Proceedings of the First International Roman and Late Antique Thrace Conference Sofia 2018 National Archaeological Institute with Museum Bulgarian Academy of Sciences The authors of each article are solely responsible for the content of the texts presented in this volume. The views and opinions expressed are those of the authors and do not necessarily reflect the views of the editors. On the covers: Building inscription on the eastern parascaenium of Philippopolis’ theatre, AD 116-117 (photo: N. Minchev); hedera carved on the seats of Philippopolis’ theatre (photo: N. Sharankov). © National Archaeological Institute with Museum, publisher © Lyudmil Vagalinski, Milena Raycheva, Dilyana Boteva, Nicolay Sharankov, editors © Guido M. Berndt, Andrzej B. Biernacki, Dilyana Boteva, Chiara Cenati, Maria Chryssaphi, Viktoria Čistʼakova, Kevin Clinton, Daniel Delchev, Zdravko Dimitrov, Nora Dimitrova, Ventzislav Dintchev, Diana Dobreva, Nadežda Gavrilović Vitas, Lily Grozdanova, Alexander Harizanov, Constantina Kallintzi, Elena Klenina, Anne Kolb, Kristina Koseva, Maya Martinova-Kyutova, Rumyana Milcheva, Miroslava Mirković, Stephen Mitchell, Johannes Nollé, Marta Oller Guzmán, Maria Gabriella Parissaki, Ulrike Peter, Svetla Petrova, Ioan Piso, Dimitrina Popova, Hristo Preshlenov, Milena Raycheva, Fraser Reed, Łukasz Różycki, Alexander Sarantis, Mustafa H. Sayar, Nicolay Sharankov, Marina Tasaklaki, Argyro Tataki, Domna Terzopoulou, Ivo Topalilov, Lyudmil Vagalinski, Étienne Wolff, Svetlana Yanakieva, Zuzana Zlámalová Cílová, authors © Rositsa Kaneva, design Printed by Bulged Ltd. ISSN 0323-9535 ИзвестИя на нацИоналнИя археологИческИ ИнстИтут, том XLIV, 2018 BULLETIN OF THE NATIONAL ARCHAEOLOGICAL INSTITUTE, XLIV, 2018 National Archaeological Institute with Museum Bulgarian Academy of Sciences Proceedings of the First International Roman and Late Antique Thrace Conference “Cities, Territories and Identities” (Plovdiv, 3rd – 7th October 2016) EDITED BY LYUDMIL VAGALINSKI, MILENA RAYCHEVA, DILYANA BOTEVA, NICOLAY SHARANKOV Sofia 2018 Contents Preface .......................................................................................................................................................9 Cities and Landscapes In Search of Territories in Southwestern Thrace: The Peraia of Samothrace, the Strategy of Korpilike and the Civitas of Traianopolis Maria Gabriella PARISSAKI .................................................................................................................................13 The Roman City of Abdera and Its Territory Constantina KALLINTZI ......................................................................................................................................21 Settlements in the Territory of a New Capital in Late Antiquity. A Contribution to the Historical Geography of the Land to the West of Constantinople in Late Antiquity Mustafa H. SAYAR .................................................................................................................................................31 Nicopolis ad Nestum and Its Place in the Ancient Road Infrastructure of Southwestern Thracia Svetla PETROVA ....................................................................................................................................................37 Influence of Landscape and Climate Conditions on the Urban Fabric of Philippopolis in Antiquity Dimitrina POPOVA ...............................................................................................................................................57 The Ancient Theatre of Philippopolis (Plovdiv, Bulgaria) Maya MARTINOVA-KYUTOVA / Nicolay SHARANKOV ............................................................................67 An Attempt at Enriching the Information on the Fortification of Colonia Flavia Pacis Deultensium Dilyana BOTEVA / Rumyana MILCHEVA ........................................................................................................77 Chronology of the Late Antique Fortifications of Deultum (Archaeological Data) Lyudmil VAGALINSKI..........................................................................................................................................85 Administration and Diplomacy Notes on Governors of Roman Thrace Nicolay SHARANKOV ..........................................................................................................................................97 Roman Thrace: Diplomacy and the Greek Cities of the Aegean Coast Domna TERZOPOULOU ....................................................................................................................................111 C. Iulius Rhaskos, Son of Rhoimetalkes, at Samothrace Kevin CLINTON / Nora DIMITROVA .............................................................................................................121 Identities and Influences Language and Ethnic Identity of the Thracians during the Roman Age and Late Antiquity Svetlana YANAKIEVA ..........................................................................................................................................131 Thrakische und moesische Regionen in Weih- und Grabinschriften stadtrömischer Soldaten (1.