Eight—
Barbarians and Politics
I—
Greeks and Romans
Perhaps this story, even though it is only a story, might hint at something more.
(De providentia 88B)
Dagron has suggested a profounder motive behind Synesius's choice of a myth of two brothers opposed in every way. De providentia is for him not only a straightforward roman à clef of the events of 400, but also a much more ambitious allegory of the division of the Roman Empire, Osiris representing the East and Typhos the West.
Dagron focused on the issue of language. For centuries, Greek and Latin had coexisted in the Roman Empire with clearly delimited roles: as he put it, "le latin est la langue du pouvoir . . . le grec est la langue de la culture."[1] This distinction and the equilibrium he saw it as maintaining were upset by the shift of political gravity to the East under Diocletian and Constantine, consolidated under Theodosius; particularly decisive was the rapid growth of Constantinople, a Greek city that became a genuine political capital. In its ambit, the provinciae orientales turned into the pars orientalis . Ambitious Greeks learned Latin and flocked into the imperial bureaucracy. Greek intellectuals were hostile to such tenden-
[1] "Aux origines de la civilisation byzantine: Langue de culture et langue detat[*] ," Revue historique 241 (1969): 23–56, at p. 25.
cies, as indeed they were to Constantinople itself. The conflict of culture and power in a unified realm Dagron saw epitomized in Julian, philhellene but also "most Roman of emperors."[2]
This situation (he claims) was profoundly changed by the effective division of the empire following the death of Theodosius in 395. Gone was "the fiction of a single empire ruled by a pair of emperors and a double capital."[3] In the East, Dagron saw political crisis accompanied by a cultural upheaval, the "Nationalist crisis." This upheaval (he believes) stands reflected in Synesius's De regno and De providentia .
According to Dagron, in De regno the political presuppositions are entirely traditional: Greek cities appeal to a Roman emperor, who is responsible for administration and defense. In De providentia , however, he saw a "Hellenic" interpretation of the division of the empire. The problem posed is the succession to a good king whose death precipitates a crisis. He leaves behind him two sons of very different character. The two brothers symbolize the two halves (partes ) of the empire, Typhos as the elder representing the West, and Osiris the East. Hereditary succession would have given Typhos the throne; which is to say that historical continuity would have given seniority to the West. But instead the decision between them was made on the basis of their natural qualities, with the result that the younger brother Osiris was chosen: that is to say, the Eastern court declared itself the sole legatee of the entire Roman Empire.
This bold hypothesis calls forth reservations at many different levels. In the first place, Dagron goes far beyond the text of De regno in reading into it signs of the reciprocal "contract" between Greek cities and Roman emperor put forward in Aelius Aristeides' Roman Oration .[4] The speech gives Synesius's mission as the delivery of crown gold (2C), but crown gold had long ceased to be a distinctively Greek tribute.[5] Synesius calls Cyrene not merely "a Greek city," but one "of ancient and lofty name, in countless songs of the wise men of old" (2C–D): his emphasis is on its venerability. He does not connect the intellectualism of his approach to his city's ethnicity.[6] Nor is ethnicity an element in the relationship he draws between the cities and the ruler; simply, they are dependent, and he is their protector.[7] "Pas de doute que le roi soit 'latin,'"
[2] Dagron 1969, 29. This position rests ultimately on his interpretation of Julian's Letter to Themistius : see Dagron 1968, 60–74. The many issues raised in this connection deserve to be discussed but would take us far from the present subject.
[3] Dagron 1969, 29.
[4] Cf. Dagron 1969, 25.
[5] In fact, as early as Augustus's triple triumph in 29 B.C.: see Millar 1977, 141.
[6] Appeal through claims of philosophy was Synesius's personal favorite tactic; see pp. 89–91 above.
[7] 27A–D, pace Dagron.
indeed; but the fact that Synesius uses exempla from both recent and remote Roman history in no way cuts him or his city off from Roman traditions.[8] Conversely, Arcadius is expected to learn from Greek exempla too.[9] More positive evidence is that in context of the one ethnic distinction Synesius does make, that with the barbarians, he uses "Romans" and the first person plural interchangeably. The most clear and concise single example is the cry "We must recover the spirits of Romans " (
Next, as to Dagron's interpretation of De providentia , it is more than doubtful whether contemporaries would have recognized anything so dramatic and permanent as a partitio imperii in 395. It is easy to multiply social, cultural, religious, and military grounds for an eventual rift between East and West, but the process neither began nor ended in 395.[11]
[8] Dagron 1969, 30; he cited 16A (court luxury), 17D–19B ("Carinus" and more recent emperors), and 24A–B (Spartacus and Crixus).
