The Ecclesiastical Histories of Socrates and Sozomen

Abstract

Evidently thinking that he could do a better job than Philip of Side, Socrates completed his Ecclesiastical History sometime around 439. Though Socrates was a pro-Nicene Christian in belief, his parents instilled in him a rare attitude of tolerance toward polytheists and dissidents of established Christian orthodoxy, a trait that is evident throughout his history. His uncommon name implies his family’s appreciation for the classical world, and Socrates confirms this suspicion by including references to his Hellenistic education from pagan priests and through many passages alluding to and defending non-Christian classics. In a similar fashion, he is singularly lenient toward minority sects among Christians: He writes favorably of the Novatians, a rigorist branch of Christianity, and even has sympathetic words for “Arians.” While details of Socrates’ personal life are unknown, a late manuscript of his Ecclesiastical History identifies Socrates as “Scholasticus,” a term usually denoting a lawyer by profession. Some scholars remain skeptical of the accuracy of this late addendum, but Socrates’ theological broad-mindedness would be surprising if he were a cleric, and subtle details found in the text itself may suggest a familiarity with the legal world.

1 Socrates of Constantinople (c. 380-c. 440?)

Evidently thinking that he could do a better job than Philip of Side, Socrates completed his Ecclesiastical History sometime around 439.¹ Though Socrates was a pro-Nicene Christian in belief, his parents instilled in him a rare attitude of tolerance toward polytheists and dissidents of established Christian orthodoxy, a trait that is evident throughout his history. His uncommon name implies his family’s appreciation for the classical world, and Socrates confirms this suspicion by including references to his Hellenistic education from pagan priests and through many passages alluding to and defending non-Christian classics.² In a similar fashion, he is singularly lenient toward minority sects among Christians: He writes favorably of the Novatians, a rigorist branch of Christianity, and even has sympathetic words for “Arians.”³ While details of Socrates’ personal life are unknown, a late manuscript of his Ecclesiastical History identifies Socrates as “Scholasticus,” a term usually denoting a lawyer by profession. Some scholars remain skeptical of the accuracy of this late addendum, but Socrates’ theological broad-mindedness would be surprising if he were a cleric, and subtle details found in the text itself may suggest a familiarity with the legal world.⁴

Socrates’ history begins with Constantine’s rise to the imperial dignity in 306 and concludes in 439, during the reign of Theodosius II (408–450). According to Socrates, a certain Theodore, whom he addresses as “a holy man of God,” supported his efforts to compose his Ecclesiastical History.⁵ Though he influenced Socrates’ history, no one has conclusively identified this Theodore, whose name, unlike that of Socrates, was common in Late Antiquity.⁶ It was at Theodore’s encouragement that Socrates undertook an extensive rewriting of the first two books of his seven-book history. Socrates determined to do so after realizing the shortcomings of Rufinus’ account.⁷ He had at first relied upon Rufinus’ earlier continuation of Eusebius of Caesarea; however, a perusal of Athanasius’ own writings revealed errors in Rufinus’ chronology and convinced Socrates to rework his previous efforts. Theodore inspired Socrates to include quotations from letters and conciliar acts to supplement his narrative.⁸ This altered the composition of these two books dramatically. Whereas the remaining five books generally have at most a tenth of their material drawn directly from quotations, the first two books consist of more than a third of quoted documents.⁹

One of the most important sources that Socrates relied upon for these passages was the fourth-century Collection of Synodal Acts.¹⁰ Sabinus of Heraclea, a Homoiousian bishop, compiled documents pertaining to the frequent episcopal gatherings of the fourth century and apparently supplemented these texts with his own reflections.¹¹ Socrates both incorporated Sabinus’ work and drew attention to the “Arian” bias of Sabinus’ remarks. For example, the Homoiousian bishop’s seemingly contemptuous dismissal of the Council Fathers of Nicaea as “laymen and simpletons” (ἰδιώτας καὶ ἀϕελεῖς) inspired a rebuttal and denunciation from Socrates, a literary tit for tat that apparently continued whenever Socrates encountered egregious examples of Sabinus’ prejudice in the Collection.¹² Unfortunately, Sabinus’ work, available as late as the seventeenth century, is no longer extant, and Socrates’ ten references in his Ecclesiastical History form our best basis for understanding this collection.¹³

Modern studies on Sabinus have reached opposing conclusions. Wolf-Dieter Hauschild depicts Sabinus as a staunch proponent of the Homoiousian creed and a determined foe of the pro-Nicene position.¹⁴ Winrich A. Löhr, in contrast, believes that Sabinus was a moderate: Whereas Hauschild (and apparently Socrates) had interpreted Sabinus’ critique of the Nicene Fathers as a grave insult, Löhr prefers to translate the description as “simple and uneducated” and concludes that this is evidence that Sabinus was not a radical Homoiousian.¹⁵ If he were, Löhr argues, terms such as “heretical” or worse would be more likely used rather than the almost-excusing word “simple.”¹⁶

What little we know about Sabinus and his text present us with a tantalizing possibility: Could he, a Homoiousian prelate, be the author behind the proposed fragments in Bidez’s reconstructed Lost Arian History? While the Collection of Synodal Acts presented a collection of documents, Socrates’ statements make it apparent that the bishop of Heraclea added his own commentary. Furthermore, if Löhr’s suggestion is correct, and Sabinus was actually attempting to be reconciliatory in his words, could this possibly link him with the Lost Arian History? As we will see, there is reason to believe that at least portions of Bidez’s reconstruction avoided harsh epithets and polemics.¹⁷ It seems almost too easy to propose Sabinus’ Collection as the source of the non-Nicene sentiments echoed in later sources. This is an attractively simple solution to the historiographical “can of worms” that is the Lost Arian History, but ultimately such a facile identification is untenable.

First of all, Sabinus may have added his own commentary, but the fragments explored in detail in Appendix 1, particularly those from the Paschal Chronicle, suggest an account (or accounts) whose range of material included far more than a commentary on the decisions of councils or important letters. The stories of the martyrs in Antioch or various astral phenomena in Bidez’s reconstruction would stretch the content of Sabinus’ Collection far beyond the little textual evidence that exists. Another example is fragment (24) and its related excerpts which feature Jacob of Nisibis and the Persian siege of that city. This episode was probably part of a non-Nicene history, and it is significant that neither Socrates nor Sozomen, both of whom used Sabinus, include any mention of Jacob. If Sabinus were the original source behind this fragment, one would expect at least one of the two to include this exciting episode. Its absence would be especially surprising in Socrates, whose understanding of the historical “sympathy” (συμπάθεια) between the Church and the Empire would find further expression and justification in this story where a holy prelate’s prayers bolstered the temporal, imperial order.¹⁸ Furthermore, as Bidez’s reconstruction now stands, the Arian History extends until Valens’ death in 378. Hauschild persuasively argues, however, that Sabinus’ work most likely ended in the 360s.¹⁹ Had it extended further Socrates’ awareness of church synods and the activity of the Homoiousian party would presumably not have declined so sharply once his narrative enters the 370s, a decade of activity for the partisans of the Homoiousian creed that is largely ignored by Socrates. Sabinus’ Collection remains a fascinating source of non-Nicene historiography, but one that appears unrelated to the Lost Arian History.

Socrates obviously felt comfortable employing the writing of Sabinus, and his text demonstrates a familiarity with the anti-Nicene and polytheist literature of his time.²⁰ It seems likely that Socrates, who otherwise appears to have recourse to whatever useful resources came his way, would have used the Lost Arian History if he were aware of it. Richard Burgess and van Nuffelen argue persuasively that he did have access to at least part of Bidez’s reconstruction, but further consideration of Socrates’ potential use of the history, which is intimately associated with Burgess’ discoveries regarding Bidez’s proposed fragments, should await the development of other arguments in Chap. 7. At present, this study concludes its introduction of Socrates by observing that he exhibits no parallels with some of the more obvious fragments identified by Bidez, particularly those concerning Julian’s persecution. Furthermore, though he used the writings of the Homoiousian Sabinus, Sabinus’ Collection of Synodal Acts appears to be unrelated to the Lost Arian History.

2 Sozomen of Bethelea (c. 400-after c. 443)

Just several years after Socrates completed his Ecclesiastical History, a second author submitted his historical manuscript to the Emperor Theodosius II. Both Socrates and Salamanes Hermias Sozomen lived in Constantinople as they composed their respective narratives; in fact, if we accept the theory that Socrates was a lawyer, the two men even shared a common profession.²¹ Nonetheless, Sozomen represents a very different sort of author than the broad-minded Socrates. Though both were pro-Nicene in theology, Socrates was willing to show empathy toward those who did not share his beliefs and also described debates within the Christian world that were not edifying. In contrast, Sozomen, a native of Palestine, decided to avoid the religious controversies surrounding “Arianism” and other heresies. He apparently believed that these internal struggles were best circumvented since he lacked the intellectual acumen to comment intelligently on them.²² The more triumphant conflict between faith in Christ and paganism thus receives a much greater focus in Sozomen’s Ecclesiastical History than it does in that of Socrates. This can readily be seen in Sozomen’s decision to include quasi-hagiographical biographies of a number of Syrian monks, some of whom are known solely through his narrative. These stories, with their emphasis on extreme ascetical practices and miraculous events, showcase Sozomen’s willingness to blend historical narrative with hagiographical devotion in order to present Christianity in what Sozomen thought was the best light possible.²³ Ironically, his narrative can be far more discrediting to the eyes of a non-Christian reader than Socrates’ candid account since Sozomen’s unsubtle attempts to evade or merely hint at the disagreements and confusion within the Christian community leave a worse impression than Socrates’ frank admissions.²⁴

Sozomen, however, did not simply write a narrative in order to further his faith’s cause against polytheism. A primary motivation was the advancement of his career. His dedication to the Emperor Theodosius II reminds the emperor of his past largesse to scholars and others in such a way that no one doubts that Sozomen hoped to be numbered among the beneficiaries of imperial patronage.²⁵ Unfortunately for the would-be social climber, Sozomen’s Ecclesiastical History apparently failed to arouse a sense of generosity in its imperial reader.

