments that have one thing in common—they are made of large, roughly dressed slabs of stone. Such monuments have been found in many parts of the world—in Europe, Asia, Africa, and in Central and South America. In the Indian subcontinent, they occur in the far south, the Deccan plateau, the Vindhyan and Aravalli ranges, and the north-west. The practice of making megaliths continues among certain tribal communities of India such as the Khasis of Assam and the Mundas of Chotanagpur. The term megalithic culture refers to the cultural remains found in the megaliths and from the habitation sites associated with them. Megaliths once used to be considered the dominant feature of a homogeneous, independent, and distinct culture. Such a view is no longer accepted. In view of the significant variations in associated cultural remains, it is necessary to use the plural term ‘megalithic cultures’ rather than the singular ‘megalithic culture’. Megaliths reflect certain burial styles that emerged at different times in different places and continued for quite some time. The origins of some of these burial practices can be traced to a neolithic–chalcolithic context. For instance, pit and urn

burials are found in the South Indian neolithic–chalcolithic sites, and two burials marked by stones have been found at Watgal. It may also be noted that a sarcophagus burial occurred in the upper level of the chalcolithic Jorwe phase at Inamgaon. The megalithic chamber tombs, however, appear to be a new development. The three basic types of megaliths are the chamber tombs, unchambered tombs, and megaliths not connected with burials (Sundara, 1975: 331–40). The chamber tombs usually consist of a chamber (the size and shape of which may vary) composed of two or four vertical slabs of stone (known as orthostats), topped by a horizontal capstone. If the chamber is underground, it is known as a cist. If it is partly underground, it is known as a dolmenoid cist. If it is fully above the ground, it is known as a dolmen. Chamber tombs can have a hole known as a ‘port hole’ in one of the vertical slabs. They may also have a passage leading up to them. The chamber is sometimes divided into sections by vertical slabs called transepts. The chamber tombs include the topikals (literally, ‘hat stones’) and kudaikals (literally, ‘umbrella stones’), which are found in Kerala and Karnataka. In the topikals, the burial urn is placed in an underground pit and is covered by a low, convex, circular capstone. In the kudaikals, the urn is placed in a chamber consisting of four orthostats capped by a large hemispherical capstone.

TOPIKAL, COCHIN

SARCOPHAGUS IN DOLMENOID CIST, SANUR

The unchambered burials are of three types—pit burials, urn burials, and sarcophagus burials. In pit burials, the funerary remains are buried in a pit. If a pit burial is marked by a circle of large stones, it is known as a pit circle. If it has a heap of large stones piled on top, it is known as a cairn. If both a stone circle and piled-up stones are present, the burial is known as a cairn stone circle. A pit burial marked by a single large standing stone slab is called a menhir. A sarcophagus burial

consists of a terracotta trough (often with legs and lid) containing the funerary remains. Urn burials consist of funerary remains placed in a large pot or urn, the mouth of which is sometimes covered by a stone slab. Urn and sarcophagus burials are often included among megalithic burials, even if they are not marked by stones, as are burials in rock-cut caves. Not all megaliths are connected with burials. Some of them consist of alignments of large stones arranged in a geometric pattern. Although such monuments seem to be related to the megalithic tradition, their precise function and significance is not always clear.

F IGURE 5.7 DIFFERENT TYPES OF MEGALITHIC MONUMENTS (AFTER G HOSH, 1989)

It is easier to describe the shape and size of the megaliths than to understand the beliefs they reflect. These structures must have been an important part of the lives and belief systems of the people who constructed them. Unlike the burials of the neolithic– chalcolithic phase, which tend to be within the habitation, megalithic burials are located in a separate area. The separation of the abodes of the living and the dead is significant, and is indicative of a shift in social organization. The megaliths reflect many different kinds of funerary practices—extended, fractional, post-excarnate, and post-cremation burials. There are instances of graves containing the remains of more than one person. Some group burials may represent family vaults. But cases where there are no signs of repeated opening are suggestive either of simultaneous death or ritual suicide. The presence of grave goods—weapons, pottery, ornaments—suggests a belief in afterlife. Some of the megaliths are clearly funerary sites, while others may have been memorials for the dead. Reference was made in earlier sections to the megaliths in the Vindhyas, which belong to a preiron chalcolithic context. The megaliths of peninsular India, on the other hand, are generally associated with iron. Not all megalithic sites are contemporaneous. Some are as early as c.1300 BCE, while others are as late as the early centuries CE. A C-14 date for the terminal date of the megaliths at Adichanallur places it as late as the 12th century CE! The important thing to remember is that in view of their extensive distribution and the wide range in their dates and contexts, the megaliths cannot be treated as representing a single, homogeneous, or contemporaneous culture.

F IGURE 5.8 B LACK AND RED WARE FROM MEGALITHIC SITES IN THE DECCAN AND SOUTH INDIA THE NORTH-WEST

Iron objects of various types—vessels, javelin heads, sword blades, arrowheads, spearheads, a horseshoe, and fishhook—have been found in cairn burial sites in Baluchistan such as Damba Koh, Jiwanri, Gatti, Nasirabad, Zangian, Mughal Ghundai, and Bishezard. It is, however, difficult to date these burials. Some scholars date them between c. 1100 and 500 BCE, but they may actually be much later. At Pirak in the Kachi plain of Baluchistan, there was a limited amount of iron in Level VI; iron artefacts increased in Levels IV and III. Arrowheads were the only iron artefact type. A blacksmith’s furnace shows that iron objects were made at the site. There was basic cultural continuity in pottery types and stone blades between the chalcolithic and early iron-bearing levels. However, a new type of pottery—a grey or black ware—made its appearance. The excavated area of Level IV revealed a set of rooms within an enclosing wall. The niches and doors had wooden lintels. There were ovens and fireplaces, and a few storage jars were found half-buried in the ground. In Level III, the houses were rebuilt, and the larger number of fireplaces, ovens, and artefacts may indicate an increase in craft activity. Some terracotta seals with compartmented designs and beads decorated with zigzag and circular patterns were also found. There were a large number of bone points, mostly made of

antler, frequently decorated with an incised circlet on each side. The earliest evidence of iron at Pirak can be dated between c. 1000 and 800 BCE. In an earlier section in this chapter, reference was made to the Gandhara Grave culture in the North-West Frontier Province of Pakistan and the cultural sequence in the Ghalighai cave. Iron objects appear in Period VII of the Gandhara Grave culture and can be dated to the beginning of the 1st millennium BCE. There was a basic cultural continuity between the earlier chalcolithic phase and the iron bearing levels. The iron objects included spearheads, arrowheads, pins/nails, spoons, rings, forks, and an axe. One of the graves at Timargarha yielded what appears to be the cheek bar of a horse’s harness. At Saraikhola, iron makes its appearance in the second phase of graves of Period III. The artefacts comprised two rings, a rod, and the iron clasp of a necklace. These may perhaps belong to the first half of the 1st millennium BCE Mention has been made of an iron object found at c. 1000 BCE megalithic levels at Gufkral in Kashmir. The real development of the iron industry at this site took place in the early historical Period III. The Kumaon–Garhwal region is rich in metals and minerals. Heaps of slag and many iron objects were found at the site of Uleni in the upper Ramganga basin in the Almora district of Kumaon. Uleni was clearly an iron smelting and working site and has given a calibrated date range of 1022–826 BCE. THE INDO-GANGETIC DIVIDE AND THE UPPER GANGA VALLEY: THE PAINTED GREY WARE CULTURE

PGW sites in the Ghaggar-Hakra area (including Bhagwanpura) and in the Bikaner region have not given evidence of iron artefacts. Elsewhere, at sites such as Jakhera and Kaushambi (and also Noh in Rajasthan), iron has been found at pre-PGW BRW levels. But in the Ganga–Yamuna doab, the earliest iron objects are generally associated with PGW. PGW was first identified at Ahichchhatra (in Bareilly district) in the 1940s, but its full significance was understood only after excavations at Hastinapur were carried out by B. B. Lal in 1954–55. PGW has a very extensive distribution, stretching from the Himalayan foothills to the Malwa plateau in central India, and from the Bahawalpur region of Pakistan to Kaushambi near Allahabad in Uttar Pradesh. Apart from the plains, it has been found at sites such as Kashipur, Thapli, and Purola in the hilly regions of Kumaon and Garhwal. Sporadic sherds have been found at other places as well—at Vaishali in Bihar, Lakhiyopir in Sind, and Ujjain in Madhya Pradesh. The main concentration of sites are, however, in the Indo-Gangetic divide, Sutlej basin, and upper Ganga plains. The dates of the PGW culture range from c. 1100 to c. 500/400 BCE, and the sites in the north-west are probably earlier than those in the Ganga valley. Given its wide geographical distribution and chronological range, it is not surprising that there are regional variations both in the pottery as well as in associated remains. In the archaeological sequence of the Ganga valley, the PGW phase is followed by the Northern Black Polished Ware (NBPW) phase, the beginning of which goes back to c. 700 BCE at Sringaverapura. The evidence from various PGW sites suggests a proto-urban phase.