[11] Even Demougeot only argued that 395 was a key stage in the inevitable movement "de lunite[*] à la division de lempire[*] romain," not an overnight turning point. And even so it has generally been felt that she overargued the case: W. E. Kaegi, Byzantium and the Decline of Rome (Princeton 1968); against the constitutional arguments of Palanque (1944, 47–64 and 280–98) that influenced Demougeot, see Cameron in CLRE 13–16.
The year 395 has become canonical in the modern historical tradition simply because no emperor unified East and West again after that date. But when Synesius wrote De providentia it was neither obvious nor inevitable that no one ever would. The notion of the Eastern court claiming seniority on the tenuous grounds Dagron suggests makes neither political nor constitutional sense.[12] Arcadius was already acknowledged to be senior Augustus by virtue of his earlier proclamation. Nor did the East dispute the automatic reversion of this title to Honorius when Arcadius died in 408; Eastern laws duly place Honorius's name before that of Theodosius II.
It is a serious error to suppose that the empire was "divided" in any significantly new sense on Theodosius's death in January 395. It had been divided between himself and the dynasty of Valentinian during most of his reign, just as it had been divided between Valens and Valentinian before that. It was only during the last three months of his life that Theodosius had reigned over a unified empire.[13] It is true that the fifth century saw the rapid growth of the idea that Constantinople was Rome's successor; not in virtue of superior merit, however, but to fill the vacuum left by the undoubted decline of the West in the half-century after 395.[14] Rome itself had long ceased to be the political capital of the West, even before the sack of Alaric. Though Milan and then Ravenna succeeded to its practical suzerainty as homes of the Western court, neither so constellated a body of traditions that it could begin to rival Rome's domination in thought. And Constantinople's eventual eclipse of Ravenna had nothing to do with the political ambitions of Arcadius and his ministers in A.D. 400.
Third, it may be doubted whether Dagron's vision of the moral superiority of literary Hellenism justifying predominance in practical affairs would have found an audience at the Eastern court. He cited all the well-known texts of Libanius, Themistius, Eunapius, and Palladas that show contempt for Constantinople or for Greeks who aspired to become Roman functionaries, owing their careers to knowledge of the law or shorthand rather than the traditional literary culture.[15] Synesius himself may have shared this attitude, but the "best men" for whom he wrote his allegory were precisely such functionaries. Indeed Libanius singles out for special mention, among the parvenus who had risen thus from the
[12] "Aux liens historiques (entre les deux partes ) on préfère les affinités politiques ou morales (la fidélité de lOrient[*] à lhellenisme[*] )" (Dagron 1969, 31).
[13] On the question of whether contemporaries would have seen a definitive partition, see above, p. 3.
[14] See Cameron, Constantinople: Birth of a New Rome (forthcoming).
[15] Dagron 1969, 28.
humblest origins to the senate of Constantinople, none other than Aurelian's father, Taurus the notary, and Philip the sausage maker's son, the grandfather of the future regent Anthemius.[16] By both origin and profession, men like Aurelian were more bureaucrats than aristocrats.
Fourth, the most curious part of Dagron's interpretation is its final stage. His view of the development of the plot is that for a while Osiris governs like a true Hellene, cultivating education and philosophy. But as a consequence he neglects the army. Profiting from the weakness of too ideal a state, Typhos usurps power with the aid of the barbarians. Osiris eventually is restored, and revises his political priorities in the light of Typhos's usurpation, "adapting them to the responsibilities of power."[17] Unfortunately, this elegant resolution of the conflict between Greek and Roman is a chimera. If the allegory is supposed to apply to the partes , no such jockeying and adjustment can anywhere be adduced. If to the ministers, it becomes relevant that Aurelian was not in fact restored to power on his return.[18] Though Synesius does his best to correct this unwelcome fact in the fiction of his myth, his readers of course knew what had really happened. At either level, it is far from clear what in practical terms Dagron's "adaptation" might involve. In any event, Synesius does not present Osiris acting any differently after his return. He offers only vague talk decked out in trite golden age imagery about how Osiris restored the prosperity of the Egyptians that Typhos had ruined (124C–125B). And since it comes before the admission that "the gods did not at once place everything in his hands" (125C), even this must describe blessings anticipated rather than achieved. It is in any case unlikely in the extreme that a thoroughgoing Hellene like Synesius would have written a work to celebrate the Romanization of a Greek ideal.