This situation may be connected to the work’s sudden ending. In its present form, Book 9 of Sozomen’s narrative is less than half as long as the preceding eight and ends abruptly around 421, leaving promises to elaborate on specific events unfulfilled.²⁶ While some believe that Sozomen died prematurely before he was able to finish his book, his dedication implies that the narrative concluded with events in 439.²⁷ It seems more likely that he completed the work and that the sudden end in Book 9 was the result of editing. Probably around 443, Sozomen presented his Ecclesiastical History to the emperor with the dedicatory statement encouraging the ruler to excise or add material according to his imperial preference. Some scholars argue that the emperor’s preference was to remove a significant portion of the concluding Book 9.²⁸ The reason for this decision, they argue, was a domestic conflict between Theodosius and his empress Eudocia. The causes of the divide between husband and wife remain obscure, though it seems she had an affair with a certain Paulinus, the Master of Offices.²⁹ The estrangement of the imperial couple did not become public until 444, and Sozomen probably unwittingly fanned the flames of domestic resentment when his narrative presumably introduced Eudocia in 421, the year Theodosius married her and, incidentally, the same year when Sozomen’s history prematurely terminates. Most likely, the embittered Theodosius excised the history’s remaining text as soon as he encountered the name of his faithless empress, an editorial butchering that permanently disrupted Sozomen’s narrative. The historian’s hopes for advancement were dashed, but he did not have as great a claim to the emperor’s esteem as he insinuated.

While Sozomen wrote in an elevated style—a feature that won the praise of Photius in his related entry in the Bibliotheca—he also masked his extensive dependence on others and gave the impression that his own work was exhaustive.³⁰ This is best seen in regard to Socrates’ Ecclesiastical History. There are remarkable parallels between the works of Socrates and Sozomen to the extent that scholarly consensus concludes that Sozomen incorporated large swaths of his fellow historian’s narrative without ever mentioning either Socrates or his Ecclesiastical History.³¹ How it was that Socrates permitted Sozomen to do such a thing or that Sozomen believed that he could get away with it remain questions beyond any conclusive answer. If one accepts the theory that the two shared a common livelihood, then it may be that Sozomen learned of Socrates’ efforts through private association. Warren Treadgold suggests that Socrates died around 440, after his own Ecclesiastical History was complete and some years before Sozomen offered his history to the emperor. When Socrates died, Sozomen perhaps trusted that few people knew of his predecessor’s work and attempted to claim much of it as his own.³²

In light of this, Sozomen’s assertion to have sought out diligently numerous materials for composition is but a half-truth.³³ He did return to a number of Socrates’ sources, such as Sabinus’ Collection of Synodal Acts, and developed or corrected certain aspects of Socrates’ account.³⁴ Furthermore, he independently consulted the writings of Athanasius, Eusebius of Caesarea’s Life of Constantine, and others.³⁵ His history is thus more than a simple rehash of Socrates, even if Socrates’ text forms its foundation. Nonetheless, it remains highly doubtful that Sozomen read all the Patristic sources that he claims to have consulted, a fact that is especially pertinent to the theological discussions within his Ecclesiastical History.³⁶ Even for his own period he chose to forego independent research and instead relied upon material from the pagan historian Olympiodorus.³⁷

Despite Sozomen’s obvious incorporation of others’ work, the Syrian material in his Ecclesiastical History is often attributed to his own personal experience and research. This, some historians believe, is where he did the real work that he only feigned elsewhere.³⁸ It is certainly possible that Sozomen drew from his own local knowledge while writing about Syrian figures and events; Sozomen’s entries regarding Syrian monasticism and Syria in general, for example, offer much more to the reader than did Socrates’ Ecclesiastical History. Nonetheless, the fact that he so frequently relied on others and hid this very fact should give the reader reason to pause. A glance through the tables of sources meticulously assembled by van Nuffelen reveals numerous points in Sozomen’s narrative where his source(s) remains unknown.³⁹ This content is often focused on events in Syria or in the West. Is it possible that Sozomen derived this material from another text(s) and laid claim to it as his own just as he had done with that of Socrates?

This suspicion is relevant since certain parts of Sozomen’s Ecclesiastical History imply a reliance on the Lost Arian History. Sozomen 5.8.2 describes the sacrilege that Count Julian, uncle of Julian the Apostate, committed against the sacred vessels of the Antiochene church. Julian’s blasphemous acts, his subsequent punishment, and Sozomen’s depiction of the divine wrath against others who likewise mocked the things of God have their parallel in Philostorgius 7.10, Theodoret 3.12–13, and, significantly, in Theophanes AM 5854, that is, in Bidez’s fragment (35) of the Lost Arian History.⁴⁰ Could this be a reflection of Sozomen’s use of the Lost Arian History?⁴¹ If so, why is it that Sozomen’s apparent reliance is limited, as far as can be discerned, to this one instance?⁴² Why are there not more parallels with, for example, the texts of Theophanes and the Paschal Chronicler?

Before pursuing this question further, this study needs to consider two important possibilities regarding Sozomen’s sources. The first looks at the probability that Sozomen used Philostorgius’ narrative directly and the second involves Sozomen’s relationship to the proposed Ecclesiastical History of Gelasius of Caesarea.

3 Sozomen’s Use of Philostorgius

Scholars are divided on whether parallels between Sozomen and Philostorgius reflect Sozomen’s immediate use of Philostorgius or whether the two drew upon a common source. The issue is one that pertains directly to the Lost Arian History: While some historians see Eunapius’ History to Continue Dexippus, a lost fourth-century text, as the bridge between Philostorgius and Sozomen, there exists the above-cited material regarding Count Julian which Sozomen includes and which has parallels to Philostorgius and other sources that used the Lost Arian Historian.⁴³ The likelihood that this explicitly Christian information derives from the pagan Eunapius is minimal.

Sozomen’s apparent incorporation of at least some of the Lost Arian History suggests that he had access to the “Arian” content either directly or through an intermediary source. It seems more likely that Sozomen did not have direct access to the Lost Arian History because of his penchant for the miraculous. The example of Jacob of Nisibis provides a good example; as seen from fragment (13) and Theodoret, EH, 2.30, this story possesses enough miraculous intervention to satisfy even the hagiographical proclivities of Sozomen. Why would Sozomen neglect to include this episode or any of the examples of martyrs suffering at the hands of Julian the Apostate if he actually had before him the source that presented this material? That Sozomen consciously ignored one of the major themes of his narrative—the conflict between pagans and Christians—is not as attractive an answer as the suggestion that he did not have access to the entire content of the Lost Arian History. It seems more probable that the traces of the lost work derive from an intermediary source. Could that source be Philostorgius’ history?

According to scholars such as Bidez, Blockley, and van Nuffelen, the answer is no: Sozomen probably employed Eunapius rather than Philostorgius.⁴⁴ Their arguments focus on the secular details that Philostorgius and Sozomen share. These they believe derive from the two authors’ independent and direct use of Eunapius. The opposing thesis that Sozomen used Philostorgius, meanwhile, focuses on the explicitly Christian material common to both Philostorgius and Sozomen, material that presumably would not have been present in Eunapius’ non-Christian narrative.⁴⁵

Eunapius’ influence may be behind Sozomen’s awareness of non-Christian perspectives or interpretations. Sozomen 1.5 cites reports that Constantine sought purification after murdering Crispus, a detail that implies his direct use of Eunapius as opposed to Philostorgius, whose 2.4, though looking upon the execution of Constantine’s son as a crime, does not mention any apparently pagan perspectives—or at least Photius’ epitome does not include such details.⁴⁶ Furthermore, Sozomen 5.1.4-5 and 5.1.8-9 present two omens that Julian apparently encountered while marching eastward. The first—dew-laden green grapes after the growing season—the emperor interpreted as a positive sign, though Sozomen is quick to adduce a more negative, Christian interpretation.⁴⁷ Perhaps Sozomen is placing a Christian spin on an event recorded by a pagan; the detail of the cross-shaped dew, for example, could be a later addition by Sozomen. The second omen points more directly, though not conclusively, to a non-Christian source. In 5.1.8-9 Sozomen records the claim that diviners had foretold Constantius’ death. He notes that the fact Constantius did die could be construed as proof of these pagan predictions had not Julian’s early death shown their hopes to be false. Sozomen’s obvious effort to distance himself from divination could be an effort to mask his dependency on a non-Christian source or perhaps an effort to compensate for (Eunapius’?) emphasis that the prophecy proved accurate.