MAP 5.7 SOME P AINTED G REY WARE SITES

Important evidence of the PGW material culture is available from excavated sites such as Hastinapur, Alamgirpur, Ahichchhatra, Allahpur, Mathura, Kampil, Noh, Jodhpura, Bhagwanpura, Jakhera, Kaushambi, and Shravasti. PGW occurs in four kinds of stratigraphic contexts. At some sites (e.g., Rupar and Sanghol in Punjab, Daulatpur in Haryana, and Alamgirpur and Hulas in western UP), it is preceded by a late Harappan level, with an intervening break in occupation. At other sites (e.g., Dadheri, Katpalon, and Nagar in the Punjab and Bhagwanpura in Haryana), there is an overlap between the PGW and late Harappan phase. At some sites (e.g., Hastinapura and Ahichchhatra in UP), it is preceded by the OCP culture, with a break in between. And at other sites (e.g., Atranjikhera in UP and Noh and Jodhpura in Rajasthan), the PGW phase is preceded by a BRW phase, with a break in between. At the upper end, PGW overlaps with the NBP culture. Structural remains at PGW levels consist mainly of wattle-and-daub and mud huts. Unbaked bricks and one baked brick were found at Hastinapura. Large baked bricks, possibly used for ritualistic purposes, were found at Jakhera. At Bhagwanpura, there were remains of a large, 13-room house made of baked bricks, but it is not clear whether this was built in the PGW or preceding late Harappan phase. There were artefacts made of stone, bone, and terracotta. Chert and jasper weights were found at Hastinapur.

P AINTED G REY WARE SHERDS FROM HASTINAPURA AND AHICHCHHATRA

Jakhera represents a fairly evolved, proto-urban or semi-urban stage of the PGW culture. An interesting piece of evidence from this site is a water channel and a bund associated with a 60 m long water channel, suggesting water management strategies. Remains of houses (many with multiple hearths), roads and lanes paved with potsherds, and an uneven mud-brick platform associated with a fire altar have also been found. A fire pit with a terracotta hooded snake, a crude handmade figurine, and a bowl were among the interesting discoveries. Square and roundish storage bins suggest surplus food production. The rich range of artefacts from Jakhera included gold and copper ornaments, 106 beads of semi-precious stones, copper artefacts of various kinds, geometric stone pieces, and ivory objects. A large number of iron objects, including agricultural implements such as hoes and sickles, were also found. PRIMARY SOURCES

Painted Grey Ware

Painted Grey Ware (PGW) is a very fine, smooth, and even-coloured pottery, with a thin fabric. Its shades range from a soft silvery grey to a strong battleship grey. It was made out of wellworked, very high quality clay. Designs, mostly simple geometric patterns, were painted on in black. The uniform colour and texture of the pots indicates very sophisticated firing techniques. A uniformly high temperature must have been maintained in the kiln. Or perhaps the pots turned grey while being fired due to the presence of black ferrous oxide in the clay. The pots were thrown on a fast-moving wheel and given an egg-shell thickness. Once they were hard, they were turned on the wheel a second time. The surface was then trimmed and smoothened with the use of

scrapers. Some sort of smoothening emulsion was also applied to give a smooth surface with a matt finish. Some PGW sherds have a reddish core, which could be the result of the use of a different kind of local clay. Simple geometric designs were painted on in black or deep chocolate brown. Several rows of lines, made with a multi-pronged brush, are the most common. Dots, dashes, circles, spirals, concentric circles, checks, swastikas, and sigmas also occur. Naturalistic designs such as floral patterns and sun symbols are less common. Some sites, especially those in Rajasthan, show stamped or incised designs on pottery of this fabric. PGW shows comparatively few shapes. Open-mouthed bowls and dishes occur often, lotas and miniature pots infrequently. PGW seems to have been a deluxe table ware, used by well-to-do people. It forms a very small percentage (3–10 per cent) of the total pottery assemblage at the levels at which it has been found, and occurs along with other pottery types such as plain grey ware, BRW, and black slipped ware. People must have used these other sorts of pottery for cooking, everyday use, and food storage. SOURCE Tripathi, 2002

F IGURE 5.9 P AINTED G REY WARE POTTERY

The PGW sites indicate a subsistence base that included the cultivation of rice, wheat, and barley. People were growing two crops a year. There is no actual evidence of irrigation facilities, but a few deep circular pits outside the habitation area at Atran-jikhera are indicative of kachcha wells. People living in the area today use such wells to irrigate their fields. Animal husbandry was also practised. PGW sites have yielded bones of cattle, sheep, and pigs, many of them charred and bearing cut marks. Fish bones and fishhooks indicate fishing. Horse bones have been found at Hastinapur. Most of the artefacts found at PGW levels seem to be connected with war or hunting— arrowheads, spearheads, blades, daggers, and lances. But there are also clamps, sockets, rods, rings, pins, chisels, axes, adzes, borers, and scrapers, some of which would have been useful in carpentry. The mature PGW phase (Period IIB) at Jakhera has also given important evidence of iron implements used in agriculture—a sickle, ploughshare, and hoe. The wide range of iron objects at PGW levels at Atranjikhera and the agricultural implements found at Jakhera show that the iron industry was well-developed in this area during this period. The chemical analysis of iron artefacts from PGW levels at Atranjikhera has indicated that they were made of wrought iron and were then carburized, probably by keeping them on a bed of charcoal for a long time at a high temperature. The composition of the objects and pieces of iron slag at the site matched that of the iron-rich rocks found in the stretch of hills between Agra and Gwalior, indicating that these were the source of the iron ore.

There are some detailed studies of settlement patterns associated with the PGW phase. Makkhan Lal’s study (1984) of the Kanpur district (UP) identified 46 PGW sites. Of these, 26 were below 1 ha, 14 between 1 and 1.99 ha, 2 between 2 and 2.9 ha, 3 between 3 and 3.99 ha, and 1 between 4 and 4.99 ha. The sites away from the rivers were smaller than those along riverbanks. The average spacing between two settlements was 10–14 km. Erdosy’s study (1988) traces the history of settlements in Allahabad district (UP) between c. 1000 BCE and 300 CE. Period I (100–600 BCE) concerns us here. There was a two-tier hierarchy of settlements. Fifteen sites were 0.42–2.80 ha in size, the average size being 1.72 ha. One site—Kaushambi—was 10 ha and clearly stood out among all the others. Its location in an area of poor soil and rugged terrain may have been in order to access the mineral resources of the Vindhyas. Assuming an average density of 160 people per ha, Erdosy estimates that between 60 and 450 people lived in these villages. A two-tier site hierarchy is also visible in northern Haryana—of the 42 PGW sites here, one site was 9.6 ha and none of the rest were more than 4.3 ha. This evidence can be compared with Mughal’s analysis of PGW settlements in the Bahawalpur area, where there are 14 sites ranging between 0.5 and 13.7 ha. Except for Satwali (13.7 ha), most of them were under 5 ha.

P AINTED GREY WARE SHERDS FROM VARIOUS SITES THE EVIDENCE FROM RAJASTHAN

Noh near Bharatpur shares a similar cultural sequence with sites in the neighbouring upper Ganga valley. Here, Period I yielded OCP and Period II was marked by BRW. Some shapeless pieces of iron were found in Period II. Period III was marked by PGW and yielded iron artefacts such as a spearhead, arrowhead with a socketed tang, and an axe with a broad cutting edge. In eastern Rajasthan, PGW levels at Jodhpura revealed a crucible-shaped furnace used for the direct reduction of ore, where the bloom was heated in an open furnace and forged on an adjacent platform.

The most important evidence comes from Ahar in south-east Rajasthan. Here, there were three chalcolithic phases, and iron occurred in Phases Ib and Ic of the chalcolithic occupation. In Phase Ib, there was an arrowhead, a ring, and slag. In phase Ic, there were four arrowheads, two chisels, one nail, one peg, and a socket. Calibrated dates for the iron-bearing chalcolithic levels at Ahar fall within the first quarter of the 2nd millennium BCE. According to some scholars, the finds come from disturbed layers. However, this has been contested, and it seems that some of the earliest dates for iron in the subcontinent are from Ahar. THE MIDDLE AND LOWER GANGA VALLEY

Recent evidence suggests the beginning of iron technology in the middle Ganga valley in the early and mid-2nd millennium BCE. Calibrated radiocarbon dates from BRW levels at Dadupur (near Lucknow) suggest that this metal may have been introduced at this site in c. 1700 BCE. The ironbearing Period II at Malhar was dated to the early 2nd millennium BCE, and iron at Raja Nal ka Tila (Period II) in the upper Belan valley goes back to c. 1300 BCE. Both these sites have given evidence of iron smelting and iron working. Iron at Jhusi (Period IB) near Allahabad is dated c. 1300 BCE. Elsewhere in the middle Ganga valley, for instance at Ganweria, iron often appears in association with black-slipped ware. At Koldihwa, iron-bearing levels follow the chal-colithic levels, without any break. The iron objects included axes and arrowheads, and crucibles and slag were also found. At Panchoh, iron nodules were found with ill-fired handmade corded and plain red wares, microliths, and small neolithic celts. At Narhan, on the banks of the Sarayu, iron objects made their first appearance in Period I (the BRW phase) and increased significantly in Period II (dominated by black-slipped ware). Period II showed an increase in the number and variety of arrowheads and bone points. There were beads of glass, agate, and terracotta; terracotta balls and dabbers; bone and terracotta dice; terracotta gamesmen; glass bangles; bone pendants; two crude terracotta female figurines; and two animal figurines representing a bull or nilgai. Crucibles made of a vitreous substance as well as of terracotta may have been connected with metallurgy or medicine. Copper objects included antimony rods, a nail parer, bangles, and a fishhook. The iron objects comprised arrowheads, spearheads, chisels, and nails. The discovery of carbonized grains of rice, barley, pea, and green gram indicate a basic continuity in agricultural practices with Period I, the only new element being safflower seeds. Remains of sisoo and jamun were found. Period II at Narhan is dated c. 800–600 BCE. In Bihar and Bengal, the earliest iron artefacts appear in a BRW context at sites such as Chirand, Sonpur, Taradih, Bahiri, Mahisdal, and Bharatpur, and can be placed in the first quarter of the 1st millennium BCE. Many sites show cultural continuity from the chalcolithic BRW phase to the early iron BRW phase. On the other hand, at Mahisdal (on the banks of the Kopai river), early iron artefacts occurred along with microliths, and at Barudih, iron was associated with neoliths. Mention was made earlier of iron artefacts found at chalcolithic levels at Pandu Rajar Dhibi in the Ajay valley in West Bengal. The sites of Bahiri and Mangalkot are also in the Ajay valley. Period I at Bahiri (dated from 1112–803 BCE onwards) gave evidence of rammed floors of wattle-and-daub houses, bone tools, BRW and associated wares, some microliths, and an extensive deposit of iron ore and slag. A piece of copper wire found at Bahiri was analysed and found to contain about 10 per cent tin alloying. Period I at Mangalkot falls within the same date range as Bahiri. Here, there were remains of wattle-and-daub houses with mud floors plastered with cow dung and sometimes rammed