Dagron's claim that Synesius draws the contrast between Osiris and Typhos throughout in terms of Greek and Roman is farfetched in itself and unconvincing in detail. For example, while it is true that Typhos is represented as contemptuous of culture and education, Synesius praises Osiris's culture much less enthusiastically than interpreters usually suppose. The character study of Osiris in De providentia 90A–C, 91D–92A,
[16] Or. 42.24.
[17] "Synésios évoque la nécessité où est lhellenisme[*] de souvrir[*] à des domaines qui lui étaient jusque-là étrangers: larmee[*] , ladministration[*] . . . . Il ny[*] a plus dès lors de fondement à la division traditionelle entre culture grecque et pouvoir romain; le pouvoir reste dexpression[*] latine et la culture dexpression[*] grecque, mais ils sont réunis" (Dagron 1969, 32–33).
[18] As established above, chapter 5, section VI.
and 102D–104C may be construed as panegyric of Aurelian,[19] and the rules and techniques of the genre are well enough understood to illuminate clearly the text between the lines.
As a child Osiris was a keen student; but it is his keenness, not his attainments, that is praised ("eager to learn all his first elements all at once," 90A). This corresponds exactly to the rules laid down by Menander for the section on upbringing: "Then you must speak of his love of learning, his quickness, his enthusiasm for study, his easy grasp of what is taught him. If he excels in literature, philosophy, and knowledge of letters, you must praise this. If it was in the practice of war and arms, you must admire this."[20] Since Synesius does not in fact go on to praise Osiris for his learning or even for his eloquence, that touchstone of literary culture in late antiquity, contemporaries would have drawn only one conclusion: Aurelian had none to speak of.[21] Moreover, the fact that he is praised for entering public office while still a youth (91D–92A) strongly suggests that he did not pursue the advanced studies in rhetoric and philosophy that were indispensable for anyone with real cultural pretensions.[22] At 102D Synesius ascribes to him the clichéd "sacrifice to Persuasion, the Muses, and the Graces," but only as an explanation of how he could govern effectively without force. Synesius directly goes on to praise his administrative abilities and his encouragement of education. In the section on the "rustic philosopher," Osiris is praised as a "discerning judge of literature" (113C); the context makes it clear that Synesius has in mind only Aurelian's appreciation of his own work. If Aurelian had really shared Synesius's enthusiasm for literature and philosophy, Synesius would have said so much more clearly. As it is, the most he could claim even in his fiction was that Osiris/Aurelian lived up to the traditional role of patron of letters.[23] No Greek panegyrist could have claimed less for his honorand, and Themistius, to quote only one example, claimed much more for so unpromising a subject as the almost illiterate soldier-emperor Valens. It is only modern readers, unfamiliar with the conventions of the genre, who have exaggerated the signifi-
[19] Which is not to say that it bears any relation to the truth. Individual actions ascribed to Osiris in the course of the fable are in a quite different category.
[20] Men. Rh. 371, p. 83.14f. Russell and Wilson.
[21] For example, Claudian manages to say something along these lines about even the dull Honorius: "quam docta facultas / ingenii linguaeque modus!" IV Cons.Hon. 515–16. Cf. I. Sevcenko, "A Late Antique Epigram," Synthronon , Bibliothèque des cahiers archéologique 2 (1968): 29–41.
[22] A. Mueller, "Studentenleben im IV. Jahrhundert n. Chr.," Philologus 69 (1910): 292f.
[23] Cf. F. Millar, The Emperor in the Roman World (London 1977), 491–506.
cance of something contemporaries would have recognized as very faint praise indeed.