-3. Jh. n. Chr.) Chiara CENATI .....................................................................................................................................................141 Thracian Veterans and the Pseudo-Tribes Ivo TOPALILOV ...................................................................................................................................................151 On Some Personal Names of Thrace Argyro TATAKI .....................................................................................................................................................161 La Thrace entre mythe et histoire : l’exemple de Claudien Étienne WOLFF .....................................................................................................................................................169 6 Identitätskonstruktionen in der Münzprägung von Philippopolis (Thracia) Ulrike PETER.........................................................................................................................................................177 Searching for Identities in the Coinage of Pautalia Lily GROZDANOVA ............................................................................................................................................187 Dionysos in Roman Imperial Thrace – What Do the Coins Tell Us about Him? Johannes NOLLÉ ...................................................................................................................................................195 What Was ‘Thracian’ in the Cult of Dionysos in Roman Thrace? Marta OLLER GUZMÁN ....................................................................................................................................211 Thracian Influence in Asia Minor Mystery Cults in the Eastern Parts of the Roman Provinces in the Central Balkans Nadežda GAVRILOVIĆ VITAS ..........................................................................................................................221 Anatolian Elements in the Order Architecture of Philippopolis Zdravko DIMITROV ............................................................................................................................................231 The Lesicheri Pillar and the Transfer of Funerary Architecture Trends from Bithynia to Roman Thrace Daniel DELCHEV / Milena RAYCHEVA ..........................................................................................................245 The Image of Sphinx in Roman Sculpture from Lower Danube Fortresses Andrzej B. BIERNACKI / Elena KLENINA ......................................................................................................259 Economy Transport in Thracia Anne KOLB ............................................................................................................................................................269 The Presence of Roman and Provincial Coins in Aegean Thrace: Quantitative and Qualitative Analysis Marina TASAKLAKI ............................................................................................................................................279 Mensae Ponderariae from the North-Western Black Sea Region. Preliminary Approach Elena KLENINA / Andrzej B. BIERNACKI ......................................................................................................289 Urban Ceramic Workshops in the Province of Thrace during the Principate: The Archaeological Evidence from Bulgaria Alexander HARIZANOV .....................................................................................................................................299 Thracian City Economy as Part of the Global Sinopean Wine Trade Diana DOBREVA ..................................................................................................................................................309 Roman Lamps from Abdera Maria CHRYSSAPHI ............................................................................................................................................323 Roman Glass from Serdica Kristina KOSEVA ..................................................................................................................................................333 Roman Glass Finds from Yurta-Stroyno. A Preliminary Report Viktoria ČIST’AKOVA / Zuzana ZLÁMALOVÁ CÍLOVÁ ..............................................................................343 Conflict and Transition Town and Church in Late Antiquity. Architectural and Urbanistic Dimensions Ventzislav DINTCHEV ........................................................................................................................................357 No City is an Island: The Impact of the Cities of Late Antique Thracia on Provincial Ecclesiastical Construction Fraser REED ...........................................................................................................................................................371 7 Hard Times? Philippopolis in the Fourth Century Stephen MITCHELL .............................................................................................................................................379 