Additional support for his theoretical reliance on Eunapius comes in Sozomen 5.19.4-5, 9-11. While the detailed, positive description of Daphne near Antioch in 5.19.4-5 may derive from Sozomen’s own memory or John Chrysostom’s sermons, 5.19.9-11 presents a pagan tradition regarding a stream in Castalia that, like the oracle at Daphne, gave one knowledge of the future.⁴⁸ Sozomen interjects a brief account of this oracle’s interaction with the emperor Hadrian, and this digression appears unlikely to have derived from a Christian source.⁴⁹ Why delay the exciting depiction of Julian at Daphne for a tradition about a Roman emperor who has nothing to do with the rest of the events in this history? Why would a Christian author, either Sozomen or his potential source, be led in his thoughts from one oracle to another that was far removed both in time and in context? What is odd for a Christian author may be natural for a pagan: It would not be surprising for Eunapius’ thoughts to leap from one imperial encounter with the gods (that is, Julian’s) to another (Hadrian’s) while extolling the beauties and the power of the oracle of Daphne.⁵⁰

Nonetheless, there is evidence that Sozomen’s parallels with Philostorgius do not derive solely from their independent use of Eunapius. For example, Sozomen 3.15.8 states that Aetius was originally a physician, a detail possibly, but not necessarily, derived from Philostorgius 3.15.⁵¹ Peter Heather argues that other internal evidence suggests that Sozomen used Philostorgius directly. In his detailed study of Sozomen 6.37, Heather argues that Sozomen’s convoluted account of the Gothic conversion to Christianity and entry into the Roman Empire during the reign of Valens is due to his additional sources, which contradicted and corrected the parallel account found in Socrates 4.33.⁵² Heather believes that Sozomen drew upon Philostorgius since Sozomen’s additional material includes both secular, political themes and a detailed knowledge of ecclesiastical events, such as the Homoian missionary-bishop Ulfilas’ participation in the embassy to Valens in Antioch. His view of Sozomen as “a compiler with limited resources” suggests that Sozomen’s interaction with Eunapius’ text probably came via Philostorgius, though he does not reject the possibility that Sozomen worked with both in front of him.⁵³ Others, such as Michael DiMaio and Noel Lenski also believe that Sozomen used the Heterousian historian directly.⁵⁴

This question of Sozomen’s sources becomes more convoluted when one introduces his portrayal of Julian’s death. In Philostorgius’ representation of the emperor’s last moments, he wrote that the dying Julian flung his blood at the sun and accused his gods, presumably for betraying him to such an end.⁵⁵ Sozomen mentions this narrative in his own Ecclesiastical History and includes a detail absent from Photius’ epitome, namely the emperor’s disappointment that the prophecy of victory promised him by astronomers had proven false. More significant than this addition, however, is Sozomen’s obvious use of a second source: Immediately preceding this detail he had related that some say Julian threw his blood at an apparition of Christ and reproached Jesus, not his gods, for his death. Photius’ epitome makes clear that Philostorgius did not include any reference to Christianity or Christ while recounting Julian’s fatal wound because Photius explicitly remarks that, unlike Philostorgius, other Christian historians had done just that.⁵⁶

Philostorgius presumably copied his material concerning these last hours from Eunapius’ account. Roger Blockley, for example, has already noted how the Heterousian historian includes medical details that suggest that his source was interested in Julian’s wound.⁵⁷ It seems quite possible that these details found their way into Eunapius’ account through his friendship with Julian’s doctor, Oreibasius, who wrote his own account of Julian’s life.⁵⁸

Sozomen did not use Socrates’ Ecclesiastical History at this point, for it has no parallel with that description of Julian’s reaction to his fatal wounding.⁵⁹ And, Sozomen did not incorporate the non-Christian perspective of Ammianus Marcellinus, whose own depiction of the emperor’s death borrows heavily from literary convention and fails to mention Julian’s outburst of anger, either at his gods or at Jesus.⁶⁰ Eutychianus, a pagan soldier and historian whose account of Julian’s death was mentioned by the sixth-century author John Malalas, seems an unlikely possibility, especially given the doubts that have been raised over the very existence of such an individual.⁶¹ In conclusion, it appears that Sozomen employed both a pagan perspective, which he shares with Philostorgius, and a second, presumably Christian source, that neither Socrates nor, apparently, Philostorgius incorporated. Sozomen’s use of a Christian source different from Philostorgius to describe the death of Julian may imply that other details presumed to derive from Philostorgius actually originate in this other source.

Sozomen’s use of Philostorgius or lack thereof has obvious implications for this study. Philostorgius most likely used the Lost Arian History in his own work, and it is possible that echoes of the Arian Historian could be found in Sozomen if he used Philostorgius or consulted the lost work directly. It is therefore important to analyze the suggested parallels between Sozomen and Philostorgius, even though the current epitomized state of Philostorgius’ text precludes any definitive comparison between the two accounts. Indeed, it is good to keep in mind all the interpretative hazards and caveats that necessarily accompany this state of affairs during the following analysis of the parallels.⁶²

3.1 Sozomen 1.10.1-2 and Philostorgius 3.12

The suggested parallel between Sozomen and Philostorgius 3.12 involves their description of various confessors who had survived the Great Persecution. Van Nuffelen apparently includes this passage in his list of parallels because of the description of Maximian of Jerusalem as a martyr for the faith who had lost an eye, though no textual clues seem to warrant any close association between the two.

3.2 Sozomen 2.6 and Philostorgius 2.5-6

Both mention Christian missionary activity among the peoples along the Roman frontier. Sozomen’s account is general, speaking of priests led away captive and then converting their new masters through their witness. Philostorgius’ account, even in epitome, has more detail. He mentions the Homoian Ulfilas and his translation of most of the Bible into Gothic; in 2.6 he recounts the success of Theophilus the Indian, a Heterousian bishop who apparently worked among the Indians of the subcontinent.⁶³

3.3 Sozomen 3.15.7-8 and Philostorgius 3.15-20 and 3.27

Sozomen here introduces Aetius, the leader of the Heterousian cause and a figure whose biography is of obvious interest to Philostorgius. Unlike Socrates 2.35, Sozomen offers details such as Aetius’ relationship with Gallus, which, in turn, may suggest that here he drew upon Philostorgius’ account. Given the incomplete nature of Philostorgius’ text, however, it is impossible to decide conclusively whether Sozomen derived this material from the Heterousian or from another, perhaps even Gelasius of Caesarea, whose own treatise against the Heterousians suggests an interest in that faction that would probably include knowledge of its founders’ biographies.

3.4 Sozomen 3.19.2-5 and Philostorgius 2.1

Photius’ epitome informs us that in Philostorgius’ account Constantine punished those bishops who had subscribed to the creed of Nicaea without actually believing in the Homoousian position. Philostorgius apparently added that Constantine himself embraced Heterousian theology. Philostorgius 2.1b states that the bishops Eusebius of Nicomedia, Theognis, and Maris told the emperor that they had signed the statement out of fear of his wrath and that after their admittance of their falsehood he banished them to Gaul and recalled Arius.

Sozomen’s account differs from that of Philostorgius while adding details not found in the epitome. It begins with Eusebius’ and Theognis’ refusal to confess the Nicene faith and thus being banished. However, after a dream or vision given to his sister, Constantine recalls those exiled and learns that the bishops had initially signed the document out of fear that any prolonged disagreement would engender another imperial persecution. Constantine embraces their explanation but, so the narrative continues, death interrupts his plans to have a pro-“Arian” council. Sozomen’s report has no parallel in Socrates, and his use throughout of “they say” (λέγουσι and ϕασι) and his repudiation of these details indicate that he wished a distinction to exist between his source and his own pro-Nicene narrative. Sozomen’s narrative reverses the events preserved in our epitome of Philostorgius, but Sozomen’s inclination to adapt things to suit his purposes and our inability to look at Philostorgius’ original text hamper our ability to know for certain whether Sozomen drew upon Philostorgius. His data seems to have its roots in a non-Nicene perspective, but it impossible for us to know in what manner he encountered this material.

3.5 Sozomen 5.18.2-6 and Philostorgius 8.11-15

Sozomen 5.18.2-6 may derive from Philostorgius since both historians closely associate the writings of Apolinarius, Basil of Caesarea, and Gregory of Nazianzus. This connection is negative in Philostorgius, who criticizes their style and mentions how Eunomius rebutted their theological arguments.

Sozomen’s similar structure reveals a dependency upon Philostorgius himself or a common source shared by the two. Sozomen writes more about Apolinarius and his attempt to turn the Old Testament into an epic poem than he does about either Basil or Gregory, despite the important legacy of the latter two for orthodox Christianity. Meanwhile, Philostorgius adds the significant detail that Apolinarius, in addition to Basil the Great, wrote a polemic against the Heterousians.⁶⁴ As noted in Chapter 3.2, Gelasius of Caesarea also wrote a work against this sect. We will explore the potential relationship between Sozomen and Gelasius in greater detail in the next section; however, we can begin to anticipate our arguments regarding Gelasius’ Ecclesiastical History by noting that Gelasius may have associated these three prominent anti-Heterousian authors before Philostorgius wrote his own Ecclesiastical History.⁶⁵

Van Nuffelen’s suggestion that the source for this particular segment of Sozomen’s history could be the same as that for 6.25.1-5 and 6.25.7 may also point to Gelasius’ history.⁶⁶ These latter sections appear to have a pro-Nicene tendency that would be odd to associate with Philostorgius: the Homoousian bishop Meletius is implicitly praised and the perspective of the bishop of Rome is favorably introduced. A Homoousian, particularly one who recognized Meletius’ claims to the episcopacy of Antioch, would be a more likely source.