Nor would the allegation that Typhos despised culture have identified him in contemporary eyes as a Roman.[24] In fact Synesius speaks of Typhos's cultural interests only in 90C–91C, where, as with Aurelian, it is his application as a child that is at issue ("when he saw his brother going to school"). It is obviously pure invention: Synesius cannot have known which of the brothers in real life was the better student. More significantly, he never claims that the adult Typhos despised culture. Nor is it really Hellenic culture that is at issue in De providentia , whether admired by Osiris or despised by Typhos. This distinction between Egyptian and foreign wisdom implies in contemporary terms the distinction between secular and Christian culture. Aurelian is praised for his Christian, as much as for his secular, learning, and Typhos is damned for his contempt of both.[25] All in all, there seems no reason why contemporary readers should have perceived Aurelian in any sense as a Hellene.
No better founded is Dagron's claim that Osiris "accède à la royauté par la voie noble, le présidence du sénat," while Typhos "devient usurpateur après avoir appartenu à ladministration[*] du fisc."[26] The truth is that in real life the careers of Aurelian and Caesarius were virtually identical: both were magister officiorum , praetorian prefect, consul, and patrician; and of course both were appointed to all those offices by the
[24] "Accusation traditionelle des 'vieux romains' contre les 'graeculi'" (Dagron 1969, 31)—but not since before the age of Cicero! During the imperial age the stock of Greek literary and philosophical culture rose steadily until for a period in the second and third centuries it bid fair to eclipse Latin as the language of culture even among Latin speakers. In real life Caesarius was a native Greek speaker who at one point kept up a house in Antioch and was on good terms with Libanius (Lib. Or . 21, esp. 33; PLRE 1.171). If he was not a Themistius whose career was built on his culture, neither was Aurelian. But it is equally obvious that Caesarius owed his precocious career in the imperial service, not to the study of Latin or the law, but, like Aurelian, to the wealth and influence of his father. This is the obvious explanation of the fact that both brothers enjoyed accelerated careers; it is clear that neither worked his way up from shorthand writer or clerk.
[26] Dagron 1974, 205; cf. 1969, 32.
emperor, as in Synesius's story by their father. The only major post Aurelian held that his brother did not is the prefecture of the city, described by Synesius in the words
Dagron characterizes Typhos's sexual orgies as "très romaines." But although the modern European tradition likes to associate orgies with Roman decadence, contemporaries did not at all view it so. No one who has even glanced at the rich material assembled in book 12 of Athenaeus's Deipnosophistai could believe that the Romans invented the orgy.[31] The Roman who developed it to an art form was that most philhellene of emperors Nero; his more exotic sex partners all bore conspicuously Greek names (Phoebus, Sporus, Pythagoras), and his more spectacular sexual escapades were staged in Greece.[32] Roman moralists regarded such be-
[27] The second phrase should not be construed as a separate office such as princeps senatus; it is merely part of Synesius's characterization of the office of city prefect, who was the senate's representative in its dealings with the emperor. In the same way, he uses two phrases to represent magister officiorum . Cf. A. Chastagnol, La préfecture urbaine à Rome sous le Bas-Empire (Paris 1960), 69–72.
[28] On the identification of these two posts, see p. 181 n. 130 above.
[29] Dagron 1974, 205.
[30] Dagron 1969, 32.
[31] See too Burgo Partridge, A History of Orgies (London 1964), 9–37.
[32] M. T. Griffin, Nero: The End of a Dynasty (London 1984), 169, 180.
havior as characteristic of Greek corruption and decadence.[33] The chastity of Osiris's wife that is held up in honorable contrast to that of Typhos's (105D) was intended to evoke Christian, as much as Hellenic, ideals.