with potsherds and granular gravels. The artefacts included stylized human terracotta figurines, miscellaneaous terracotta objects (beads, bangles, sling balls, net sinkers), beads of semi-precious stones, some microliths, lots of different kinds of bone tools, and copper spiral bangles and fishhooks. Iron artefacts included a point, spearhead, and knife; iron slag and bloom have also been found. CENTRAL INDIA

Iron is found at BRW levels at sites such as Nagda on the banks of the Chambal and Eran on the banks of the Bina river. There is broad cultural continuity between the chalcolithic and early iron age levels. At Nagda, Period I belongs to the Malwa culture. The site was reoccupied after a short break of occupation. Period II was marked by BRW, although the earlier pottery types continued, as did the microliths. Iron objects occurred throughout and included a double-edged dagger, an axe socket, axe with broad cutting edge, spoon, ring, nail, arrowhead, spearhead, knife, and sickle. There was a red or cream pottery with designs (mostly geometric) painted on in black. Similarly at Eran, Period I belonged to the Malwa culture, while Period IIA had BRW and iron. At Ujjain, the iron artefacts found at BRW levels included a spearhead, arrowhead, knife, crowbar, and spade. As the iron-bearing BRW level at these sites directly follows Malwa culture levels, it can be dated c. 1300 BCE. Such a dating is also indirectly supported by calibrated C-14 dates from chalcolithic levels at Eran. There are a number of iron-bearing megalithic sites in Madhya Pradesh. The important ones include Dhanora, Sonabhir, Karhibhandari, Chirachori, Majagahan, Kabrahata, Sorara, Sankanpalli, Timmelwada, Handaguda, and Nelakanker. THE DECCAN

The earliest iron artefacts in the Deccan occur at BRW levels, and many of them are associated with megaliths. The relationship between these levels and the preceding chal-colithic Jorwe culture is not clear. Many of the Jorwe sites seem to have been deserted for four to five centuries, and were reoccupied in about the 6th/5th century BCE. At other places, there seems to be some cultural continuity between the Jorwe phase and the succeeding iron age phase. Prakash has a cultural sequence similar to that of Nagda: Malwa culture levels, followed by a short break in occupation, then a BRW deposit yielding iron artefacts, followed by an early historical NBPW level. The iron artefacts found at BRW levels at Prakash comprised the following types—tanged arrowhead, celt-like axe head, knife blade, sickle, chisel-edged tanged object, clamp, lance or spearhead, ferrule (a metal joint or protective cap), and nails. Similar evidence was found at Bahal. Several megalithic burials and associated habitational deposits in Maharashtra have yielded iron objects. Important sites include Takalghat-Khapa, Naikund, Mahurjhari, Bhagimohari, Borgaon, Ranjala, Pimpalsuti, and Junapani. The calibrated range of dates from Naikund are 800–420 BCE and 785–410 BCE. These sites seem to have been flourishing agricultural settlements. Barley, rice, and lentil grains were found on house floors at Naikund. There were a wide range of copper and iron artefacts. The iron artefacts included ladles, nails, dagger blades, arrowheads, knives, chisels, spikes, axes, double-edged adzes, blades, bars/rods, fishhooks, horse bits, bangles, nail-parer-cum-

earpicks (?), tridents, a spearhead, sword, and cauldron. Iron hoes were found at Naikund, and there was also evidence of the local smelting of iron. The remains of a workshop included a furnace made of small curved bricks with a cylindrical terracotta pipe. Iron ore was found in a nala about a kilometre away from the smelting site. Mahurjhari was an important bead-manufacturing site and the exceptional richness of grave-goods in the burials may be related to this fact. Bead manufacture at this site continued from the megalithic to the early historic phase (Mohanty, 1999). The remains of horses, replete with iron bits and bedecked with copper ornaments, were found at almost all the stone circles at Mahurjhari and Naikund. One of the Ma-hurjhari burials revealed the complete skeleton of a horse, cut marks suggesting that it had been sacrificed and then buried with the human. There were two other dramatic burials—one grave contained the remains of an adult male, his mouth gaping, an arrow embedded near his collarbone. The second contained the top part of the body of an adult male, a dagger with an iron blade and copper hilt rested on his chest. Such burials speak eloquently of a warrior tradition. SOUTH INDIA

In South India, the earliest iron objects appear in the overlap between the neolithic and megalithic phases. Megaliths are widely distributed in South India. In Tamil Nadu, the sites include Adichanallur, Amritamangalam, Kunnattur, Sanur, Vasudevanallur, Tenkasi, Korkai, Kayal, Kalugumalai, Perumalmalai, Pudukkotai, Tirukkampuliyar, and Odugat-tur. Important sites in Kerala include Pulimattu, Tengakkal, Cenkotta, Muthukar, Peria Kanal, Machad, Pazhayannur, and Mangadu. On the basis of the typology of the arte-facts, Machad and Pazhayannur have been dated between the 2nd century BCE and the 2nd century CE. The megaliths at Mangadu in Kollam district of Kerala have a range of c. 1000–100 BCE. Among the important megalithic sites in Karnataka are Brahmagiri, Maski, Hanamsagar, Terdal-Halingali, T. Narsipur, and Hallur. Hallur has a radiocarbon date of c. 1000 BCE. Kumarnahalli has an even earlier thermoluminescence date of 1300–1200 BCE. Sites in Andhra include Kadambapur, Nagarjunakonda, Yelleswaram, Gallapalli, Tadapatri, Mirapuram, and Amaravati. Megaliths associated with BRW have also been found in Sri Lanka. Some of the megalithic types are associated with specific regions—for instance, the kodaikals and topikals with Kerala and Karnataka, and the menhirs with Kerala, Andhra Pradesh, and Karnataka.

B RAHMAGIRI : CHAMBER TOMB WITH PORT HOLE

Megalithic sites were intitially understood as settlements of nomadic pastoralists. However, the evidence clearly indicates that early iron age communities in the far south lived on a combination of agriculture, hunting, fishing, and animal husbandry. There is also evidence of well-developed craft traditions. These features, along with the megalithic monuments themselves, suggest sedentary living. People grew cereals, millets, and pulses. Charred grains of horse gram, green gram, and possibly ragi were found at Paiyampalli. Rice husk occurred at Coorg and Khapa (in Karnataka), and Hallur yielded charred grains of ragi. Rice grains were found in one of the tombs at Kunnatur (in Tamil Nadu). Naturally, there were some regional variations in the crops grown. Pestles and grinding stones have been found at some megalithic sites. For instance, a granite grinding stone was found in a cist at Machad (in Kerala). K. Rajan’s recent (2003) study of the megaliths of the Pudukottai region of Tamil Nadu suggests that the location of megalithic sites close to irrigation tanks (mostly rain fed, some fed by streams) was more than a coincidence. Some clues to subsistence practices also come from paintings and figurines. Hunting scenes are depicted at Marayur and Attala (in Kerala). At Hire Benkal (in Karnataka), there are scenes of hunting, showing peahens, peacocks, stags, and antelopes, as well as scenes of people dancing in groups. The frequent occurence of animal bones—of both wild and domesticated species—indicates domestication and hunting. The cow, sheep, dog, and horse were among the domesticated animals. Cattle were the most important domesticated animal, and in this respect, there was continuity in subsistence practices from the earlier neolithic–chalcolithic phase. Fishhooks have been found in some megalithic graves in Tamil Nadu.

CLOSE-UP OF CHAMBER

The megalithic sites of South India give evidence of well-developed traditions of specialized crafts. Different kinds of pottery have been found, including BRW. Some pots with lids with decorated finials in the shape of birds or animals appear to be ceremonial wares. There is evidence of bead making. Grave goods included etched carnelian beads and beads of other material as well. There are copper and bronze artefacts such as utensils, bowls, and bangles; a few silver and gold ornaments also occur. FURTHER DISCUSSION

The enigma of the megalithic anthropomorphs

Over two decades ago, a huge anthropomorphic figure was discovered at Mottur in Chengam taluka of Tamil Nadu. The figure was part of an arrangement of stones in three concentric circles. The two outer circles were made of stone slabs. The anthropomorph stood in the innermost circle, facing south. It had been embedded in the ground by digging out the bedrock to a depth of 75 cm, and was given additional support by stone packing on both sides at ground level.