Late Antique Mesembria: (Re)Shaping of Public Spaces Hristo PRESHLENOV ..........................................................................................................................................393 The Roman Army and Roman Cities Miroslava MIRKOVIĆ..........................................................................................................................................409 The Strategic Role of Thracian Fortifications in the Balkan Wars of Late Antiquity Alexander SARANTIS ..........................................................................................................................................415 Das verhängnisvolle Jahr 262 und die amissio Daciae Ioan PISO ...............................................................................................................................................................427 Theoderich vs. Theoderich. Zur Konkurrenz gotischer Warlords im Osten des Römischen Reiches Guido M. BERNDT...............................................................................................................................................441 Early Byzantine Arms and Weapons from the Episcopal Complex in Novae Andrzej B. BIERNACKI / Łukasz RÓŻYCKI ....................................................................................................453 Identities and Influences BULLETIN OF THE NATIONAL ARCHAEOLOGICAL INSTITUTE, XLIV, 2018 Proceedings of the First International Roman and Late Antique Thrace Conference “Cities, Territories and Identities” (Plovdiv, 3rd – 7th October 2016) Language and Ethnic Identity of the Thracians during the Roman Age and Late Antiquity Svetlana YANAKIEVA Abstract: The paper examines two interrelated issues: the Thracian language during the Roman Age and Late Antiquity, and the link between the language and the ethnic identity of the Thracians. The two theories in the literature on the Thracian language are discussed, notably: its preservation until the end of Antiquity, and its extinction and replacement by Latin or Greek already during the Roman Age. The theory about the preservation of the Thracian language until the end of Antiquity and the settling of the Slavs on the Balkan Peninsula is supported by research on the available data: evidence provided by Greek and Roman authors, the data from inscriptions and onomastic material – anthroponyms, toponyms, oronyms and hydronyms. On the second issue the author presents her views on the role of language for the preservation of the Thracian ethnic identity within the same chronological boundaries. Key words: Thracian language preservation, ancient authors, inscriptions, onomastic material, ethnic identity. The first researchers of the Thracian language and ethnos write about total Romanisation of the Thracians (Иречек 19783, 87-88; Tomaschek 1893, 10; Zilliacus 1935, 21, 129), which meant in linguistic terms the extinction of the Thracian language during the Roman Age and transition to the Latin linguistic practice. Thus, according to W. Tomaschek, the Thracian language had long disappeared in the 5th – 6th century (Tomaschek 1893, 77), and all peoples accepted in the Roman province were transformed into Latin-speaking provincial inhabitants (Tomaschek 1893, 110). This conviction seems to be due to the existence of a contemporary Romance language originating from Latin in the northern territories of the ancient Thracian linguistic space. Tomaschek even claims that the Christianised Bessi already spoke “limba Rumanêsca” (Tomaschek 1893, 77). Progress in the research on Thracian history, culture and language raised doubts in the earlier theory on the total linguistic Romanisation of the Thracians, and revealed that the available data are not sufficiently convincing for maintaining it (Бешевлиев 1955, 300-302; Beševliev 1964; Геров 1971, 36). The second half of the 20th century was a time of controversial opinions on the fate of the Thracian language during the Roman Age. А. Vraciu, who is researching the substrate elements in the Romanian language, accepts О. Densușianu’s view on the strong Romanisation of most of the Balkan Peninsula, prompted by the transition of the 132 SVETLANA YANAKIEVA indigenous population to new, more civilised traditions, and the wish of the inhabitants to assume public positions, which resulted in the need to master the Latin language better. Consequently, the “autochthonous population was completely Romanised,” whereby the process was slower among the lower strata and in the smaller towns and villages (Врачу 1966, 100-101). No concrete evidence by ancient authors or onomastic data are cited or interpreted. Sometimes it is indeed difficult to understand what some authors think. Thus, I. Russu writes in some places that the Thracian language was alive until the end of the 6th century (Russu 1969, 194), and elsewhere he writes about a “powerful denationalising process” that resulted in the gradual ousting of the Thracian language and the adoption of Greek and Latin (Russu 1969, 193). R. Schmitt also doubts the existence of the Thracian language during the Roman Age. He