3.6 Sozomen 5.21.1-3 and Philostorgius 7.3

Sozomen here relates the fate of the statue of Christ that, according to tradition, had been erected to commemorate the miraculous healing of the woman with the issue of blood.⁶⁷ This statue was in Caesarea Philippi, and Sozomen’s narrative describes both the statue’s destruction during the reign of Julian by the pagans of that city and the Christian effort to collect the various broken pieces, which were placed in a church for safekeeping. Philostorgius’ narrative, as preserved in Photius’ epitome, offers a number of details that enhance our knowledge: There was a miraculous herb that grew at the base of the statue, the statue itself had been moved to a church interior, and, in the wake of the pagan violence, the preserved parts included Christ’s head, which Philostorgius claimed to have seen himself.

The details in Philostorgius round out the outline presented by Sozomen, though there is one point where Sozomen offers a more complete account than Philostorgius. Sozomen declares that Julian replaced the statue of Christ with one of himself, which God’s divine power destroyed through fire from the sky. The historian delights in describing how the different pieces of Julian’s image suffered under the heavenly onslaught. Philostorgius, at least in the epitome, fails to describe the destruction of this image; in fact, he states that an image of the god Pan, not Julian, replaced the statue of Christ. Philostorgius further explains that it was from this statue that the city received its name Paneas.

While it is tempting to see Philostorgius’ account as the basis for Sozomen’s, the contrast in statues erected is significant. Other sections of Philostorgius’ history reveal that he is eager to use classical labels when discussing contemporary peoples and to explore the etymology of place names.⁶⁸ Some are conjectural on his part, and others reveal his penchant for archaic terminology. His reason why the city was known as Paneas in the fourth century is superfluous: Already since the second century the city had been commonly called Caesarea Paneas.⁶⁹ The city clearly did not derive that name from a sculpture purportedly built a century later, and Philostorgius’ attempt to link the city’s name with the statue may indicate that he himself supplied the detail of the god’s statue.

Even if Philostorgius’ claim that the statue was of Pan is accurate, it does not imply that an effigy honoring the emperor was never erected. Some cities placed images of the emperor in prominent areas, and Caesarea may have done likewise.⁷⁰ Meanwhile, Sozomen’s detailed depiction of the statue’s fate has no parallel in the extant record from Philostorgius. One wonders whether the Heterousian, who claimed to be an eyewitness of the broken statue of Christ, either failed to mention the destruction of the pagan replacement or Photius did not include it among his notes. Perhaps Sozomen is here drawing upon a different source from Philostorgius. Both the difference between the statues and Sozomen’s detailed account of the pagan image’s destruction seem to have a different focus than that of the account of Philostorgius, which highlights the rediscovery of the statue of Christ, the miraculous herb, and efforts to preserve the statue’s fragments.

There are too many gaps in our knowledge to decide one way or the other, but Sozomen need not be drawing directly from Philostorgius’ history in order for us to account for the similarity between the two. Another Christian narrative, such as that of Gelasius of Caesarea or the Lost Arian History, may have served as the basis for Sozomen’s account. This is especially pertinent since Philostorgius’ account mentions the miraculous herb said to grow at the base of the statue and to have curative properties for consumption. Sozomen often includes miraculous details such as this, but here he merely mentions vaguely that Eusebius of Caesarea had mentioned the herb’s healing virtue. This is surprising, particularly when the rest of his chapter highlights similar stories of miraculous places or things associated with the life of Christ.

It is also worth pointing out that the rest of Sozomen 5.21, according to van Nuffelen’s meticulous research, derives from an unknown source. Van Nuffelen tentatively suggests local tradition inspired Sozomen’s discussion of the miraculous places in Nicopolis and Hermopolis, which cities he links to events in the New Testament.⁷¹ Philostorgius, or at least the epitome, has no such material. Sozomen, a native of Palestine, may have known of Nicopolis’ traditions, but it is not certain why traditions in Egypt would have been familiar to him.

3.7 Sozomen 6.6.1-8 and Philostorgius 7.7 and 8.8

The points of apparent parallel in these sections involve the death of Jovian and the dramatic change in fortune for Valentinian I. Jovian, Sozomen states, may have died from excessive eating, but his narrative also mentions the theory that he died from asphyxiation. This theory parallels that given in Philostorgius, and both historians include the details regarding the recently applied plaster and the dampness of the bedroom.

Regarding the early career of Valentinian I, Sozomen begins with an aside about the Christian Valentinian’s violent reaction to a pagan rite that he was forced to attend as an officer under Julian’s command. This outburst of anger results in his banishment to Melitene in Armenia, from which humiliation Sozomen’s narrative quickly shifts to his accession to the throne and his decision to share the rule with his brother Valens.

Philostorgius 7.7 hints at Valentinian’s encounter with Julian’s anti-Christian pressure but offers little detail except the notice that the future emperor suffered exile in Thebes, an obvious contradiction to the account of Sozomen above.⁷² 8.8, meanwhile, includes Valentinian’s speech to his soldiers defending his newly gained imperial prerogative to select a co-emperor, something which is echoed in both Sozomen and Theodoret 4.6.2.

Theodoret 3.13, too, offers an account of Valentinian’s career prior to his elevation. His depiction of Valentinian’s encounter in a pagan temple echoes that of Sozomen, though there are discrepancies between the texts. For example, Sozomen records that the event happened in Gaul, whereas Theodoret states that the event occurred in Antioch.

Given these contradictions among the texts, many historians have discounted their historical reliability. Maurizio Colombo, for example, dismisses these details as fictitious.⁷³ He argues that the sources’ confusion indicates their lack of historical accuracy, though he does believe that there was a minor incident that inspired the accounts of Valentinian’s courageous confession of faith and subsequent exile.⁷⁴ Colombo’s article aims, in part, to refute the “bizarre reconstruction” (la bizzarra ricostruzione) offered by David Woods.⁷⁵ Colombo’s reconstruction of Valentinian’s early military career has much to commend it vis-à-vis Woods’ theory, but it still seems worthwhile to explore elements of Woods’ argument, especially since it is seems reasonable to propose the existence of an underlying textual tradition behind Sozomen’s, Theodoret’s, and Philostorgius’ story of Valentinian’s confession of faith. Whether the tradition is historically accurate is not our concern so much as exploring the potential textual connections among the three fifth-century historians who included this episode.

David Woods argues that these various accounts, including even that of an ambiguous episode in Ammianus Marcellinus 23.1.5-6, are compatible and even historically accurate, with one crucial alteration.⁷⁶ He claims that these stories involve not Valentinian but his brother Valens. At some point, some individual through either careless mistake or more likely, conscious decision, changed the original account.⁷⁷ What had originally been biographical details for Valens morphed into those for his brother Valentinian. Woods offers little speculation about who might have done this or when this alteration might have taken place, and he sees no obvious candidate among those who wrote prior to Philostorgius and the other fifth-century ecclesiastical historians. Perhaps the best guess would be that Philostorgius made the change himself, for his account occasionally altered pro-Nicene stories into an opposing perspective.⁷⁸ Yet Woods rightly points out that such an alteration seems highly unlikely.⁷⁹ Philostorgius, a Heterousian, had no reason to prefer Valentinian, a lukewarm, but pro-Nicene, emperor, to Valens, an aggressive advocate for the Homoian position. Furthermore, there are the contradictions between Theodoret and Sozomen. Both are probably referring to the same source, yet one places the event in Antioch, the other in Gaul.⁸⁰

Woods’ circumstantial evidence in favor of his theory has attracted much criticism. Noel Lenski, for example, highlights several difficulties presented by Woods’ thesis, including the ambiguous archeological evidence that Woods uses to identify Valentinian’s military unit as well as the absence in Socrates of any trace of the Valentinian/Valens story as presented by the other ecclesiastical historians.⁸¹ As Lenski assumes that Socrates employed Gelasius, whose account would be the earliest of the ecclesiastical historians and presumably the most accurate, he believes that Woods’ theory loses crucial support if Socrates’ account is different.⁸²

Lenski proposes three additional objections to Woods’ thesis, though two of these are not compelling as the third. The first involves the discrepancies between Theodoret’s narrative and that of Ammianus Marcellinus.⁸³ Lenski suggests caution when dealing with Theodoret due to the bishop’s propensity to alter his narrative in order to suit his didactic purpose.⁸⁴ This warning is worthy of consideration, but, as studies on Ammianus Marcellinus have shown, neither historian can be trusted more than the other in this regard.⁸⁵ Ammianus’ polemical relationship with Christians and Christianity is too complex to assume he would not alter a story under certain circumstances, especially if a tradition had arisen which praised Valens, whom he disliked, as a confessor.

Lenski’s contention that Eutropius, another late fourth-century historian, would have included such a detail in his Brevarium is also worth considering.⁸⁶ Eutropius, an official at Constantinople and a participant during Julian’s Persian expedition, wrote a history of Rome from its founding to the present at the behest of Valens and dedicated it to him. Why would Eutropius not mention something so advantageous to the reputation of his patron? A potential answer lies in the purpose of the text. Eutropius’ goal was to provide an abbreviated history of Rome, and his account of the persecution of Julian is so brief that there are no details.⁸⁷ Furthermore, the purpose behind the Brevarium seems to have been the promotion of popular support among the imperial officials for an aggressive stance against Persia.⁸⁸ Given that the Eastern Roman Empire was still recovering from the situation Julian’s failed campaign had created, it is possible that Eutropius (and Valens) wanted to avoid any note of discord among the recent emperors in order to promote as wide a base of support as possible. This may explain why, contrary to many Christian sources, Eutropius places emphasis on Jovian’s forced withdrawal from Persia and describes Jovian’s ascension without any reference to the religious issue.⁸⁹ Maybe this work was intended to win the support of both Christian and pagan officials and thus avoided exacerbating any tensions by ignoring Julian’s hostility to Valens and vice versa.