The anthropomorph was 3.25 m wide and 3.25 m tall. It had curved arms, 0.92 m long, and the neck and head were represented by a semi-circular projection above the shoulder. Instead of legs, there was a pedestal, making it look like a sitting figure. An identical figure had been discovered a few years earlier at Udayarnattam at Villupuram taluk. This figure—3 m tall and 1 m around the waist—formed part of a stone circle marking a cist burial. A small triangular projection above the shoulder looked like a neck. Local tradition has an interesting explanation for such figures. It tells us that long ago, the Valiyars (pygmies) learnt that a ‘rain fire’ was about to break out. They decided to flee southwards to save themselves. They requested their god to come with them, but he refused. So, as they left, they cut off his head and took it along with them. Hence the figure stands headless. Anthropomorphic figures have been found at 15 megalithic sites stretching from the central Godavari valley to the Tamil Nadu hills. These include Kaperlaguru on the Godavari, Amabala Vayal in Kerala, Midimalla near Chittoor, and Kumati in Bellary district. At Eguvakantala Cheruvu in Chittoor district, three anthropomorphic figures were found associated with each other; the one on the east had a round port hole. Anthropomorphs with heads but no arms have also been reported from northern Andhra Pradesh, particularly at sites on the south bank of the Godavari such as Tottigutta and Dongatogu. What exactly these giant anthropomorphs symbolized is difficult to say. They usually occur in association with chamber tombs and dolmens. They may have been connected with ancestor worship. SOURCE Rajan, 1998a

Iron objects generally outnumber objects made of other metals at megalithic sites. The large volume and variety of iron artefacts—utensils, weapons (arrowheads, spearheads, swords, knives, etc.), carpentry tools (axes, chisels, adzes), and agricultural implements (sickles, hoes, coulters—the vertical blade fixed in front of a ploughshare)—indicate the metal’s widespread use in everyday life. Other more elaborate objects found in burials may have had ritualistic functions. Different sorts of metallurgical techniques were used in the manufacture of metal artefacts. Some of the copper and bronze objects were evidently cast in moulds, others were hammered into shape.

Some communities knew how to alloy metals. An analysis of iron artefacts at Pazhayannur and Machad (Mehta and George, 1978) indicates that the metal was relatively pure with very small traces of other elements. Most of the metal objects at these two sites seem to have been made by forging thin strips, which were then joined by beating them together. One of the objects, a hook, was moulded. There is evidence of local smelting of iron at Paiyampalli (Karnataka). Some megalithic sites must have been centres of craft production linked to networks of exchange. This is suggested by the location of several large megalithic settlements on the trade routes of the early historical period. Inter-regional trade is also suggested by the distribution of non-local items of precious metals and semi-precious stones.

MEGALITHIC CIST, B RAHMAGIRI

Recent excavations at Kudatini in the Bellary district have revealed an exceptionally wellpreserved late neolithic/early iron age sarcophagus burial (Mushrif et al., 2002–03). This was a secondary burial. The sarcophagus and the pots around it contained the remains of a single individual—a child who probably died at the age of 6 or 7. Excavations at megalithic Kodumanal (dated between the 3rd century BCE and 1st century CE) in Erode district, Tamil Nadu, revealed several new features. A cist contained a deer buried in an urn along with etched carnelian beads, a sword, and axes. It seems that in cist burials, the function of the passage was to provide enough space to perform rituals against the port hole. Graffiti marks in archaic Tamil–Brahmi on the grave goods were another major discovery at Kodumanal (Rajan, 1998b). Some megalithic graves reveal continued use of the same burial area over many centuries. However, it seems that the graves were not used more than once or twice in a generation. They probably represent a small elite group within a ranked society. Compared to the earlier neolithic– chalcolithic burials, fewer megalithic graves contain burials of children and young adults, and there is a very high percentage of burials of adult males. Rock paintings found at megalithic sites show fighting scenes, cattle raids, and hunting scenes. At the megalithic habitation site of Mallapadi in Tiruppattur Taluk in Tamil Nadu, rock shelters contained paintings made with white kaolin. One scene showed two horse riders fighting each other with poles. Another showed a human figure with raised arms, holding a stick or weapon. At Paiyampalli, the paintings include a fighting scene, dancing figure, horse raiders, flora, birds, and sun motifs. Such paintings give us an interesting insight into the lives and experiences of megalithic communities. The construction of megaliths must have involved community endeavour. These monuments must also have been sites of rituals that formed an important part of the social and cultural lives of people.

Ethnographic studies of modern megalithic communities suggest that the building of such monuments may have been connected with feasting, gift exchange, and the forging of alliances.

www.pearsoned.co.in/upindersingh PHOTOGRAPHS OF MEGALITHS FROM VARIOUS SITES THE IMPACT OF IRON TECHNOLOGY

There is often a time lag between the beginnings of a technology, its maturation, and its significant impact. Small quantities of iron occur at a few sites in early 2nd millennium BCE contexts. The metal became more widely prevalent in c. 1000–800 BCE. During c. 800–500 BCE, the use of iron was known in virtually all regions of the subcontinent, and by this time, most regions (including the Ganga valley) seem to have entered the iron age. However, in certain areas, this transition took place much later. There has been a decades-long debate over the impact of iron technology on the history of ancient India (see Sahu, 2006 for the different perspectives). This debate has to do partly with the larger question of the role of technology in history, and partly with assessing the literary and archaeological evidence of iron in different areas at different points of time. The debate has especially focused on the Ganga valley in the 1st millennium BCE. Some of the older hypotheses, thought provoking as they were in their time, are not supported by evidence and need to be discarded. For instance, many decades ago, D. D. Kosambi suggested that the eastern movement of the Indo-Aryans was in order to reach the iron ores of south Bihar, and that a near-monopoly over these ores was responsible for the political dominance attained by the state of Magadha (in south Bihar) in early historical times. These hypotheses are untenable, given the very wide distribution of iron ores in the subcontinent. As mentioned earlier, chemical analysis of early iron artefacts at Atranjikhera points to the hills between Agra and Gwalior, not Bihar, as the probable source of ores. R. S. Sharma highlighted the role of iron axes in clearing the forests of the Ganga valley and iron ploughs in agricultural expansion in this area. He argued that the use of these implements was responsible for generating an agricultural surplus, which paved the way for the second phase of urbanization. Religions such as Buddhism were a response to the new socio-economic milieu generated by iron technology. Many aspects of this hypothesis were questioned. A. Ghosh and Niharranjan Ray argued that the forests of the Ganga valley could have been cleared through burning. It was pointed out that Sharma’s argument was not supported by archaeological data, that the impact of iron technology was gradual, that it manifested itself in the mid-NBPW phase when urbanization was well underway, and that socio-political factors had an important role to play in the historical transformations of the Ganga valley in the 1st millennium BCE. Makkhan Lal described the idea of large-scale forest clearance through the use of the iron axe and the generation of an agricultural surplus through the use of the iron plough as a myth. He argued that there was no significant increase in the use of iron from PGW to NBPW levels, that iron technology was not an essential prerequisite for an agricultural surplus or urbanization, that the Bihar iron ores were not tapped during this

period, and that the Ganga plains in fact remained heavily forested till as late as the 16th and 17th centuries CE. Technology is certainly an extremely important factor in history, but it has to be considered along with other variables. Archaeological data indicates that the beginning of iron technology in parts of the Ganga valley can be traced to the 2nd millennium BCE. The earliest iron artefacts occur in BRW or PGW contexts. The use of iron and its impact increased gradually over the centuries and is reflected in the increase in the number and range of iron objects in the NBPW phase. While the expansion of agriculture must certainly have involved some amount of land clearance, large tracts of land continued to be forested. Massive deforestation in the Ganga valley and in the subcontinent in general is actually a feature of the colonial period, when the extension of the railways, increase in population, and the commercialization of agriculture led to a dramatic, unprecedented reduction in forest cover (Williams, 2003: 346–69). Detailed studies of archaeological data from the various regions and sub-regions highlight the complexity of the relationship between technological change and history. For instance, in the far south, the early advent of iron was not followed swiftly by socioeconomic transformations. Rajan Gurukkal ([1981], 2006) points out that iron ploughshares tended to be restricted to the wetland areas. He also argues that notwithstanding the knowledge of iron technology, the larger sociopolitical context of war and plunder hindered the process of agrarian growth in Tamilakam. The simplistic technological determinism that marked the early phase of the iron debate is no longer tenable. The Problem of Correlating Literary and Archaeological Evidence