claims that the absence of texts in the Thracian language in the many inscriptions in Greek and Latin in the region suggests that the Thracians had not preserved their language and ethnicity for a long time. According to him, at any rate the literate people were certainly strongly Hellenised and Romanised. He also cites evidence pointing to the contrary, but believes that evidence not to be absolutely reliable and finally comes to the rather obscure conclusion that the “theory of the preservation of the Thracian language and ethnos until the end of the Antiquity comes as little close to reality as the alternative of the complete Hellenisation, or Romanisation accordingly” (Schmitt 1983, 564-565). V. Georgiev, the most renowned Bulgarian authority in the sphere of Thracian linguistics from the second half of the 20th century, rarely expressed a view on the fate of the Thracian language during Late Antiquity. According to an earlier statement of his, the Thracians, the Moesians and the Getae started using more and more Greek words and grammatical forms, until finally their language became almost completely Hellenised – with only a certain number of substrate elements remaining. On the other hand, the beginning of the new era marked the start of gradual Romanisation of Moesia and Dacia (Georgiev 1966, 22-23). However, that opinion is in contrast with the theory supported by him of the Thracian (Dacian-Moesian, according to his terminology) origin of the Albanian language (Георгиев 1977, 212-215), which presupposes preservation of the Thracian language until the end of Antiquity. Still, most of these scholars admit that the linguistic Romanisation basically affected the upper social stratum of the urban population, whereas the rural population demonstrated poor affiliation to the Latin language. For example, according to V. Velkov, the Hellenised and Romanised Thracians in Thrace and Moesia Inferior left clear traces in the epigraphic monuments in the 2nd – 3rd century, the predominant urban population in the Thracian lands was Hellenised and Romanised in the 4th – 5th century, but – in his opinion – the rural population preserved its customs, cults and language until the end of Antiquity (Велков 1979, 326-327, 414-415). It may be assumed that I. Duridanov accepts the theory of the preservation of the Thracian language until the end of Antiquity. His study on the Slavonicisation of Thracian toponyms contains numerous examples from the entire Thracian linguistic territory LANGUAGE AND ETHNIC IDENTITY OF THE THRACIANS DURING THE ROMAN... 133 (Дуриданов 1968), which presupposes direct contact between the bearers of the two languages. In recent years, there are rare studies on the issue of the linguistic Romanisation and on the fate of the Thracian language during the Roman Age, with a tendency towards accepting its preservation (Янакиева 2009, 192-196; Paliga 2015, 39-41). According to G. Mihailov, too, in the second edition of his book, in spite of the increasingly wider use of Greek and Latin, “the Thracian population in its majority, especially in the rural areas, preserved its native language and the ancient bequeathed tradition” (Михаилов 2015, 460). These controversial views are due to the fact that they were expressed on the basis of general impressions, without detailed study of the entire linguistic material and the data in the literature, although different researchers cite separate pieces of evidence by ancient authors and categories of names. After Rome conquered the Balkan lands and transformed Moesia, Thrace – and later Dacia as well – into Roman provinces, the penetration of Latin into the Thracian linguistic space does not evoke doubts. The problem is to identify the symptoms on the basis of which that process could be referred to as linguistic Romanisation in the sense of complete ousting of the Thracian language and mass transition of the Thracian population to Latin linguistic practice. In view of the absence of written monuments in the Thracian language from the Roman Age, this needs to be done on the basis of the available evidence by the ancient authors, Greek and Latin inscriptions and the data of onomastics: personal, settlement and river names. Inscriptions The spreading of the Latin language in the northern Balkan provinces predominantly and of Greek to the south of the Haemus Mountain is beyond any doubt. There are more than 3,000 Greek inscriptions only from the contemporary Bulgarian lands, whereby in the interior of the country they were from the Roman Age – with few exceptions. The use of Greek administrative, military and religious terminology becomes apparent from them, there is evidence of teachers in Greek, about reading of and familiarity with ancient Greek literature, Greek personal names gradually permeated anthroponymy. Apparently, in addition to the actual Greek settlers and of settlers from the Eastern Hellenistic world, there were also Thracians who knew Greek and were literate, who lived in the cities in Roman times. That was natural, in view of the fact that Greek was the official administrative language in Thracia. At least the representatives