A third, and weightier, objection derives from Lenski’s observation that the Paschal Chronicle follows the Valentinian-as-confessor narrative and does not place Valens in this role.⁹⁰ That the Paschal Chronicle tells the story about Valentinian and not his brother suggests that the chronicler’s source, typically identified as the Lost Arian History, did not have Wood’s hypothetical version involving Valens. Since the Arian Historian would have no reason to substitute the pro-Nicene Valentinian for the pro-Homoian Valens, it would appear that Woods’ theory has no true foundation.

This last argument is particularly compelling, and, yet, there must be some reason(s) why the particulars in the story of Valentian’s confession of faith have so much variety. Even if we accept Colombo’s belief that these are fictitious, the authors were clearly drawing upon an established tradition despite presenting contradictory details. In Chap. 7, we explore the possibility that there existed a martyrology focused on preserving the memory of those who suffered during the reign of Julian.⁹¹ This proposed martyrology, which likely highlighted the suffering of Antiochene Christians in particular, appears to have come from a non-Nicene pen; in fact, Chap. 7 argues that it most likely derives from a Homoian perspective, a theological position shared with the contemporary emperor Valens. It may be that this proposed non-Nicene martyrology was the source for the tradition that Valentian was a Christian confessor. It would not be surprising for this source to honor the Homoian Valen’s brother and co-ruler, even if Valentinian was technically pro-Nicene in his official stance. After all, such flattering testimony of the Western Roman emperor’s character would reflect positively on Valens since it was Valentinian who promoted his brother to the throne.

This, of course, brings us no closer to understanding why there exists the variety of details. At least in regards to Sozomen, Woods’ theory seems to be plausible: Woods believes that it may be Eunapius who provided the alternative information regarding Valentinian’s early career—including his dishonorable discharge from the army due to his poor performance in Gaul.⁹² Sozomen, Woods argues, had both Eunapius’ and another account before him and sought to reconcile them, not realizing that he was attempting to resolve discrepancies that arose from a martyrology whose account of two individuals had been conflated and possibly combined. In order to preserve details from both works, Sozomen likely transferred the episode of the violent encounter with the pagan priest from Antioch—which is where, according to Woods’ theory, it actually happened and where Valens’ exile began—to Gaul, the location where Eunapius described the unrelated events that led to Valentinian’s discharge. We do not have to accept the particulars of Woods’ argument to see that this situation, or one similar to it, could have arisen in Sozomen’s text.⁹³ Why Theodoret and Philostorgius have the discrepancies they do is not as easy to suggest. It may be that they drew upon an alternative traditions, but we simply cannot speak with confidence.

In conclusion, there is little that can be written with certainty regarding the historiographical roots of the tradition that Valentinian was a confessor for Christianity during the reign of Julian. Woods’ theory has engendered spirited rebuttals that require many of his specific claims to be reevaluated. Nonetheless, we acknowledge that some martyrological source(s) lay behind the testimony of Philostorgius, Sozomen, and Theodoret. It may even be, as Woods argues, that details from different individual confessors became conflated and confused by these fifth-century writers; but whether this is so remains beyond our ability to confirm or reject. In light of this situation, it is impossible to say whether Sozomen drew upon Philostorgius for his account.

3.8 Sozomen 6.26.2-6 and Philostorgius 10.4 and 6.1-3

Sozomen 6.26 reveals an intimate knowledge of events in the Heterousian community, information that is commonly linked to his supposed use of Philostorgius. Even a casual perusal of Sozomen’s knowledge of the Heterousian community reveals that his source(s) presents much more information than that used by the parallel Socrates 5.24. Eunapius is an unlikely candidate for such details, and historians have considered this information to be rooted by default in the narrative of Philostorgius.⁹⁴ For example, while Socrates states that Eunomius’ disciples changed the formula for baptism, Sozomen apparently engages with a number of sources, some of which claim that Eunomius himself was the author of the novelty. Sozomen then enters into a relatively lengthy polemical argument rooted in a thorough understanding of Eunomian practice: He claims that their new baptism in the name of the death of Christ, as opposed to that in the name of the Persons of the Trinity, leads the Heterousians into a paradox. They deny the validity of the latter baptismal formula and yet claim that only those who are validly baptized can administer that sacrament. Sozomen uses this point to craft a logical argument against the Heterousians’ alteration of the baptismal formula. After all, reasons Sozomen, if the uninitiated cannot baptize and the Heterousian leaders themselves received an invalid Trinitarian baptism, how is it that this technically unbaptized leadership can now bestow that sacrament on their followers?

Our initial reaction may be to identify Sozomen’s source as Philostorgius. Though Photius’ epitome glosses over those details of the later Heterousian community that would be most helpful, Philostorgius did write an encomium for Eunomius and would have undoubtedly followed the history of his own community in his narrative.⁹⁵ Those in favor of Sozomen’s direct dependence on Philostorgius see Philostorgius’ Ecclesiastical History as the most likely source for Sozomen’s knowledge. And yet, even here, a direct relationship between Sozomen and Philostorgius, while possible, does not seem probable in light of another option.

The clue is the logical argumentation that Sozomen employs. His account is known for its generally maladroit handling of religious matters. Whether it is his combination of history with hagiography or his desire to gloss over theological divisions, Sozomen’s theological discourse is usually not very astute; he often avoids or clumsily handles such information.⁹⁶ In 3.15.10, for example, he declares that it is the task of the theologian and not the historian to make judgments regarding orthodoxy, and he himself admits in 7.7-8 that he lacks the dialectical skill to engage a particular theological subject. Interestingly, the subject about which he refrains from further discussion is none other than Eunomian baptism and the divisions that arose from their diverging practices. What a contrast with the preceding 6.26, in which the reader discovers a detailed, logical argument based on thorough knowledge of Heterousian beliefs or, to be more precise, the multiple beliefs of the Heterousian community as it began to fragment. As van Nuffelen has demonstrated, Sozomen operated on a principle of accumulation: He incorporated any number of sources and used them side-by-side to craft his narrative.⁹⁷ The result is occasionally a collage of contradictory material when a statement in one place explicitly or implicitly gives testimony against a previous declaration. The material from 6.26—a confident rebuttal of Heterousian theology and practice—stands in clear contrast with Sozomen’s apparent hesitation to engage in tangled polemics.

At this point it becomes useful to anticipate arguments from the following section regarding the existence and content of the Ecclesiastical History by Gelasius of Caesarea.⁹⁸ We recall that scholars accept as an authentic writing from Gelasius his non-extant treatise against the Heterousians.⁹⁹ Gelasius had an evident interest in anti-Heterousian polemics, and this concern seems either to have found its way into his Ecclesiastical History prior to his independent treatise or, if the treatise came first, he, like many scholars, may have incorporated his previous work into his later project. Either way, we suggest here, in anticipation of arguments later this chapter, that Gelasius’ Ecclesiastical History may have provided the details and arguments discerned in Sozomen 6.26 rather than Philostorgius’ history.

3.9 Sozomen 6.37.2-5 and Philostorgius 9.17

The sources underlying Sozomen’s account of the conversion of the Goths and their entrance into the Roman Empire has long been contested. Though Peter Heather has argued that Sozomen’s account—with its depiction of the Goths’ conversion after their entrance into the Empire—is a corrective to that of Socrates, Noel Lenski has demonstrated that this is not likely to be the case.

Chronology is the crux of the debate. In what order did the Goths convert to Christianity, have a civil war, and ultimately enter the Empire with the fateful permission of Valens? Socrates 4.33-34 unequivocally states that the conversion occurred north of the Danube, that is before the Goths entered Roman territory. According to his narrative, a Gothic civil war ended with Roman assistance for one faction, which, in turn, converted to Homoian Christianity. Only after a persecution by an opposing Gothic leader and the invasion of the Huns did the Goths appeal to Valens and receive permission to cross the Danube.

Sozomen 6.37 revises this order significantly. He begins with the Hunnic threat and the successful appeal to Valens. Next followed the civil war and the subsequent conversion out of gratitude for Roman support. In addition to this alteration, Sozomen presents a number of other details, such as an anecdote regarding how the Huns discovered they could cross a lake and invade the lands of the Goths.

Theodoret 4.37.1-5 follows Sozomen in placing the conversion of the Goths south of the Danube. However, Theodoret’s account is particularly poor. His narrative credits the conversion of the Goths, at least in part, to Eudoxius, who had died several years before the Goths crossed the river. Furthermore, it appears that he simply conflates all material he had on hand regarding the Goths into a compressed timeframe so that events that occurred years before—such as Valens’ visit to Tomi—seem to happen just before his fateful encounter at Adrianople.¹⁰⁰

Noel Lenski’s attempt to make sense of this confusing array of information and error is not flattering to Sozomen. In contrast to Peter Heather, who believes Sozomen’s account is based on sources unavailable to Socrates and therefore more accurate, Lenski attributes the shift of the conversion from the north to south of the Danube to simple carelessness: Sozomen, relying on one source that described the Hunnic invasion, became “carried away by the force of his narrative” and told the story of the Huns and the Gothic flight before consulting Socrates’ account of their earlier civil war and conversion.¹⁰¹ Sozomen, in Lenski’s opinion, decided not to alter his account but rather to force the data from Socrates into his new, erroneous chronology.¹⁰² While this view certainly does not bolster one’s esteem for Sozomen, Lenski’s arguments are plausible and coincide with the perspective that this study has been forming regarding Sozomen’s scholarship.