A great deal of the contentious debate about the co-relation of literature and archaeology during the period c. 2000–500 BCE revolves around the Aryan issue and the relationship between the Vedic and Harappan cultures. The possible co-relations between Sangam literature and the later stage of the megalithic culture of South India, which will be discussed in a later chapter, are considerably less polemical. Many different attempts have been made to connect the evidence from Vedic literature with the Harappan and post-Harappan archaeological cultures of northern India. The relationship between the Indo-Aryans (about whom we know through their texts) and the Harappan civilization (about which we know through archaeology) is a controversial issue. As we saw earlier, some scholars argue that the Harappan civilization was destroyed by the Vedic Aryans. Others point to an overlap between the mature/late Harappan phase and the Indo-Aryan immigration. Still others maintain that there was no Indo-Aryan immigration and that the Harappan civilization represents the culture of the Vedic Aryans. The problems in co-relating literary and archaeological evidence are their completely different nature, their ambiguity, and the problem of dates. We do not know what language the Harappans spoke, and in the absence of any deciphered written evidence, it is difficult to link the Harappan sites with linguistic, ethnic, or cultural groups known from texts. Kenneth Kennedy’s analysis (1997) of the skeletal record reveals that the first phase of discontinuities in physical types in the north-west occurs between c. 6000 and 4500 BCE, and the second one after 800 BCE. There is no evidence of demographic disruption in the north-west during and immediately after the decline of the Harappan civilization. Clearly, no invasions took place

during the period when the Indo-Aryans are supposed to have entered India, nor were there any large-scale migrations. A series of small-scale inflows are a possibility. Many attempts have been made to identify the Indo-Aryans in archaeology. As mentioned in Chapter 4, some archaeologists identified the Cemetery-H culture with the Indo-Aryans. Others have identified foreign elements in the post-urban phase at Chanhudaro (although M. R. Mughal emphasizes cultural continuity rather than discontinuity at that site). Some have sought to identify the Aryans with changes in funerary practices, fire worship, and the use of the horse at Gandhara Grave culture sites. The copper hoards have been variously connected with the early Indo-Aryans, Harappan refugees, and the pre-Aryan inhabitants of the doab. A connection between PGW and the later Vedic Aryans has been suggested on the basis of a chronological and geographical overlap and some similarity in their cultural elements. The PGW culture has also been linked to the Mahabharata events. The chalcolithic cultures of Rajasthan, central India, and the Deccan have been variously identified with pre-Aryans, Aryans, or non-Vedic immigrants. Out of all these corelations, many scholars accept the later Vedic culture–PGW correlation. However, the central problem that has not been properly worked out is: On what basis are connections between material culture—especially pottery—and historically known groups of people to be drawn? It is clear that ceramic cultures cannot be mechanically identified with specific linguistic groups, ethnic groups, lineages, or political units. The spread of similar craft products may have to do with the spread of craft traditions or trade rather than the migration of people. Historians and archaeologists need greater methodological clarity about how to interpret continuity and change in ceramic traditions before making historical inferences on their basis. CONCLUSIONS

Literature and archaeology reveal the varied cultural mosaic of the subcontinent between c. 2000 and 500 BCE. During these centuries, many parts of the subcontinent made the transition from the chalcolithic to the iron age. Historians have used the Vedic texts to identify broad patterns of historical change in the north-west and the upper Ganga valley. Archaeology outlines the features of the everyday life of people living in these and other parts of the subcontinent. The evidence indicates a large number of settlements, many relying on a well-established and stable agricultural base with a two-crops-a-year cycle, supplemented by animal domestication and hunting. In some areas, there was a two-tiered hierarchy of settlements, with a small number of fairly large settlements, sometimes fortified, supporting substantial populations. Traditions of specialized crafts and metallurgical techniques for iron crafting become visible in most areas. There is also evidence of inter-regional and long-distance trade in raw materials and finished goods. All this suggests increasing levels of socio-economic complexity. Archaeological evidence from Inamgaon in the Deccan reflects a chiefdom stage of society and polity, while later Vedic texts reflect the process of transition from tribe to territorial state in the Ganga valley. Towards the end of this period, north India stood on the threshold of urbanization.

www.pearsoned.co.in/upindersingh Further resources

Chapter Six Cities, Kings, and Renunciants: North India, c. 600–300 BCE Chapter outline THE SOURCES: LITERARY AND ARCHAEOLOGICAL THE 16 GREAT STATES THE GANAS OR SANGHAS POLITICAL CONFLICTS AND THE GROWTH OF THE MAGADHAN EMPIRE THE PERSIAN AND MACEDONIAN INVASIONS LAND AND AGRARIAN EXPANSION FROM VILLAGE TO TOWN: THE EXAMPLE OF ATRANJIKHERA THE EMERGENCE OF CITY LIFE ARCHAEOLOGICAL AND LITERARY PROFILES OF EARLY HISTORICAL CITIES URBAN OCCUPATIONS, CRAFTS, GUILDS, AND MONEY THE NEW SOCIAL ELITES: GAHAPATI AND SETTHI TRADE AND TRADERS CLASS, KINSHIP, VARNA, AND JATI GENDER, FAMILY, AND HOUSEHOLD THE RENUNCIATORY TRADITION THE AJIVIKAS EARLY BUDDHISM EARLY JAINISM CONCLUSIONS

SILVER PUNCH-MARKED COINS

The Buddha arrived in a grove outside Kusinara accompanied by his disciple Ananda, weary after a long journey. The two talked of many things, and the conversation turned to the Buddha’s imminent

demise. The Buddha instructed Ananda that his remains should be treated in the same manner as those of a king of kings. The disciple implored him not to die in Kusinara. This small, nondescript town with mud huts in the middle of a jungle was unworthy of being the scene of the Buddha’s final passing away. There were other great cities such as Champa, Rajagriha, Shravasti, Saketa, Kaushambi, and Varanasi, Ananda argued, more worthy of this honour. The Buddha brushed aside these objections. He told Ananda that long ago, Kusinara was a great city named Kusavati, capital of a mighty king named Maha-Sudassana. It measured 12 yojanas east to west and 7 yojanas north to south. Crowds of people thronged its streets and there was prosperity everywhere. Day and night, the city resounded with the sounds of elephants, horses, chariots, drums, tabors, lutes, songs, cymbals, gongs, and with the cry, ‘Eat, drink, and be merry!’ Kusavati was no less than Alakananda, the royal city of the gods. It was indeed worthy of being the place where the Buddha would breathe his last.

This conversation between the Buddha and Ananda takes place in the Mahaparinibbana Sutta of the Digha Nikaya. In the Ganga valley, the 6th century BCE was an age of exceptionally intense and varied philosophical inquiry, as well as one of significant political, social, and economic changes. These changes were the culmination of processes outlined in Chapter 5. While c. 600 BCE has been taken to mark the beginning of the early historical period in north India, this date should be understood as an approximate point along a much longer historical continuum stretching across many centuries. The continuing debate on the date of the Buddha’s death —an event known in Buddhist tradition as the parinibbana—is central to the chronologies of early historical dynasties, post-Vedic texts, and the historical changes reflected in these texts. According to the Dipavamsa and Mahavamsa, 218 years elapsed between the parinibbana and Ashoka’s consecration. On this basis, it has been argued that the Buddha died in 544/543 BCE. Scholars refer to this as the ‘uncorrected long chronology’. Theravada communities in South and Southeast Asia accept this date as the beginning of the Buddhasasana or Buddhist era. The ‘uncorrected long chronology’ was subsequently modified by advancing the date of Ashoka’s consecration to give the ‘corrected long chronology’, according to which the demise of the Buddha occurred some time between 486 and 477 BCE. Another hypothesis, known as the ‘short chronology’, is based on Sanskrit and Chinese sources, which assert that 100 years elapsed between the Buddha’s death and Ashoka’s consecration. Accordingly, the parinibbana has been dated c. 368 BCE. Recently, the question of the date of the Buddha has been subjected to fresh scrutiny. Largely on the basis of the ‘lists of Elders’ in the Pali chronicles, it has been suggested that the parinibbana should be dated somewhere between c. 400 and 350 BCE (Bechert [1991], 1995). The Buddha is supposed to have lived for about 80 years, so depending on which date we accept for his death, we would get a date falling in the 7th, 6th, or 5th century BCE for his birth. Most Indian scholars still favour c. 480 BCE for the parinibbana, while most Western scholars are inclined to accept more recent dates. The Ahraura version of Ashoka’s minor rock edict 1 may hold the key to solving the problem (see Narain, 1993). This inscription suggests that 256 years had elapsed between the death of the Buddha and the issuing of the edict. If the latest possible dates for the events are taken, i.e., if Ashoka’s consecration is dated 264 BCE and the edict belongs to the last year of his reign, i.e., 227 BCE

(although the minor rock edicts are generally assigned to the early part of the reign), the parinibbana would have to be placed in 483 BCE (256 + 227 BCE) or earlier. If conclusive evidence or argument proves a later date for the Buddha, chronologies for early historical texts and events will have to be re-adjusted. Till then, c. 480 can be retained as the date for the parinibbana. The Sources: Literary and Archaeological

For the period c. 600–300 BCE, there is for the first time the possibility of comparing evidence from different kinds of literary sources. As explained in Chapter 1, the Pali canon is not a homogeneous source of history. The first four books of the Sutta Pitaka (the Digha, Majjhima, Samyutta, and Anguttara Nikayas) and the entire Vinaya Pitaka were composed between the 5th and 3rd centuries BCE. The Sutta Nipata also belongs to this period. The Khuddaka Nikaya (the fifth book of the Sutta Pitaka) and the Abhidhamma Pitaka are later works. The geographical context of the composition of the canon corresponds roughly to the middle Ganga valley (modern Bihar and eastern UP). Many historians use the Jatakas (one of the 15 books of the Khuddaka Nikaya) indiscriminately as a source for the 6th century BCE and the Maurya, and post-Maurya periods. Alhough they may contain older legends, in their present textual form, the Jatakas belong to the 3rd centuries BCE–2nd century CE and should not, therefore, be used as a source for the earlier period. In this chapter, they are cited only occasionally, to fill gaps in the detail of political narrative, confirm points emerging from contemporary sources, or to provide a long-term perspective on specific issues. PRIMARY SOURCES