of the upper social strata, who were aspiring to a career in the administration, as well as to integration with Greek culture, knew Greek. Nevertheless, the claim that the Hellenisation process affected all social strata, whereby Greek replaced Thracian even as a spoken language (Sharankov 2011, 139), does not seem sufficiently substantiated. N. Sharankov himself indicates the existence of numerous inscriptions set up by Thracians, which testify to their illiteracy and semi-literacy, even among the representatives of the elite (Sharankov 2011, 137-138). One of the examples cited refers to a person who gave a text to be translated into Greek, in which he was even unable to identify separate letters and respectively the words composed of them, i.e., he did not know the words at all in the text that 134 SVETLANA YANAKIEVA had been translated to him. Probably many people used translators for setting inscriptions in Greek. At the same time, even the good knowledge of the language did not mean necessarily that the native language had been totally abandoned. The spreading of Latin in the Northern Balkan provinces passed through several stages. Initially (until Hadrian’s time), it was connected exclusively with the Roman troops and the Danubian limes, and was restricted to the military camps and the settlements near the camps, the bearers of the language being predominantly of Italic origin. During that period, the Thracian population did not participate in the cultural life of the Roman settlements near the camps and was totally absent from the Latin inscriptions (Геров 1949, 6869). During the 2nd and 3rd centuries, the territory of Latin gradually spread in the territory between the Danube and the Balkan Range along the Danube tributaries from Timok to Dobroudja, and beyond the Danube – in Dacia that was conquered by the Romans. During that time Latin was already used by Thracians as well, mostly soldiers and veterans. According to B. Gerov, the upper stratum of the urban population can be considered to have been Romanised. However, in the rural areas along the upper courses of the tributaries and along the northern foothills of the Haemus, the absence or the very rare occurrence of Latin inscriptions in the 3rd century as well suggests that Latin was completely unknown or slightly known among the Thracian population there (Геров 1952/1953, 226-330, 347-350; Gerov 1980, 24-27, 33; Геров 1971, 36). The small number of Latin inscriptions south of the Haemus, mainly of official nature, suggests even poorer knowledge of Latin in Thracia. In Dacia more than 3,000 inscriptions testify to the spreading of Latin (Mihăescu 1960, 281). Roman rule in that province lasted a little more than a century and a half, and after the Romans left under Emperor Aurelian, Roman population was transferred across the Danube into the newly-created part of Dacia Ripensis in the lands of Moesia Superior. After the Romans abandoned Dacia, setting Latin inscriptions on its territory was sharply discontinued, which means that if some bearers of the Latin language had remained, their number would have been minimal. D. Boyadjiev’s study on the late Latin inscriptions in Bulgaria is of particular interest in connection with the issue of the linguistic Romanisation. He has found that the number of Latin epigraphic monuments is very limited, their geographic dissemination followed the urban centres, and the persons related to them were veterans, Christians and representatives of the official rule, with evidence that the compilers were bilingual in some cases. On the whole, the findings present Latin during that time as a language spoken by a small population, which had undergone considerable phonetic and morphosyntactic changes (Бояджиев 1990, 24-30). The general impression of the ancient Greek and Latin inscriptions in the territory of the Thracian linguistic space is that they were the product of a definite literate and educated layer of the population, comprising – in addition to the Greek and Roman settlers, accordingly – also a limited part of the indigenous Thracian population, predominantly in the cities, i.e., they were not proof of mass mastering of the two languages by the Thracians. LANGUAGE AND ETHNIC IDENTITY OF THE THRACIANS DURING THE ROMAN... 135 Ancient authors Information about the Thracian language is very rare in the works of Greek and Roman authors. In the beginning of the Roman Age, such information is found in two authors. Strabo narrates that the Getae spoke the same language as the Thracians, and that the Dacians spoke the same language as the Getae (Str. 7.3.10; 7.3.13; see also Janakieva 2002), while Ovid, living in exile in Tomis, mentions in a letter that he wrote a poem in the Thracian language, referring to it as Getic by the name of the local tribe (Ovid. Ex Ponto 4.13.19). During the last centuries of the Empire, John Chrysostom’s known evidence about translation of the Bible into Thracian (probably the New Testament) (Ioann. Chrysost. Hom. 8.1) was most impressive and repeatedly commented upon in various studies (Кацаров 1924, 46-47; Велков 1958, 739; Бешевлиев 1965, 55, 84; Михаилов 1972, 