Our main interest is Sozomen’s source. His account is particularly noteworthy for its awareness of Homoian Gothic martyrs during a period of persecution. In a similar vein, he also presents a number of details about the bishop Ulfilas, the Homoian missionary, that Socrates does not have. Philostorgius is an obvious potential source. Philostorgius reveals his sympathy for Ulfilas’ activity in 2.5, and, though his account of the Goths in 9.17 fails to mention any conversion or persecution, this may be due to the epitome rather than the original text. Alternatively, Sozomen perhaps used Eunapius for the secular account and, as Lenski suggests, advanced his narrative too far before turning to Socrates and/or someone else for the ecclesiastical details.¹⁰³ This could explain how he presents both religious and secular details—such as the story of how the Huns learned they could cross the lake—with his convoluted mangling of Socrates’ account forming the link.

There is a historiographical danger of suggesting lost sources as a general panacea whenever a problem is encountered, but it is worthwhile to point out that another connection may exist here between Sozomen’s account and that of Gelasius. Gelasius, as Lenski notes, probably had an interest in Gothic Christianity, and it is no stretch to imagine that he would include their conversion and subsequent persecution.¹⁰⁴ Nor, given the parallels discussed above, would it be surprising if in Gelasius’ history the Church’s story eclipsed any political narrative regarding those north of the Danube. Perhaps Gelasius himself made a mistake and led both Sozomen and Theodoret astray, or, as Lenski indirectly suggests, maybe Gelasius, like Theodoret, altered his account to serve a polemic purpose.¹⁰⁵ A revised or tendentious chronological presentation could do for Gelasius what it does in Theodoret’s Ecclesiastical History: The Gothic conversion to Homoian Christianity and Valens’ subsequent death at Adrianople would be firmly linked and it could serve as a salutary warning against imperial support for heresy. At the very least, Gelasius’ Ecclesiastical History represents a viable option in solving the puzzle of Sozomen’s sources and how it is that he possesses more data than Socrates does and yet offers a poorer narrative.

3.10 Sozomen 7.6.1-6 and Philostorgius 9.4

As with Sozomen 6.26.2-6 above, 7.6.1-6 is an independent chapter rich in details only hinted at elsewhere. Sozomen records Eunomius’ popularity despite his withdrawal from his see of Cyzicus. So great, Sozomen claims, was his eloquence esteemed that Theodosius himself sought to attend his discourses; it was only his pro-Nicene wife Flacilla who prevented this.

Many suspect that this material derives from Philostorgius. Philostorgius 9.4 describes Eunomius’ retirement and how solicitous he was for his visitors. This account also informs us that, though Eunomius no longer performed liturgies or acted as bishop himself, all the bishops of the Heterousian communities sought his approval before acting. We also note that Gelasius’ own evident interest in the Heterousian world makes it just as probable that he would write of Eunomius’ affairs while in retirement.

Our incomplete knowledge of Philostorgius’ actual text again hampers our ability to make any conclusive judgments. Arguments could be made for the Heterousian’s influence while counterarguments could be made regarding Gelasius’ possible input, particularly regarding the role of the empress in guarding Theodosius’ orthodoxy. The episode immediately following in Sozomen’s account makes it clear that some pro-Nicene source did not indulge in unqualified adulation of the pro-Nicene Theodosius.¹⁰⁶ The reader learns that the emperor once suffered a rebuke from an old priest for continuing to hold discourse with those who dishonored God’s Son. According to the story, this priest addressed Theodosius with all due decorum but treated his son, the future emperor, in a familiar manner and without any becoming respect. When Theodosius grew angry at the indignity paid to his son, the priest used the moment to explain how God became angry when His Son was not honored by the “Arians.” According to Sozomen’s account, Theodosius took the reprimand to heart and eventually promulgated a law to limit public discourse on the subject of the Son’s relationship to the Father.

Theodoret 5.16 also includes this tale of Theodosius’ edification, though he adds a particular detail that changes the entire nature of the story. Whereas Sozomen speaks of an old priest, “simple and inexperienced” (ἁπλοῦς καὶ πραγμάτων ἀτριβής), Theodoret identifies the man as Amphilochius, a prominent member of the Cappadocian hierarchy and an astute bishop. It is probable that Theodoret and Sozomen are drawing from the same, unidentified source, so whence this discrepancy?

Though Theodoret could easily have tampered with his material, as he did with his chronology regarding the Empress Flacilla’s death, Sozomen is just as likely to change his material to suit his narrative needs.¹⁰⁷ Why replace a prominent bishop with the stock character of an anonymous wise, old priest? Sozomen’s preceding Book 6 presents a lengthy catalogue of holy hermits and wonderworkers. Perhaps, having established for the reader the effectiveness of these simple, yet influential, souls, Sozomen decided to expand that theme by replacing Amphilochius with someone who could easily have fitted among the figures he had described just shortly before.¹⁰⁸ That Sozomen removed Amphilochius rather than Theodoret adding him is hinted at in the text of Sozomen. When he introduces the episode he states that the bishops then residing in Constantinople (τῶν ἐνδημούντων ἐπισκόπων τῇ Кωνσταντινουπόλει) came to salute the emperor. How it was that among this company would be a humble priest is not explained. Meanwhile, Amphilochius was holding regional synods against other sects in his diocese the year this episode happened.¹⁰⁹ Given this bishop’s efforts on behalf of Nicene orthodoxy, it seems likely that, if this episode did derive from Gelasius’ account, it did so in a manner closer to Theodoret’s presentation.

3.11 Sozomen 7.17 and Philostorgius 10.6

Socrates 5.20.4-5 and 5.24.3-5 includes information regarding Eunomius’ last days in the suburbs of Constantinople and the division of the Heterousian community into at least two opposing camps, but Sozomen 7.17 is far more specific. He informs his audience where Eunomius was buried and the specific theological opinions of two Heterousian factions, the Theophronian and the Eutychian. Sozomen’s knowledge of the particulars within the Heterousian community is impressive, and it would be helpful to learn what Philostorgius’ narrative said at this point. Sadly, Photius’ epitome fails here: It presents only hints of the evolving situation among the Heterousian communities and cannot independently prove or disprove Sozomen’s dependency on Philostorgius. Philostorgius 10.6 is usually given as the potential parallel, though the majority of the information Photius preserved regards Philostorgius’ personal meeting with Eunomius and the physical attributes of the heresiarch. 10.6 does, however, relate how it was that Eunomius came to Dacora, his birthplace and the site, according to Sozomen, of his burial. Sozomen’s material may have come from Philostorgius, but it is impossible to state with certainty whether it did. It is also possible that this information could have derived from Gelasius and his particular interest in the Heterousian community.

3.12 Sozomen 7.24.3-4 and Philostorgius 11.2

Sozomen’s account of Theodosius’ battle against the pagan Eugenius is linked to that of Philostorgius on account of the treachery that enabled Theodosius to secure both the high ground to his rear and the enemy soldiers supposedly guarding that area. While Sozomen’s narrative presents an elaborate story of the exchange between Theodosius and the traitors about to switch to his side, all that remains in the epitome of Philostorgius is the notice that the mountain passes were betrayed to Theodosius. Unfortunately, there is simply too little here to attempt either to establish or refute dependency on Philostorgius’ narrative. Socrates’ narrative, meanwhile, is generally superior and has more details, lacking only the treason and Sozomen’s account of how John the Baptist answered Theodosius’ prayers.¹¹⁰

4 Sozomen’s Use of Gelasius of Caesarea

The parallels between Sozomen and Philostorgius do not prove conclusively that Sozomen relied immediately upon the Heterousian historian. Though that possibility remains, this study suggests instead a connection between Sozomen and the Ecclesiastical History of Gelasius. Gelasius’ interest in the leaders and history of the Heterousian community could explain Sozomen’s awareness of the same. So, too, other aspects of Sozomen’s history, such as his focus on Antioch and interest in ecclesiastical events in the West, fit, or at the least do not contradict, our expectations of what an ecclesiastical history of Gelasius would presumably have included.

Van Nuffelen, as we have seen above, argues convincingly that Socrates was unlikely to plagiarize Gelasius of Caesarea if he had actually used any writings from the bishop. While Socrates may not have acted in such a way, Sozomen certainly could have. This suggestion is, of course, complicated by the recent doubts as to whether Gelasius wrote a history. Even if one accepts the history’s existence, there is little to guide any study other than the confusing statements from Photius noted in Chap. 3. Too many holes riddle our knowledge to make any one theory safe from criticism, and the ideas proposed below remain merely an interpretation of the few remaining clues. If Gelasius’ account existed and was used by Sozomen, it opens new insights regarding the relationships among the first church historians. The theory also could help us propose which parts of Sozomen’s narrative possibly have a relationship to the Lost Arian History.

The first task is to establish the plausibility that Sozomen included another source as a framework for his own narrative in addition to Socrates’ history. Van Nuffelen’s table of Sozomen’s sources proves invaluable in this search for evidence of common themes or emphases within Sozomen’s History.