Panini and his Ashtadhyayi

Panini (this seems to have been his gotra name) was a grammarian who lived in the 5th or 4th century BCE. His Ashtadhyayi, the oldest surviving Sanskrit grammar, represents a brilliant intellectual achievement. It sums up the rules of Sanskrit grammar in 3,996 aphorisms (sutras)— short, highly compressed, and condensed statements—combining brevity with clarity and comprehensive coverage. Panini mapped out the grammatical rules of Sanskrit as it existed in his time. His book became a landmark in the history of the Sanskrit language and literature, marking the transition from Vedic Sanskrit to classical Sanskrit. There must have been earlier Sanskrit grammarians, but by founding his own school of grammar and becoming a widely acknowledged authority on the subject, Panini surpassed and eclipsed them completely. Panini’s grammar was handed down over the centuries from teachers to students, to a large extent through oral instruction. Panini was the precursor of other Sanskrit grammarians such as Katyayana (4th century BCE) and Patanjali (2nd century BCE). Both acknowledge their respect for him by giving him the honorific bhagavan. Patanjali describes Panini as a great teacher and the Ashtadhyayi as a vast ocean of learning (mahat-shastra-ugha). He states that students had become careless and indifferent

towards the study of grammar, and that Panini wrote in order to change this attitude. His work is said to have become very popular among young students. Some scholars suggest that Panini was also a poet, but the evidence for this is inconclusive. Little is known about Panini’s life. He was a Brahmana, and seems to have belonged to a place called Shalatura in Gandhara country in the north-west. The Chinese pilgrim Xuanzang visited Shalatura in the 7th century CE. He mentions a statue of Panini standing in the town, and tells us that its children pursued the study of grammar and held the great grammarian in high esteem. The 19th century archaeologist Alexander Cunningham suggested the identification of ancient Shalatura with Lahur, a town four miles north-west of Ohind, close to the confluence of the Kabul and Indus rivers. Xuanzang records some traditions about Panini that were current in the 7th century, and later sources also preserve some stories about his life. The 9th century Manjushrimulakalpa refers to his association with the court of the Nanda king at Pataliputra. The 10th century writer Rajashekhara refers to the institution of a board of examiners at Pataliputra, before whom great grammarians such as Panini and many others appeared. Eleventh-century texts such as Somadeva’s Kathasaritsagara and Kshemendra’s Brihatkathamanjari preserve some legendary stories about the great grammarian, based on the earlier Brihatkatha of Gunadhya. They tell us that Panini was a student of a teacher named Varsha. The story goes that he was a slow learner and lagged behind his friends in studies. At some point of time, he went off to meditate in the Himalayas. Pleased with his efforts, the god Shiva revealed a new system of grammar to him. The Ashtadhyayi is a work on grammar. But in order to illustrate the rules of grammar, Panini referred incidentally to many aspects of his time—to places, people, customs, institutions, coins, weights and measures, and peoples’ beliefs and practices. This is why historians use the Ashtadhyayi as a source of information on the 5th/4th century BCE. SOURCE Agrawala, 1953

Texts belonging to the Brahmanical tradition include the Puranas, which provide useful information on dynastic history. The later sections of the Puranic king-lists clearly have a historical basis, but they present several problems. The Puranas contradict each other in places. Rulers of different lines are sometimes mixed up and presented as members of the same dynasty. Contemporary rulers may be described as successors, collateral rulers as direct descendants. Certain kings known from other sources are not mentioned. Due to their complex internal chronology, it is difficult to use the Sanskrit epics— the Ramayana and Mahabharata—as sources for any specific period. They can at best be used in a very general way for a comparative perspective on cultural practices. The Grihyasutras and Dharmasutras form the earliest segment of the vast corpus of Dharmashastra literature. Kane ([1941b], 1974: xi–xii) dates the Dharmasutras of Gautama, Apastamba, Baudhayana, and Vasishtha between c. 600 and 300 BCE, while Olivelle ([2000], 2003: 10) suggests later dates—the early 3rd to mid-2nd centuries BCE for the first three texts, and a slightly later date

for Vasishtha. We will go by Kane’s chronology. It is difficult to ascertain the precise region where the Dharmashastra texts were composed; they seem to broadly belong to north India, although it is possible that Apastamba belonged to some area in the south. In addition to the Dharmasutras, the principal Shrautasutras (those attributed to Apastamba, Ashvalayana, Baudhayana, Katyayana, Shankhayana, Latyayana, Drahyayana, and Satyashadha) and early Grihyasutras (e.g., those attributed to Ashvalayana, Apastamba, Shankhayana, Gobhila, Paraskara, Kathaka, and Manava), dated c. 800– 400 BCE by Kane, can also be used as sources for this period. All these texts are normative and cannot be treated as simple reflections of prevailing social practices. Nor do they reflect identical points of view. They have to be read as attempts of the Brahmanical tradition to engage with and regulate widely divergent practices. Jaina texts represent the third major tradition that can be used as historical source material for this period. They include the canonical texts and other works such as the Bhagavati Sutra and the Parishishtaparvan. A comparison of Buddhist, Puranic, and Jaina texts on details of dynastic history often reveals more disagreement than agreement. This may be due to incomplete or incorrect information available to their composers, or the fact that they were compiled at different times, but it also has to do with different perspectives. Apart from indigenous literary sources, there are a number of Greek and Latin narratives of Alexander’s military career by writers such as Arrian, Curtius Rufus, Diodorus Sicilus, Plutarch, and Justin. Written several centuries after the events they describe, they recount Alexander’s invasion of India (327–26 BCE) and the political situation prevailing in the north-west at the time. Alexander’s life and career had become the stuff of legend in the Graeco-Roman world. Archaeology continues to be an important source for the history of the subcontinent in c. 600–300 BCE. In north India, the focus is on the culture associated with a pottery called Northern Black Polished Ware (NBPW). The NBPW phase is dated between the 7th century BCE and 2nd/1st centuries BCE, and can be sub-divided into at least two phases—early (7th–3rd centuries BCE) and late (3rd–1st centuries BCE). In this chapter, we are concerned with the former. It may be noted that a series of recent radiocarbon dates from Ayodhya suggest that the NBPW phase could possibly go back to as early as c. 1000 BCE. The evidence from NBPW sites includes an early series of punchmarked coins, which mark the beginning of the use of money in the subcontinent. PRIMARY SOURCES

Northern Black Polished Ware

This pottery’s name is misleading, because it is not only found in north India, it is not always black, nor is it necessarily polished. The NBPW is a well-fired, wheel-made deluxe pottery made of well-levigated clay. It is a fine ware, sometimes as thin as 1.5 mm. Apart from black, it is also found in other shades and colours. The shapes include bowls with straight, convex, tapering, and corrugated sides, dishes with incurved rims and convex sides, dishes with straight sides, knobbed lids, sharply carinated handis, and miniature vases.

The pottery has a glossy surface. How exactly this was achieved is not certain. One theory is that some ferruginous compound was applied to pots before they were fired, and that the black colour was the result of firing the pots in a reducing condition. Another view is that the shiny surface was achieved by applying some material, such as oil or plant juice, on the pots after they were fired, while they were still hot. Yet another study suggests that magnetic iron oxide gave the pottery its black glassy look, while the shine was the result of the application of liquid clay, perhaps containing haematite, along with a natural alkaline substance before firing the pots under reducing conditions. The NBPW is usually unpainted, but there are some instances of designs (bands, wavy lines, dots, concentric and intersecting circles, semi-circles, etc.) painted on in yellow and light vermillion. Pottery matching NBPW was first found at various sites in the 19th century and at Taxila in 1913. It was subsequently reported at many sites in the Ganga valley and beyond. This pottery has been identified at almost 1,500 sites stretching from Taxila and Charsada in the northwest to Amaravati in Andhra Pradesh, and from Prabhas Patan in Gujarat to Tamluk in Bengal. There is a concentration of sites in the Punjab, Haryana, north Rajasthan, UP, Bihar, and West Bengal. The main excavated sites are Rupar in the Punjab; Raja Karna ka Qila and Daulatpur in Haryana; Bairat, Noh, and Jodhpura in Rajasthan; Hastinapur, Atranjikhera, Kaushambi, and Shravasti in UP; and Vaishali, Patna, and Sonepur in Bihar. At sites in the Punjab, Haryana, Rajasthan, and western UP, the NBPW phase is preceded by the PGW phase, with an overlap between them. In eastern UP and Bihar, it is preceded by the Black and Red Ware (BRW) phase.

F IGURE 6.1 NORTHERN B LACK P OLISHED WARE

The 16 Great States

Later Vedic texts, the epics, and the Puranas name many ancient kings and dynasties. The difficulties these texts present as sources of political history have been touched on in Chapter 1, and for the pre6th century BCE period, there is no way of cross-checking and corroborating their details. The earlier part of the epic-Puranic genealogies are clearly mythical; the later ones have a historical basis, while the historicity of the ones in between is uncertain. For instance, it is possible (but not certain) that Parikshit, who is supposed to have become king of the Kurus after the Mahabharata war, his son Janamejaya, and Janaka, king of Videha, were historical figures who lived and ruled between the 9th and the 7th centuries BCE. From the 6th century BCE onwards, the outlines of the political history of north India become clearer, and kings and religious teachers mentioned in different literary traditions can be identified as real, historical figures. State polities and societies emerged in the 6th/5th century BCE in a belt stretching from Gandhara in the north-west to Anga in eastern India, also extending into the Malwa region. The inclusion of Assaka (Ashmaka) in the upper Godavari valley in lists of the great states of the time suggests that similar processes were underway in parts of trans-Vindhyan India as well. Cities and states become visible in South India a few centuries later.1