The whole house of cards collapses. There is no such imperial allegory in De providentia . In fact not the least interesting aspect of De providentia is that Synesius's allegory has no place for the part played by the West in the crisis of 399–400. We know, in most detail from Claudian, that successive Eastern ministers of the age continued to live in fear that Stilicho would march East and establish his "regency" over Arcadius by force. Eutropius went so far as to proclaim Stilicho a hostis publicus , probably in 397. Aurelian too must have taken a hostile line in 399 if Stilicho refused to acknowledge his consulate in January 400. Claudian refers to Aurelian's government as iners atque impia turba . A fragment of Eunapius shows Fravitta reproaching Count John for attempting to undermine the concord of the brothers Augusti.[34] Another fragment gives a vivid picture of the man in the street trying to make sense of the conflicting rumors:
During the time of Eutropius the eunuch it was impossible to include in a history an accurate account of events in the West . . . . Any travelers[35] or soldiers with access to political information told the story as they saw fit, biased by friendship or hostility or a desire to please someone. And if you brought together three or four of them with conflicting versions as witnesses, there would be a furious argument, a pitched battle proceeding from passionate and heated interjections like the following: "Where did you get this from?" "Where did Stilicho see you?" "Would you have seen the eunuch?" so that it was quite a task to sort out the tangle.[36]
It is not too much to say that the Eastern and Western courts were in a more or less permanent state of cold war between 395 and Stilicho's
[33] This is not to say that there were no differences between Greek and Roman orgies, notably in the role of food and women. At Greek symposia, "women were present only for the purposes of entertainment and sexual pleasure: the hetaira, the dancer and the flute girl" (Oswyn Murray, JRS 75 [1985]: 40). Romans allowed their wives to attend, which inevitably complicated matters (Murray, 48–49). Since Typhos's wife attended his orgies, to that extent they might be classified as Roman rather than Greek, but it is hard to believe that Synesius, who had no knowledge of Latin literature, was aware of this distinction.
[34] 85M, quoted in full above, p. 246.
[36] Frag. 74M = 66.2B; the translation is modified from Blockley's.
death in 408, with a slight temporary thaw during 401–3.[37] One later letter to his brother shows Synesius at least curious about news from the West,[38] but De providentia is concerned exclusively with the struggle for power in the East.
This is not in itself surprising. In the West Stilicho ruled without rival in the name of Honorius, whom he soon took the extra precaution of making his son-in-law. He controlled the army as well as the court. In the East, however, there was a constant struggle for power, both among individual civilian ministers and between civil and military authorities. Gaïnas's military coup in 400 brought about a major internal crisis, and it is understandable that Synesius should have concentrated on it. Even so, there must have been a constant fear that Stilicho would take this opportunity of intervening in Eastern affairs. In late 399 his propagandist Claudian had gone so far as to suggest that only Stilicho could save the East from the ravages of Tribigild. But Synesius completely ignored this dimension of the crisis.
In this respect De regno , which Dagron contrasted with De providentia , shows exactly the same limitations. Claudian's panegyrics on the third and fourth consulates of Honorius, delivered at Milan in January 396 and 398, make frequent reference to the East and to Honorius's brother and colleague Arcadius. Synesius writes as if there were only one emperor and one court, in Constantinople.[39]
Under the circumstances, there is little enough to suggest to the reader that De providentia , which already contains one sort of political allegory, might also contain another on a subject in which Synesius nowhere else shows any interest.[40] It might be added that any contemporary tempted to interpret the death of the old king and the struggle between his two sons in terms of the imperial crisis after Theodosius would naturally have thought first of Arcadius and Honorius rather than East and West. And most would have abandoned this line the moment they discovered that it was the younger son (Honorius) who won the election and the older (Arcadius) who was the villain of the tale. There are only so many balls that even the most sophisticated political allegorist can keep in the air at once.
[37] See pp. 246–50.
[40] "Dobbiamo subito dichiarare che da tutti i suoi scritti non traspare mai il problema dei rapporti fra Roma e Costantinopoli" (Bettini 1938, 40).
II—
Political Exploitation of the Myth
The philosophical and religious terms in which Synesius tells his tale suggest that the truths of philosophy and morality are the primary aims of his book. The preface states that he has investigated "many philosophical issues hitherto undecided," described lives that are "examples of vice and virtue," and "elaborated and embroidered the work throughout with a view to its utility" (88B). Most important of all, he claims that his tale justifies a thesis about divine Providence.