The episcopacy of Cyril of Jerusalem, Gelasius’ uncle and the reputed instigator of Gelasius’ historical work, presents a number of problems in terms of its chronology and is a good point to begin the reevaluation of Sozomen’s sources.¹¹¹ There are two traditions in the historiography. One, preserved by Jerome and Epiphanius, highlights a series of depositions and re-installments for Cyril during the turbulent decades between Constantius’ last years and the death of Valens; the other relies on the testimony of the ecclesiastical historians Socrates and Sozomen, whose accounts—with one crucial variation—imply a continuous non-Nicene occupation of the see of Jerusalem before Cyril’s restoration.¹¹²

The actual events do not concern us as much as the differences between Socrates and Sozomen. As noted above, Sozomen operated on a principle of accumulation: He incorporated different sources and used them side-by-side to inform his history.¹¹³ The result can be a contradictory narrative as a statement in one place explicitly or implicitly gives testimony against a previous declaration. So it is in Sozomen 4.30.3 and 7.2.2. The latter, which mirrors Socrates 5.3.1, presents us with a Cyril who is in control of the see of Jerusalem at the time of Theodosius’ accession. However, in 4.30.3 Sozomen clearly states the opposite of this position: Cyril’s deposition was followed by three “Arian” interlopers, and only after Theodosius came to the throne was Cyril restored to his rightful place.

Socrates’ presentation may also give evidence of incorporating contradictory material.¹¹⁴ While he maintains the intermission of three “Arian” bishops, he clearly states that Cyril was in control not only when Theodosius came to power, but also during the reigns of Julian and Valens.¹¹⁵ This presentation, if taken at face value, would imply that Cyril lost his see during the council of Constantinople in 360, witnessed three intervening bishops between the years 360 and 363, and then returned as the bishop of Jerusalem before May of 363 in order to be present when the Jewish community attempted to rebuild the temple.¹¹⁶ Van Nuffelen argues that Socrates most likely had two sources in front of him, a chronicle that mentioned the three bishops and another, known to be the history of Rufinus, which highlighted Cyril’s presence in Jerusalem during the reign of Julian.¹¹⁷ These Socrates combines into a seemingly contradictory or at least highly unlikely chronology.

Sozomen independently consulted sources employed by Socrates, but he also incorporated materials that Socrates did not. If Sozomen is here independently utilizing the same source as Socrates then van Nuffelen’s conclusion seems the most likely: Sozomen saw the improbability inherent in Socrates’ chronology and avoided the situation by ignoring Cyril’s activity during the reign of Julian and emphasizing Cyril’s deposition until the rise of Theodosius.¹¹⁸ Yet Sozomen may have drawn upon a source unknown or unused by Socrates in framing his more believable chronology for the bishops of Jerusalem in 4.30.2. Furthermore, given Cyril’s troubled legacy and the suspicions that many held regarding his theology in the wake of events in the 360s and 370s, those in favor of the bishop, as presumably his nephew Gelasius was, probably desired to emphasize his punishment during the reigns of those recent anti-Nicene villains, Julian and Valens.¹¹⁹ Even if Cyril exercised (or attempted to exercise) some sort of authority during these two decades as Jerome and Epiphanius’ accounts propose, it is understandable how a sympathetic historian would stress instead his lack of authority until the reign of Theodosius.¹²⁰

Sozomen’s account of Cyril’s deposition supports the belief that he is using a different source from Socrates. Having run afoul of Acacius of Caesarea, with whom Cyril struggled for regional ecclesiastical supremacy, the bishop of Jerusalem was deposed at the council of Constantinople in 360 and replaced by someone more sympathetic to Acacius’ authority and Homoian theology. Socrates merely mentions that Cyril stood accused and adds the detail that this accusation against the bishop of Jerusalem had already been raised before (ἤδη πρότερον), though he admits that he is ignorant of the precise nature of the charge leveled against Cyril.¹²¹ Sozomen, in contrast, has a number of details. He writes that Cyril had admitted several Homoiousian prelates to communion despite their opposition to the rising Homoian party and that Acacius and Cyril already mistrusted each other’s religious orthodoxy. The climax was Cyril’s decision to sell church goods in order to feed the poor during a time of famine in Jerusalem. According to Sozomen, a man recognized an item that he had dedicated to the church while gazing at the details of an actress’ costume and realized that Cyril had sold his offering. His complaint served as the pretext for Acacius’ action against Cyril in 360.

Theodoret, interestingly, also includes this story, though he introduces a number of variations that add to the drama.¹²² In Theodoret’s rendition, the important historical figure of Constantine is associated with the sold church goods, and the bishop-historian even includes a cautionary death. These dramatic flairs may enhance excitement, but they do not change the substance of the narrative: For both Theodoret and Sozomen, Acacius used the sale of church property as a pretext to depose Cyril.

Van Nuffelen does not believe that this parallel between Theodoret and Sozomen is indicative of any new source used by the latter. He argues that Sozomen’s account, despite its obvious difference with that of Socrates, is a continuation of their common source Sabinus.¹²³ However, his argument is only partially convincing. True, Sozomen gives notice that Cyril’s reception of Basil of Ancyra and George of Laodicea caused friction between Cyril and Acacius. Sabinus could have mentioned the condemnation of these two prominent Homoiousian prelates at the hands of Acacius’ Homoian party. Furthermore, this connection between Cyril and the Homoiousians would likely derive from either a Homoiousian source, like Sabinus, or another that did not have a vested interest in promoting Cyril as a staunchly pro-Nicene figure. Nonetheless, the concluding episode, which tells of Cyril’s sale of items dedicated to the church, seems to do violence to what little is known of Sabinus’ Collection of Synodal Acts. Sabinus did offer his own commentary to the collection of documents that he had gathered, but these editorial comments do not imply any extensive narrative, unless someone recorded at length the testimony used against Cyril at the Council of Constantinople. Sozomen is here possibly combining two sources (if we accept that the reference to Basil and George derives from Sabinus or elsewhere), one of which Socrates did not possess and which told the story of the dedicated goods.¹²⁴ Otherwise, Socrates’ concluding comment becomes unintelligible: Why would Socrates write that he is unable to give the reasons for Cyril’s deposition (οὐκ ἔχω ϕράσαι) if Sabinus had not only listed several but included a lengthy narrative to that effect?¹²⁵

Theodoret’s version presents a different problem. According to Ernest Bihain, Theodoret likely drew his material from the writings of Theodore of Mopsuestia.¹²⁶ Theodore had written about the struggle between Acacius and Cyril in his Contra Eunomium; van Nuffelen accepts that Theodoret may have used Theodore or his own particular knowledge to write of this event as well as Cyril’s exile to Tarsus, a detail not found in Sozomen.¹²⁷ It is impossible to know with certainty where Theodoret received his information, but the possibility that Theodoret (or Theodore) and Sozomen consulted the same source for this episode remains plausible.

In fact, other parallels shared between Sozomen and Theodoret open the possibility that both accessed material unavailable or unused by Socrates. There are a number of points in the two narratives where a common, but unknown, source seems possible. What is fascinating about these particular moments is the focus they have on ecclesiastical events in the West. No fewer than seven times Sozomen and Theodoret include parallel information that provides the reader with western events or writings.¹²⁸ While there is no conclusive evidence to confirm a single underlying source, it appears likely that both used some source that was either focused on the western church or included stray details about events and personages there.

This western-focused source may also inform Sozomen 7.16. Though there are parallels with Socrates’ narrative (5.19), Sozomen presents liturgical practices in the western, particularly Roman, church that are absent in Socrates. Furthermore, Sozomen clarifies a crucial detail in Socrates’ account of a certain scandal within the Constantinopolitan church: Whereas Socrates simply mentions that a female penitent slept with a deacon after receiving her penance and thus outrage arose, Sozomen explains that the deacon raped the woman while she was performing the prescribed prayers and fasts in the church. Sozomen’s clarification helps explain why there was such a tumult among the people and how it was that the office of the penitentiary presbyters was abolished in Constantinople. Though Theodoret does not include this episode, this absence presents no difficulty in the theory of a common source shared by Theodoret and Sozomen. Such an unbecoming act by a member of the clergy would not fit well into Theodoret’s narrative which so often seeks to elevate the position and authority of ecclesiastics.