NORTHERN B LACK P OLISHED WARE FROM VARIOUS SITES

Buddhist and Jaina texts list 16 powerful states (solasa-mahajanapada) that flourished in the early 6th century BCE. (Janapada also meant a region consisting of urban and rural settlements, along with its inhabitants.) Apart from these, there must have been smaller states, chiefdoms, and tribal principalities. The Anguttara Nikaya’s list of the mahajanapadas is as follows: Kasi (Kashi), Kosala (Koshala), Anga, Magadha, Vajji (Vrijji), Malla, Chetiya (Chedi), Vamsa (Vatsa), Kuru, Panchala, Machchha (Matsya), Shurasena, Assaka (Ashmaka), Avanti, Gandhara, and Kamboja. The Mahavastu has a similar list, but substitutes Shibi (in the Punjab) and Dasharna (in central India) for the north-western states of Gandhara and Kamboja.2 The Bhagavati Sutra gives a somewhat different list: Anga, Banga (Vanga), Magaha (Magadha), Malaya, Malava, Achchha, Vachchha (Vatsa), Kochchha, Ladha (Lata or Radha), Padha (Pandya or Paundra), Bajji (Vajji), Moli (Malla), Kasi (Kashi), Kosala, Avaha, and Sambhuttara.3 While some of the names in the two lists are common, the Bhagavati Sutra list seems to be later and less reliable. Two kinds of states are included in the list of mahajanapadas—monarchies (rajyas) and nonmonarchical states known as ganas or sanghas. The latter two terms are used synonymously in the political sense in the Ashtadhyayi and Majjhima Nikaya, and are used interchangeably in this chapter. The translation of gana and sangha as ‘republic’ is misleading. These were oligarchies, where power was exercised by a group of people. The most powerful states in the 6th century BCE were Magadha, Kosala, Vatsa, and Avanti. The relations among the states fluctuated over time and included warfare, truce, and military alliances. Marriage alliances too were an important aspect of inter-state relations, but often became irrelevant when it came to realizing political ambitions. Outlines of political history can be reconstructed by using the various literary sources of this as well as later periods (Raychaudhuri [1923], 2000: 85–210).

The kingdom of Kashi was bounded by the Varuna and Asi rivers to the north and south respectively. It is from the names of these two rivers that its capital city Varanasi (modern Benaras), on the banks of the Ganga, got its name. The Jatakas indicate that several Kashi kings aspired to the status of political paramountcy. They refer to a longstanding rivalry between the kingdoms of Kashi and Kosala. Kashi was also involved in occasional conflicts with Anga and Magadha. At one time, one of the most powerful states of north India, Kashi was eventually absorbed into the Kosalan kingdom.

SILVER PUNCH-MARKED COINS OF KASHI , KOSALA, AND MAGADHA

The powerful kingdom of Kosala was bounded by the Sadanira (Gandak) on the east, the Gomati on the west, the Sarpika or Syandika (Sai) on the south, and the Nepal hills to the north. The Sarayu river divided it into a northern and a southern part. Shravasti (identified with modern SahethMaheth) was the capital of north Kosala, and Kushavati the capital of south Kosala. Saketa and Ayodhya were two other important towns and may once have been political centres. Lesser towns included Setavya, Ukkattha, and Kitagiri. Kosala succeeded in conquering Kashi. It extended its power over the Sakyas of Kapilavastu, and also probably over the Kalamas of Kesaputta and other states in the vicinity. Pasenadi (Prasenajit), king of Kosala, was the Buddha’s contemporary and is frequently mentioned in Pali texts. Kosala and Magadha were linked through matrimonial ties during the time of Prasenajit and the Magadhan king Bimbisara, but a bitter struggle between the two kingdoms ensued after the latter’s death. Anga corresponded roughly to the present-day Bhagalpur and Monghyr districts of Bihar. The Ganga bordered it on the north. The Champa river (which can probably be identified with the Chandan) was its boundary with Magadha, which lay to its west. Its capital Champa (once known as Malini) was one of the greatest cities of the 6th century BCE. Located at the confluence of the Ganga and Champa rivers, it has been identified with modern Champanagara or Champapura village near Bhagalpur. Champa was also an important commercial centre on the trade routes of the time. Merchants are described as sailing overseas from here to Suvarnabhumi (probably in Southeast Asia).

The kingdom of Magadha roughly covered the modern Patna and Gaya districts of Bihar. It was bounded by the Ganga, Son, and Champa rivers on the north, west, and east respectively and the Vindhyan spurs on the south. Its capital was initially Girivraja or Rajagriha, near modern Rajgir. The Puranas give lists of Magadhan kings, starting with Brihadratha. This dynasty came to an end in the 6th century BCE, making way for the Haryankas. The detailed history of Magadha is given later in this chapter. The principality of the Vajji (Vrijji) was in eastern India, north of the Ganga, extending up to the Nepal hills. Most historians consider the Vajji a confederacy of eight or nine clans. This is based on a reference in Buddhaghosha’s Sumangala Vilasini to the atthakulikas of the Vajjis as one of the legal tribunals of Vaishali. However, the atthakulikas may actually have been a body consisting of the eight leading families of Vaishali. The most important members of the confederacy were the Vajjis, Lichchhavis, Videhas, and Nayas/Jnatrikas. We know little about other affiliated clans such as the Ugras, Bhogas, Kauravas, and Aikshvakas. Vaishali was both the capital of the Lichchhavis and of the Vajji confederacy, and is identified with modern Basarh in north Bihar. The Lichchhavis are mentioned often in Buddhist texts. They were on good terms with Kosala and the Mallas, but were embroiled in conflict with Magadha. Jaina tradition states that the nine Lichchhavis formed a league with the nine Mallas and 18 clan lords of Kashi and Koshala. The Videhas had their capital at Mithila, identified with modern Janakpur in Nepal. The Jnatrikas were based in Kundapura (or Kundagrama) and Kollaga, suburbs of Vaishali. Mahavira belonged to this clan. The Vajji confederacy is said to have been led by Chetaka, who was the brother of Trishala (mother of Mahavira) and father of Chellana, wife of the Magadhan king Bimbisara.

MAP 6.1 THE 16 MAHAJANAPADAS

The Malla principality was located to the west of the Vajjis and consisted of a confederacy of nine clans. There were two political centres—at Kusinara and Pava. Kusinara has been identified with Kasia on the smaller Gandak, about 77 km east of Gorakhpur, and Pava with Padaraona village, about 26 km north-east of Kasia. The Mallas are said to have originally been a monarchy. The Vajjis and Mallas seem to have been allies, although there were some episodes of conflict between them. The Chedi kingdom was situated in the eastern part of Bundelkhand in central India. Its capital was Sotthivatinagara, probably the same as the Shuktimati or Shuktisahvaya of the Mahabharata. Vatsa or Vamsa, south of the Ganga, was noted for its fine cotton textiles. Its capital was Kaushambi, identified with Kosam village on the right bank of the Yamuna. Legends recount the rivalry between kings Udayana of Vatsa and Pradyota of Avanti, and refer to a love affair and marriage between Udayana and Vasavadatta, Pradyota’s daughter. Udayana also seems to have

entered into matrimonial alliances with the ruling families of Anga and Magadha. This king went on to become the romantic hero of three Sanskrit dramas of later times—the Svapna-Vasavadatta of Bhasa and the Ratnavali and Priyadarshika of Harsha. According to Buddhist tradition, the Kuru kingdom was ruled by kings belonging to the Yuddhitthila gotta, i.e., the family of Yudhishthira, from their capital at Indapatta (Indraprastha). In the Buddha’s time, Kuru was a minor state ruled by a chieftain named Koravya. The Jaina Uttaradhyayana Sutra refers to a Kuru king named Isukara who ruled from the town of Isukara. The Kurus established matrimonial relations with the Yadavas, Bhojas, and Panchalas. Up to the time of the Buddha, the Kurus were a monarchy; subsequently, they became a sangha. The kingdom of Panchala included the Rohilkhand area and part of the central doab region, and was divided into two parts by the Ganga. The capital of Uttara (north) Panchala was Ahichchhatra (identified with modern Ramnagar in Bareilly district, UP), and that of Dakshina (south) Panchala was Kampilya (identified with Kampil in Farukkhabad district, UP). The famous city of Kanyakubja or Kanauj was located in this kingdom. Several ancient texts mention a king named Chulani Brahmadatta. Going by the testimony of the Arthashastra, the Panchalas seem to have later switched to an oligarchic form of government. The principality of the Matsyas was located in the Jaipur area in Rajasthan, extending into the Alwar and Bharatpur areas as well. Their capital was Viratanagara (modern Bairat), named after Virata, founder of the kingdom. Buddhist texts usually associate the Matsyas with the Shurasenas. The Shurasenas had their capital at Mathura (also known as Madura) on the Yamuna. Buddhist tradition describes Avantiputra, king of the Shurasenas, as a disciple of the Buddha. This king’s name (literally, ‘son of Avanti’) suggests the possibility of a matrimonial alliance between Shurasena and Avanti. The Mahabharata and the Puranas refer to rulers of the Mathura region as the Yadu or Yadavas, who included the Vrishnis. Texts such as the Ashtadhyayi, Markandeya Purana, Brihatsamhita, and the Greek accounts suggest that the Assaka (Ashmaka/Ashvaka) kingdom was situated in the northwest. However, the Assaka of Buddhist texts is firmly located on the Godavari river. Its capital was Potana/Podana or Potali, identified with modern Bodhan. The Godavari separated Assaka from the neighbouring kingdom of Mulaka or Alaka with its capital at Pratishthana (identified with modern Paithan). Jataka stories suggest that Assaka may at some point have come under the sway of Kashi and that it achieved a military victory over Kalinga in eastern India. Avanti, in the Malwa region of central India, was divided into a northern and a southern part by the Vindhyas. The two important towns of this kingdom were Mahishmati (identified with modern Maheshwar) and Ujjayini (near modern Ujjain), both of which are mentioned in ancient texts as its capital. These two cities were important points on trade routes that connected north India, both with the Deccan and with ports on the western coast. Pradyota was a famous king of Avanti, during whose time this kingdom was involved in conflicts with Vatsa, Magadha, and Kosala. The kingdom of Gandhara comprised modern Peshawar and Rawalpindi districts of Pakistan and the Kashmir valley. Its capital Takshashila (Taxila) was a major centre of trade and learning. King Pukkusati or Pushkarasarin ruled over Gandhara in the mid-6th century BCE. He had cordial relations with Magadha, and waged a successful war against Avanti. The Behistun inscription of the Achaemenid emperor Darius indicates that Gandhara was conquered by the Persians in the later part of the 6th century BCE.