284-285; Тодоров 1984; Янакиева 2010; Sharankov 2011, 138). The existence of liturgical books in Thracian is also confirmed in the Vita of St. Theodosius, compiled by Theodore of Petra in 536 AD: “the clan of the Bessi addressed their prayers to the Almighty in their native language” (Vita S. Theodosii; see also Бешевлиев 1965, 62 for other data as well). The need to make such a translation for the conversion of the Thracians to Christianity in the 4th century clearly suggests that Greek and Latin were not understood by the vast majority of them. In spite of the lack of texts in the Thracian language, the great available quantity of onomastic material and its careful analysis can give an idea about the linguistic situation in the Thracian linguistic space during the Roman Age and Late Antiquity. Personal names The data from the inscriptions indicate a large number of Greek and Latin personal names from the Thracian lands. Greek names occurred initially in the Greek colonies along the two sea coasts, and at the dawn of the Roman Age Latin names were brought by Italic settlers. Gradually both increased in numbers, and after Caracalla’s edict Latin names became a mass phenomenon with the conferring of Roman citizenship, but they were a sign precisely of citizenship and could not serve as indication of a change in language. Moreover, the three names are not always Latin when citizenship is conferred: the preservation of the Thracian name as cognomen was a frequent occurrence. After the adoption of Christianity, Greek names were predominantly a sign of religious affiliation. It should be borne in mind that the data from anthroponymy present the disparity between the situations of the urban and rural population, because the epigraphic monuments from the cities are incomparably more numerous. As regards the situation in Thracian villages, the inscription from the beginning of the 3rd century of the marketplace Pizus by the peasants from nine villages in the region is very indicative. Only several of about 160 founders had Greek names and several had Latin names. The remaining 140-150 people had Thracian names (IGBulg III.2 1690). Settlement names V. Beševliev has proven convincingly that the numerous Latin settlement names in Moesia and Thrace cannot be considered as evidence 136 SVETLANA YANAKIEVA of Romanisation of the Thracian population (obviously, perceived by him precisely as affiliation of the Thracians to the Latin language). They were exclusively names of sites built or exploited by the Romans, or roadside taverns that served all, irrespective of the language they spoke (Бешевлиев 1955, 300-302; Beševliev 1964). The data on the settlement names present Dacia Ripensis as the linguistically most Romanised area in the eastern part of the Balkan Peninsula (Gerov 1971, 165). The same and the neighbouring northern parts of Dardania and Dacia Mediterranea were the place of origin of the highest number of Latin and Latinised settlement names mentioned in the Roman itineraries, inscriptions and later in the lists of castella in Procopius: Ratiaria, Variana, Valeriana, Ad Malum, Remetodia, Pomodiana, Regianum, Augustae, Montana, Vorovum minor, Florentiana, Romuliana, Castra Martis, Combustica, Mediana, Radices, Ulmo, Remesiana, Latina, Turres, Translitis, Ballanstra, etc. (TP; IA 135, 219; IH 566; Proc. De aed. 4.4; and other sources). The picture is even more vivid with the Greek settlement names, whereby only individual settlements, mostly newly-established, had Greek names (Михаилов 1971, 30). On the whole, the principal mass of settlement names (about 1,000) remained Thracian. Hydronymy Hydronymy could be a serious indicator of a change in language on account of the fact that it implies names of natural sites and the process of their naming could not be influenced by social and political factors (service in the army, acquiring citizenship), or some kind of fashion as for personal and to a certain extent for settlement names (naming by foreigners, imitation). Out of a total of about 150 hydronyms in the Thracian linguistic space there is evidence of only about ten Greek and Latin names. Greek river names occur only along the Aegean and Black Sea coasts, i.e., close to the Greek colonies: Ἐχέδωρος and Ῥήχιος, rivers in the Thessaloniki region, Ζυγάκτης, river name in the area of Philippi, Μέλας, a river flowing into Melas Bay, today Saros Bay (the older Thracian name of the river was Ἄψινθος), Χάβρις, river on the Chalkidiki Peninsula. Three of the Istros Delta distributaries have Greek names: Ἱερὸν στόμα, Ψευδόστομον and στόμα Καλόν. One distributary has two names: Θιαγόλα (Thracian) and Ψιλόν (Greek). What does this small number of Greek hydronyms suggest? Mass influx of river names from a new language in a certain territory would be reliable evidence either of mass settlement of bearers of that language, or of transition of the indigenous population to the new language, abandoning the old one. Such was the case with Asia Minor – a linguistically strongly Hellenised area, where 200–210 out of about 350 known hydronyms were Greek and only about 140 were of local, Asia Minor origin (Tischler 1977). It can be seen that the situation was radically different in Thrace. In the territory of the eastern part of the Balkan Peninsula there is information about only two Latin river names: Picusculus and Turgicu[lus] in the inscription determining the territory of Histria in the early 2nd century (ISM I 68). In Late Antique and Early Medieval authors (Cassiodorus, Marcellinus Comes, Jordanes, Constantine Porphyrogenitus and Hierocles), there were some river names of LANGUAGE AND ETHNIC IDENTITY OF THE THRACIANS DURING THE ROMAN... 