In addition to this western perspective which Sozomen shares with Theodoret, Sozomen’s Ecclesiastical History appears to have had a pro-Nicene source that focused on the ecclesiastical confusion in Antioch. 3.20.4 and 3.20.8-9—with its corollary 4.28—have many independent details that enhance the material found in Socrates and do not derive from that earlier account.¹²⁹ Sozomen’s 6.21.1-2 and 6.25.1-5 may also draw on some source for Antiochene information that Socrates lacks: Both of these sections offer specific material from an anti-“Arian” perspective regarding the upheaval in the Antiochene church.¹³⁰ 6.25.1-5, it should also be noted, flows naturally from its account of intra-Antiochene disputes into the wider, western perspective already noticed in Sozomen’s narrative, perhaps indicating a single source for Sozomen’s information for this chapter.¹³¹

Here, too, possible parallels with Sozomen occur in Theodoret’s history. For example, their two accounts of Meletius’ preaching in Antioch dovetail nicely. Theodoret 2.31 shares Sozomen 4.28’s positive portrayal of Meletius and recounts many of the same specific details, such as Meletius’ triumphal entry into the city under the gaze of both Christians and non-Christians. Moreover, their accounts supplement each other since each lacks an important detail found in the other: In Theodoret one reads some of Meletius’ very words during his fateful, pro-Homoousian sermon; Sozomen, meanwhile, supplies the crucial detail, lacking in Theodoret, that an indignant “Arian” deacon tried to cover Meletius’ mouth during this sermon.¹³²

Sozomen’s account of Antioch in 6.21 may indicate another parallel between the two. Following this notice, Sozomen presents a detailed narrative of the pro-Nicene Scythian bishop Vetranio and his refusal to accommodate the heretical tendencies of Valens when that emperor visited the churches under Vetranio’s charge. Sozomen relates Valens’ decisions to exile and, later, to recall the bishop and concludes with statements regarding the Scythians’ esteem for Vetranio’s virtuous example. Theodoret 4.35 lacks many of these details though he stresses that Vetranio, the bishop of the Scythians, protested the heretical decisions of the emperor and used the words of the prophet David to that effect. Though Theodoret’s account is briefer than Sozomen’s, his inclusion of Vetranio’s use of the psalm may reflect a detail in a common source that he incorporated and that Sozomen chose not to include.¹³³

In a similar manner, Theodoret 5.4.1 mentions the same Antiochene presbyter Vitalius as does Sozomen 6.25.1-5, though, again, Theodoret foregoes many specifics. Either Theodoret’s edition of their presumed shared source was somehow lacking many details or, more likely, he chose to excise more details for his briefer narrative. Interestingly, the Paschal Chronicle also mentions this Vitalius, though from a very different perspective.¹³⁴ Whereas Theodoret and Sozomen represent Vitalius’ embrace of Apolinarius’ heresy as a fall from the faith, the Paschal Chronicle exhibits continuous hostility toward both the pro-Nicene Melitius and his presbyter Vitalius. This is not surprising since that section from the Paschal Chronicle appears to have derived from the Lost Arian History.¹³⁵

There are any number of possible solutions to the questions subsequently raised regarding Sozomen’s and Theodoret’s source(s). The proposed history of Gelasius of Caesarea seems the most economical answer to the different narrative threads identified above. No information survives regarding Gelasius’ movement (or lack thereof) while he was unable to fulfill his role as bishop of Caesarea. It is at least possible that he, like Athanasius and others, journeyed westward to Rome in order to plead his case. Such travel could explain why it is that information regarding the western church, which Socrates so often ignores, features so prominently in Sozomen’s account. Alternatively, Gelasius could also have possessed an interest in the struggle for orthodoxy in the west and included it in his narrative. It is also significant that the pope of Rome recognized and confirmed Cyril in his episcopacy, a decision that may have of its own accord won from Cyril’s deferential nephew a prominent place for that see in his narrative.¹³⁶

That this western-focused data probably derived from a single source can be deduced from the aforementioned parallels between Theodoret and Sozomen, but doubts may remain as to whether this should be attributed to Gelasius. Fortunately, there is other evidence that implies a Gelasian source. Sozomen 6.22 treats of the Macedonian heresy.¹³⁷ This material, whose source is unknown, combines the above-mentioned western focus—Sozomen introduces the bishop of Rome’s reaction—with another area of probable interest for Gelasius. The Macedonian heresy would likely have been important to Gelasius, a Council Father at the ecumenical gathering in Constantinople that addressed that very heresy. There is also Sozomen’s Antiochene source, which has a clear pro-Nicene tendency, unlike the fragments identified by Bidez as belonging to the Lost Arian historiographical tradition. In addition there are the details which Sozomen included regarding the Heterousian community, particularly his polemic argument of 6.26.2-6.

Even the terminus of the history appears to coincide with what little is known of Gelasius’ biography. According to van Nuffelen’s study regarding Sozomen’s source material, the frequency of an unknown source(s) dwindles significantly in Sozomen’s Books 8 and 9.¹³⁸ This change could imply the end of Gelasius’ Ecclesiastical History and would leave us with events such as Theodosius’ victory over Eugenius or an otherwise-unattested miracle of the Nile (7.20.3-5) as possible termination points for Gelasius’ history. Both would be fitting conclusions to demonstrate how Christianity had survived the turmoil of the fourth century and was now pursuing its victory over paganism, a worthy theme for the successor to Eusebius of Caesarea.¹³⁹ It is also possible that Gelasius wrote until the death of Theodosius.¹⁴⁰

If this hypothetical ending is accurate, it explains how it was that Gelasius’ proposed history never became better known. If he died shortly after its completion or perhaps even in the act of writing it, only a handful would probably have known of it.¹⁴¹ Furthermore, the history would likely have remained in Palestine, a situation that would simultaneously help explain Socrates’ apparent ignorance of it and Sozomen’s awareness of it. Sozomen, a native of Palestine who, based on the content of his history, continued to have an obvious interest in that region, may have had connections in that part of the empire who alerted him to the existence of Gelasius’ history. It may be that circumstances akin to those of Socrates surrounded Gelasius, so that Sozomen believed he could incorporate the works of both without risk of censure. Theodoret, meanwhile, being a well-connected bishop in the Syrian and Palestinian world, could have learned of Gelasius’ history in a number of ways and used it as his resource, even if he wrote in exile. This would thus explain the parallels that Theodoret and Sozomen have with each other.

There are a number of remaining riddles, such as Sozomen’s source for the discovery of the True Cross, that could be solved by considering a Gelasian influence.¹⁴² These, however, also rely on circumstantial evidence and a detailed discussion would take us even further afield from our present study on the Lost Arian History.¹⁴³

One last observation, while not conclusive, provides additional support for the theory of Sozomen’s dependence upon Gelasius. Gelasius, one of the attendees of the Ecumenical Council of Constantinople in 381, would have presumably written of it at length if he did compose a historical narrative. In this regard Sozomen’s account does not disappoint. He presents a dramatically fuller account of the gathering than his predecessor Socrates does.¹⁴⁴ Clearly, Sozomen has another source at his disposal. Van Nuffelen tentatively suggests that he consulted the Acta of the Council, though the proposed history of Gelasius would likely have included similar material.¹⁴⁵ Given what is known of Sozomen’s habits as a historian, he may have sought his information in a single source, and his account of the council not only includes its decisions but also narrative details that might indicate more source material than that gathering’s Acta.¹⁴⁶

And yet there is a difficulty here: 4.27 states that the Macedonians were without churches until the reign of Arcadius, but Socrates 2.45 (and Sozomen 7.2) informs us that they regained them under Gratian. Furthermore, the fact that this passage refers to Arcadius may signify a non-Gelasian source since Gelasius had likely died around the time Arcadius came to the throne or shortly thereafter. See also 7.21 and 8.5 in which Macedonian events are mentioned that, at least in the case of 8.5, clearly take place after the presumed death of Gelasius and come from an unknown source. Does their presence undermine the theory above? While these reflect a later source, or maybe Sozomen’s own knowledge, their presence does not automatically contradict the collected evidence for Gelasius. The fact that Theodoret, whose account offers so many parallels otherwise, includes neither may suggest that these instances derive from another source than the one he and Sozomen seem to have shared. Perhaps 8.5 comes from a source focused on John Chrysostom, while 7.21 may reflect local tradition. It is simply impossible to say with certainty.

Based on the circumstantial evidence presented above, it is possible that none other than the Ecclesiastical History of Gelasius of Caesarea formed the basis for much of the Palestinian, Syrian and western content of Sozomen’s account. The details about Cyril of Jerusalem, the Council of Constantinople, and heresy all reflect interests that Gelasius would presumably have had.¹⁴⁷ Furthermore, the numerous parallels between Sozomen and Theodoret, who as a bishop of Syria might have also learned of Gelasius’ history, suggest that a single source provided both this western perspective and details focused on Antioch. This source, furthermore, was pro-Nicene, and we can exclude much of Sozomen’s un-attributed material from our present study on the Lost Arian History.

Nonetheless, there remains the potential link with fragment (35) of the Lost Arian History in Sozomen 5.8; this has obvious similarities with sources that formed part of Bidez’s reconstruction, though this is not in itself conclusive.¹⁴⁸ Van Nuffelen believes that Sozomen 2.29.5, with its awareness of the “Arian” theory that Arius died through black magic, also ultimately derives from the lost history.¹⁴⁹ There is also fragment (30) and its parallel with Sozomen 4.16. Van Nuffelen believes that this derives from Sabinus and that Sozomen’s reaction to the non-Nicene sentiment should be seen as a reaction against that Homoiousian author.¹⁵⁰ It seems just as possible that Sozomen’s reaction is actually an echo of Gelasius’ Ecclesiastical History and its indignation at the Lost Arian History’s claims. Gelasius, who we know possessed a keen interest in the theological disputes of the day, may have been aware of the Lost Arian History through his own research.¹⁵¹ In fact, perhaps he even wrote to counteract its narrative or to transform it into a pro-Nicene one. Thus, it may be that Sozomen’s echoes of the Lost Arian History—like so much of his Ecclesiastical History—is derived from others.

5 Conclusion

Socrates, as Chap. 7 discusses more fully, may have incorporated elements from the Lost Arian History while Sozomen seems not to have consulted the lost text directly. Previously it was thought that Sozomen drew from Philostorgius and, through Philostorgius, indirectly incorporated material from the Lost Arian History, but Sozomen’s use of Eunapius adequately explains those parallels with Philostorgius’ secular material. Furthermore, Sozomen may have incorporated material from the lost history of Gelasius of Caesarea, thus explaining the pro-Nicene, western-looking material in his history. Indeed, it may be from Gelasius—and not Philostorgius—that we encounter the faint traces of the Lost Arian History within Sozomen’s narrative.¹⁵²