Ancient texts and inscriptions usually associate the kingdom of Kamboja with Gandhara. Kamboja included the area around Rajaori, including the Hazara district of the North-West Frontier Province of Pakistan, and probably extended up to Kafiristan. The Kambojas were a monarchy in the 6th century BCE, but the Arthashastra refers to them as a sangha.

G ANDHARA PUNCH-MARKED COIN

FURTHER DISCUSSION

The identification of Taxila

ALEXANDER CUNNINGHAM (1814–93)

Many of the major cities mentioned in ancient Indian texts were identified in the 19th century. One of the men who made a major contribution in this field was Alexander Cunningham, an archaeologist who went on to become the first Director General of the Archaeological Survey of India in 1871. Like others, Cunningham pieced together valuable clues to the location of ancient Indian cities from details given in the classical Graeco-Roman accounts and the travelogues of the Chinese pilgrims Xuanzang and Faxian. Unlike others, however, he routinely confirmed his identifications through field surveys. Cunningham counted among his major achievements the identifications of Aornos, Taxila, Sangala, Srughna, Ahichchhatra, Bairat, Sankisa, Shravasti, Kaushambi, Padmavati, Vaishali, and Nalanda. Cunningham had initially identified the site of ancient Taxila (Takshashila) with Manikyala. However, his explorations at Shah-dheri in 1863– 64 convinced him that the correct identification lay here. The following extract reveals how he reached this conclusion: The position of the celebrated city of Taxila has hitherto remained unknown, partly owing to the erroneous distance recorded by Pliny, and partly to the want of information regarding the vast ruins which still exist in the vicinity of Shah-dheri. All the copies of Pliny agree in stating that Taxila was only 60 Roman, or 55 English, miles from Peucolaitis, or Hashtnagar, which would fix its site somewhere on the Haro river, to the west of Hasan Abdal, or just two days’ march from the Indus. But the itineraries of the Chinese pilgrims agree in placing it at three days’ journey to the east of the Indus, or in the immediate neighbourhood of Kala-ka-sarai, which was the third halting-place of the Mogul emperors, and which is still the third stage from the Indus, both for troops and baggage. Now as Hwen Thsang, on his return to China, was accompanied by laden elephants, his three days’ journey from Takhshasila [sic] to the Indus at Utakhanda, or Ohind, must necessarily have been of the same length as those of modern days, and, consequently, the site of the city must be looked for somewhere in the neighbourhood of Kala-ka-sarai. This

site is found near Shah-dheri, just one mile to the north-east of Kala-ka-sarai, in the extensive ruins of a fortified city, around which I was able to trace no less than 55 stupas, of which two are as large as the great Manikyala tope (i.e., stupa), 28 monasteries, and nine temples. Now the distance from Shah-dheri to Ohind is 36 miles, and from Ohind to Hashtnagar is 38 more, or altogether 74 miles, which is 19 in excess of the distance recorded by Pliny between Taxila and Peukelaotis [sic]. To reconcile these discrepant numbers I would suggest that Pliny’s 60 miles or LX, should be read as 80 miles or LXXX, which are equivalent to 73½ English miles, or within half a mile of the actual distance between the two places…. Then follows a discussion of the history of Taxila and a detailed description of the mounds at the site, namely Bhir, Hathial, Sirkap, Kacha-kot, Babar Khana, and Sirsukh. …In closing my account of the extensive ruins near Shah-dheri, which I have endeavoured to identify with the famous Taxila of the Greeks, I may remark that the identification is most satisfactorily confirmed by the bearings and distances of the next two places visited by Hwen Thsang, both of which will be now described under the names of Hasan Abdal and Baoti Pind. The ruins at these places form, what may be called, the western group of the suburban or outlying remains of Taxila, the ancient capital of the Panjab. SOURCE Cunningham, 1871: 111–35

The Ganas or Sanghas

Ancient Indian texts recognize the difference between the political structure and functioning of the rajyas and the ganas or sanghas. Two of the mahajanapadas, the Vajji and Malla, were sanghas. Buddhist texts mention others as well—the Sakyas of Kapilavastu, Koliyas of Devadaha and Ramagrama, Bulis of Alakappa, Kalamas of Kesaputta, Moriyas of Pipphalivana, and Bhaggas (Bhargas) with their capital on Sumsumara hill. It is interesting to note that most of the ganas, especially the politically important ones, were located in or near the Himalayan foothills in eastern India, while the major kingdoms occupied the fertile alluvial tracts of the Ganga valley. The ganas had greater vestiges of tribal organization than the monarchies. Some may have simply been more complex political forms of older tribal formations. Others may have been created through the subversion of monarchical rule: For instance, the Videhas were apparently originally a monarchy, but had become a gana by the 6th century BCE. The Kurus were a monarchy at this time, but became a gana a few centuries later. There were two kinds of ganas—those that consisted of all or a section of one clan, e.g., the Sakyas and Koliyas; and those that comprised a confederation of several clans, like the Vajjis and Yadavas. The confederacies suggest the existence of a self-conscious political identity among the ganas. The Sakyas claimed to belong to the Ikshvaku family and the solar dynasty. Their principality was bounded on the east by the Rohini river, the Rapti river to the west and south, and the Himalaya mountains to the north. There is still considerable debate over the location of its capital, Kapilavastu. Some scholars locate it at Tilaurakot, but there is better reason to locate it at PiprahwaGanwaria. The amount of detail regarding the Sakyas in Buddhist texts is due to the fact that the

Buddha belonged to this clan. The Sakyas were connected through marriage to the royal house of Kosala. Buddhist texts clearly indicate that the Sakya assembly gathered to discuss important business such as forging alliances, embarking on war, and concluding peace. The principality of the Koliyas of Ramagrama lay to the east of the Sakyas, the Rohini river forming the boundary between the two. Some texts suggest that the two peoples were related to each other. The Bhaggas seem to have been located in the Vindhyan region, between the Yamuna and Son rivers, and were apparently subordinate to the Vatsas. Little is known about the other ganas. Early studies on the ganas by Nationalist historians (see, for instance, Jayaswal, 1943) tended to glorify them by exaggerating their democratic features. Comparisons were made with the republics of Greece and Rome and with modern political institutions. A lot of this was no doubt to disprove the assertions of Western scholars that Indians had never known anything other than despotic rule. Later writings (e.g., J. P. Sharma, 1968) adopted a more dispassionate approach. Governance in these polities was certainly marked by a corporate element. The Arthashastra, a later text, outlines the special strategies that the would-be conqueror could use to vanquish the ganas. Because they were different, the strategies recommended to defeat monarchies would not work, and Kautilya’s advice focused on creating dissension among their ranks. FURTHER DISCUSSION

The conflict between the Sakyas and Kosalans

Buddhist tradition narrates the following story of how relations between the Kosalans and Sakyas soured due to a trick played by the latter on Prasenajit: Prasenajit, king of Kosala, was an admirer of the Buddha, and hit upon the idea of marrying into the Sakya clan, to which the Buddha belonged. He requested the Sakyas for one of their princesses as a wife. The Sakyas were too proud of their lineage to hand over one of their women, but given the power of Prasenajit, were afraid to refuse. So they resorted to a strategem. They passed on a slave girl Vasabhakkhattiya, daughter of a Sakya chief Mahanaman by a slave woman, concealing her low parentage on her mother’s side. Vidudabha and Vajira were the son and daughter born of this marriage. At some point of time, Vidudabha paid a visit to his maternal grandfather’s home and learnt of his mother’s background. The news reached Prasenajit as well. When the king realized that he had been tricked, he immediately disowned his wife and son. The Buddha mollified him by telling him that it was the father’s social status that counted, not the mother’s. Vidudabha and his mother were restored to favour due to this intercession. Vidudabha eventually succeeded his father as king, and massacred the Sakyas as revenge for their trickery. It is impossible to know for sure whether there is any historical basis to this story. But the Kosalan massacre of the Sakyas became one of many episodes remembered in later Buddhist tradition, and it is depicted vividly in relief sculpture at many stupa sites.

The ancient Indian ganas were not, however, democracies. Power was vested in the hands of an aristocracy comprising the heads of leading Kshatriya families. There was no single hereditary monarch. Instead, there was a chief (known variously as ganapati, ganajyestha, ganaraja, or sanghamukhya) and an aristocratic council which met in a hall called the santhagara. Effective executive power and day-to-day political management must have been in the hands of a smaller group. Even in Athenian democracy (with which the Indian ganas were often compared), there was a tendency for power to be concentrated in a few hands. The political system of the ganas seems to have been a compromise between government by assembly and by an oligarchy within this assembly. Later texts offer many details about the Lichchhavis. For instance, the Ekapanna Jataka states that in the Lichchhavi capital of Vaishali, there were always 7,707 rajas (kings) to govern the kingdom, and a similar number of uparajas (subordinate kings), senapatis (military commanders), and bhandagarikas (treasurer