137 Germanic and Hun origin (Erac, Gilpil, Nedao, Var) in the Thracian linguistic territory, but no Latin names. Such a small number of Greek and Latin hydronyms in the Thracian lands, whose geographic location obviously linked them to Greek and Roman settlers, as well as the total absence in the interior of the country, are yet another clearer sign than the data on personal and settlement names that Greek and Latin in Thrace were not spoken on a mass scale by the Thracian population (Янакиева 2009, 185-196). Bilingualism probably existed in certain strata of the urban Thracian population during the Roman age, namely among the more educated, among officials at different levels in the administration, among the priests, as well as among the descendants of mixed marriages. There is no doubt that in Thrace, where Greek was the language of the administration, that bilingualism was most frequently manifested as speaking Thracian and Greek, to the north of the Haemus – of Thracian and Latin, and certain individuals probably spoke all three languages. Bilingualism in Thrace existed in certain groups of the urban population, it did not spread to affect the rural population and hence did not result in the ousting of the Thracian language by Greek or Latin. The data on the province of Dacia Ripensis can probably be interpreted as evidence of lasting linguistic Romanisation, i.e., abandoning of Thracian and transition to Latin spoken practice. The preservation of a Romance element in some places in Trajan’s Dacia even after the Romans left cannot be ruled out, but there is no evidence that it was on a mass scale. Quite to the contrary, the phonetic form of some Thracian hydronyms preserved in the contemporary Romanian language suggests that they had been adopted through Slavonic intermediation: Alutas > Olt, Marisos > Mureş, Samus > Someş (Янакиева 2009, 196). It may be assumed that during Late Antiquity (4th–6th century), after the barbarian incursions, the Thracian linguistic space was already fragmented, which created prerequisites for the extinction of the Thracian language and the assimilation of the Thracians when the Slavs settled on the Balkan Peninsula. Language and ethnic identity Modern social sciences, which are oriented predominantly towards studying personal identities, impose the constructivist view in ethnology that the ethnic communities were not formed due to concrete characteristics, but were artificially constructed based on subjective faith (myth) of common origin (Max Weber, Fredrik Barth, Ernest Gellner, Benedict Anderson, Eric Hobsbawm, and others). Nevertheless, a number of cultural anthropologists admit the existence of principal elements of ethnicity (Nash 1996, 25) or cultural markers. According to A. Smith, the ethnic links in a certain group are revealed by external observers based on one or more cultural markers: community of language, customs and religion, although the actual members of the group may not have a myth of common origin (Smith 2009, 27). This is a very long discussion that cannot be covered here. At any rate, I believe that the so-called principal elements or cultural markers are actually real ethnic characteristics. There is a lot of scope for contemplating and arguing on what, how many and to what extent the 138 SVETLANA YANAKIEVA particular characteristics are mandatory for identifying the ethnos. At the same time, the ethnic communities exist in dynamic processes that were subordinated to different factors. In that scheme, language is at the same time a characteristic of the ethnos and a factor of the ethnic processes. The former had been formulated as early as Herodotus in his famous definition of the Greeks (i.e., of the Hellenic ethnos) as a community having common blood, a common language, common sanctuaries of their deities and sacrificial rites, and common customs (Hdt. 8.144). Again Herodotus cites an example of the latter as well, in his narrative about the pre-Greek population of Hellas, the Pelasgians: “... the Attic nation, being of Pelasgian blood, must have changed its language too at the time when it became part of the Hellenes”1 (Hdt. 1.57). Translated into modern academic language, this means that the Pelasgians were finally assimilated by the Hellenes when they stopped speaking their own language (Yanakieva 2015, 178-179). In the light of the data analysed here on the Thracian language it can be said that no linguistic assimilation of the indigenous population (Hellenisation or Romanisation) took place in the Thracian linguistic space, possibly with the exception of some of its groups. This means that the Thracian language was preserved during the Roman Age as one of the principal characteristics of the Thracian ethnos and as a factor for the preservation of the ethnic identity of the Thracians until the end of Late Antiquity. Bibliography Beševliev, V. 1964. Über manche ältere Theorien von der Romanisierung der Thraker. – Etudes Balkaniques 1, 147158. Georgiev, V. 1966. Die Deutung der altertümlichen thrakischen Inschrift aus Kjolmen. – Linguistique Balkanique 11, 1, 9-23. Gerov, B. 1980. L'aspect ethnique et linguistique dans la région entre le Danube et les Balkans à l'époque romain (Ier – IIIe s.). In: Beiträge zur Geschichte der römischen Provinzen Moesien und Thrakien. Gesammelte Aufsätze. Amsterdam. 