It was the events of the spring and summer of 400 that inspired Synesius to these far-reaching conclusions. But it was the conclusions, not the events, that were important to him. The events were known to all. He was not writing a history. Modern historians prepared to translate any piece of narrative into the events of A.D. 400 must pause and duly weigh the essential nature of their source. We cannot simply transfer to Caesarius Typhos's every word and act. Nor can we take the speech of Typhos's wife, whose dramatic context presents it as a tissue of skillful lies (108C), out of that context, and represent it as a report that reached Gaïnas's ears in the spring of 400.[41] It is instructive to contrast the very different emphasis of a literary study: R. Helm classified Synesius's book first and foremost as a work of fiction with a moral purpose.[42]
What was the advantage of casting his pamphlet in this form? It was not that he needed allegory to license imagination and invention. Claudian's poems well illustrate how thoroughly contemporary events can be transformed without mythical disguise. The main advantage is that the myth chosen predetermines what sort of behavior will be expected of the real-life models. To give a simple example, once the author has begun to depict the rivalry of two men in terms of, say, Hare and Tortoise, then there are certain elements both in their characterization and the development of his plot that he can take for granted. However well the Hare character seems to be doing at a given moment, we know that he cannot win. However black things look for Tortoise, we know that he will win in the end. A skillful writer will be able to play with or against these expectations in a variety of ways.
What about the real-life struggle between Aurelian and Caesarius did Synesius see reflected in the myth of Osiris and Typhos? At a simple level, there is the opposition between good and evil. A Typhos cannot be capable of a single decent thought or action. So whereas we might
[41] "On racontait au chef" (Demougeot 1951, 246 n. 57).
[42] Der antike Roman (Berlin 1948), 22; the rest of his treatment is disappointing, consisting of little more than a lengthy plot summary.
question whether a former prefect and consul like Caesarius would see the destruction of Constantinople as his best way of recovering the prefecture, we are not at all surprised that Typhos should act with irrational violence to gain his ends. More important are the expectations that the myth creates. Synesius's original plan extended only to book 1. While he wrote, Caesarius was all-powerful and Aurelian far away in exile, and Synesius so represented Typhos and Osiris at the close of book 1. But the myth guarantees that Typhos will be overthrown in turn. Synesius's representation transfers this expectation to Caesarius.
But a further element beyond the ultimate victory of good over evil is more important still. In the myth Typhos kills Osiris. Osiris never becomes king of Egypt again. He becomes ruler and judge of the Underworld. It is Osiris's son Horus who finally deposes Typhos and becomes king in his stead. Synesius points to this aspect when he closes book 1, looking forward to "the years yet to come, when Osiris's son Horus would decide to select the wolf rather than the lion as his ally" (115B).
It was because he did not fully appreciate this feature of book 1 that Barnes mistakenly detected "signs of retouching," additions Synesius made to book 1 when he composed book 2.[43] He had in mind anticipatory passages like 112A and 115A–B that prophesy the reversal of barbarian fortunes. That is to say, he assumed that Synesius added them to an original work written when there was no light at all at the end of the tunnel. But why make such an assumption? Without the expulsion, what would have been left for the god to prophesy? Since Synesius did continue the story in a second book, it would have been much simpler for him to place all the new material there.
The explanation is that Synesius wrote book 1 after 12 July 400. Hence the unmistakable allusion to the massacre at 115A: "It was beyond human prediction that a vast armed force . . . would be defeated with no force to resist it." The massacre of the Goths did not immediately end Gaïnas's regime or Aurelian's exile, but it was an encouraging sign. At this moment Synesius had the idea of casting his story in the form of a myth where good triumphs only after many tribulations, and of cutting the tale short in the midst of the tribulations. It is worth emphasizing how pointed, not to say eccentric, it must have seemed to end the story of Osiris with the victory of Typhos: like ending the story of Oedipus with his wedding to Jocasta, or the Gospel story with the Crucifixion. It cannot have been accidental. It was deliberately calculated to make the reader look forward with the philosopher in the tale to Horus's triumph
[43] Barnes 1986a, 95.
in the next generation. The pointers to a brighter future to which Barnes draws attention are not isolated additions, but integral and indeed climactic parts of Synesius's original plan for reformulating the myth of Osiris.
In order to create this optimistic aura, Synesius stopped his Egyptian narrative just before the massacre but alluded under the guise of prophecies to it and its two harbingers, the comet and Gaïnas's request for an Arian church (114D–115B). The natural implication is that since these three "prophecies" have all come true, the further prophecies that the Goths will be driven out and Typhos's regime will last "months, not years" will also be fulfilled. Synesius says as much quite openly in the preface: "Since the events foretold by the gods appeared to be coming to pass, they [the returned exiles] wanted the tale to go on to their better fortunes" (88B).