21-39 (= Studi Urbinati 1959, 173-191). Gerov, B. 1971. Ein neues Buch über die Romanisation im zentralen Teil der Balkanhalbinsel. – Etudes Balkaniques 7, 3, 159-167. IGBulg III.2 = Mihailov, G. (ed.). Inscriptiones Graecae in Bulgaria repertae. Vol. III.2. Serdicae 1964. ISM I = Pippidi, D. (ed.). Inscriptiones Scythiae Minoris Graecae et Latinae. Vol. 1. Bucureşti 1983. Janakieva, S. 2002. La notion de ὁμόγλωττοι chez Strabon et la situation ethno-linguistique sur les territoires thraces. – Études Balkaniques 4, 75-79. Nash, M. 1996. The core elements of ethnicity. In: Hutchinson, J. / Smith, A. D. (eds.). Ethnicity. Oxford. 24-28. Mihăescu, H. 1960. Limba Latină în provinciile dunărene ale imperiului Roman. Bucureşti. Paliga, S. 2015. Ethnic and linguistic distinctions in the ancient Balkan area. How many languages were there in the Antiquity? In: Асенова, П. / Станчева, Р. / Алексова, В. / Бейлери, Р. (eds.). Балканските езици, литератури и култури. Дивергенция и конвергенция. София. 37-43. Russu, I. 1969. Die Sprache der ThrakoDaker. Bucureşti. Sharankov, N. 2011. Language and society in Roman Thrace. In: Haynes, I. P. (ed.). Early Roman Thrace. New English translation by A. D. Godley. The Greek word used is ethnos. 1 Evidence from Bulgaria (= Journal of Roman Archaeology, Suppl. 82). Portsmouth. 135-155. Schmitt, R. 1983. Die Sprachverhältnisse in den östlichen Provinzen des Römischen Reiches. In: Aufstieg und Niedergang der römischen Welt, II 29.2. Berlin / New York. 554-586. Smith, A. 2009. Ethno-symbolism and nationalism. A cultural approach. London / New York. Tischler, J. 1977. Kleinasiatische Hydronymie. Wiesbaden. Tomaschek, W. 1893. Die alten Thraker. Eine ethnologische Untersuchung. I. Übersicht der Stämme. Wien. Yanakieva, S. 2015. Language and ethnicity in Herodotus’ work. – Linguistique Balkanique 54, 2/3, 173-181. Zilliacus, H. 1935. Zum Kampf der Westsprachen im Oströmischen Reich. Helsingfors. Бешевлиев, В. 1965. Проучвания върху личните имена у траките. София. LANGUAGE AND ETHNIC IDENTITY OF THE THRACIANS DURING THE ROMAN... Бешевлиев, В. 1955. Латинските местни имена в Мизия и Тракия. – Известия на Археологическия институт 19, 279-303. Бояджиев, Д. 1990. Езикът на късните латински надписи от България. Част 2. – Годишник на Софийския университет, Факултет по класически и нови филологии 78, 2 (1984), 3-33. Велков, В. 1958: Към въпроса за езика и бита на траките през ІV в. от н.е. In: Бешевлиев, В. / Георгиев, В. (eds.). Изследвания в чест на академик Димитър Дечев. София. 731-741. 139 Геров, Б. 1971. Въпросът за романизацията на тракийското население в българските земи. In: Етногенезис и културно наследство на българския народ. София. 33-37. ство в София 6, 31-47. Геров, Б. 1952/1953. Романизмът между Дунава и Балкана. Част 2. От Хадриан до Константин Велики. – Годишник на Софийския университет, Историко-филологически факултет 48, 307-415. Михаилов, Г. 1972. Траките. София. Дуриданов, И. 1968. Славизация на предславянски географски имена от Балканския полуостров. – Славянска филология 10, 133-141. Михаилов, Г. 2015. Траките. Второ допълнено и преработено издание. София. Михаилов, Г. 1971. Гръцко езиково и етно-културно влияние сред траките. In: Етногенезис и културно наследство на българския народ. София. 27-31. Тодоров, Т. 1984. Библия Бесика (социолингвистична студия за тракийския език в късната античност). – Thracia 6, 259-275. Врачу, А. 1966. Современное состояние вопроса о субстратных элементах румынского языка I. – Linguistique Balkanique 11, 1, 71-107. Иречек, К. 19783. История на българите. София (Geschichte der Bulgaren, Prag 1876). Янакиева, С. 2010. Още веднъж за Biblia Bessica. – История 18, 3, 54-58. Георгиев, В. 1977. Траките и техният език. София. Кацаров, Г. 1924. Тракийските беси. – Известия на Историческото друже- Янакиева, С. 2009. Тракийската хидронимия (= Studia Thracica 12). София. Prof. Svetlana Yanakieva Institute of Balkan Studies & Centre of Thracology Bulgarian Academy of Sciences 13 Moskovska Str. BG-1000 Sofia s.yanakieva@mail.bg Proceedings of the First International Roman and Late Antique Thrace Conference “Cities, Territories and Identities” (Plovdiv, 3rd – 7th October 2016) Edited by Lyudmil VAGALINSKI, Milena RAYCHEVA, Dilyana BOTEVA, Nicolay SHARANKOV Bulgarian First Edition National Archaeological Institute with Museum Bulgarian Academy of Sciences Bulletin of the National Archaeological Institute XLIV, 2018 Printed by Bulged Ltd.