A more puzzling example is what Synesius himself refers to as "the riddle of the wolf" (88A).[44] Generations of scholars have taken for granted that there is a political solution to the riddle, with the animals standing for real-life equivalents in the world of A.D. 400. The favorite hypothesis is that the lion stands for the Goths and the wolf for the Huns.[45] It is unlikely enough that Synesius, a philosopher from a provincial backwater whose patron was in exile from Constantinople, was in any position to offer a suggestion that might be taken seriously. Moreover, the political interpretations of the riddle offered require chronological or psychological impossibilities: either Synesius devised the riddle after the death of Gaïnas, which he cannot have done, or he wrote with genuine prescience of its surprising agent, or he made a totally uncharacteristic exception to his judgment on barbarian allies at the same moment he warned against any such alliance. In fact the wolf in the riddle is purely literary. It is the form in which, in one variant of the myth, Osiris returned from the dead to aid Horus against Typhos.[46] If anything, the wolf represents a return to Aurelian's policies by some successor Horus. The lion, in Plutarch's version of the myth, is a force too strong in its own right to be entirely trustworthy. Horus says that it "was helpful to one who needed assistance, but a horse routed the enemy in flight and utterly destroyed him" (De Is. et Os. 358C). The difference is that the rider remains in control of the horse and uses it for his own purposes. There is no bridle to guarantee that the lion, once having helped "one who needed assistance," may not turn and maul him next. In this sense,
[44] See the detailed analysis by J. Long (1987).
[45] Originally proposed by Grützmacher (1913, 58) and argued for by Lacombrade (1946).
[46] Diod. 1.88.6.
the lion in Synesius represents not only Gaïnas's Goths, who were currently demonstrating the danger, but any barbarian force used by the Romans. How universally he held this opinion is demonstrated in his account of the massacre in book 2, by his reuse of material from De regno directed against Alaric. For what solution he expects, Synesius looks to a Horus, who will one day overthrow Typhos and restore peace and honor.[47] He is promised by the myth; and he will be, in reality, whoever turns out to fulfill this role.
The political realities of Constantinople demanded that Synesius choose, of all myths representing the struggle between good and evil, one where good does not triumph till the next generation. For his composition, the fact that Osiris's son rather than Osiris himself takes revenge on Typhos is the crucial element of the myth. It might be added that, frequently as this particular myth is used in Greek and early Christian literature, it is normally the struggle between Typhos and Horus that symbolizes the struggle between good and evil. Naturally enough: Horus, not Osiris, defeated Typhos. It is therefore Horus, not Osiris, who symbolizes good triumphant.
This original purpose of Synesius has to some extent been obscured by book 2. At the time he finished book 1, on any reasonable forecast Aurelian was politically as dead as his mythical counterpart. The past few years had seen the ruin of a succession of contenders for power at the Eastern court. Rufinus and Eutropius were killed; others like Tatianus, Timasius, and Abundantius exiled; it made little difference, since the exiles never returned. No one could have foreseen that Aurelian would return, if not to power, at any rate to his estates and past honors. When he did, Synesius was obliged to resurrect Osiris, at the cost of violating the myth and destroying the subtlety of book 1. But although Osiris was back in Thebes, he was not Horus and did not do what only Horus could. Typhos remained in power, however hard Synesius tries to suggest that his regime was tottering and that it was only because of Osiris's "clemency" that he was not condemned and executed. Once more Synesius looks to the future for hope: the gods have allowed Typhos to continue in office so that men will come to appreciate good the
[47] In a letter written to Aurelian at some later date (Ep. 31), Synesius referred to Aurelian's own son as "the new Taurus, the good hope of the Romans." The letter begins with an obvious allusion to the flattery he had already administered to Aurelian in De providentia , and, now complimenting him as a father, he suggests that young Taurus (later a most distinguished man: consul, prefect, and patrician) might prove the very Horus figure he had conjured. But such succession could not literally fit the Roman context. It is a flattering literary allusion, not a serious suggestion.
more after knowing evil (122D). The concluding quotation from Pindar looks forward to "the remaining days as the wisest witnesses."
It was because he did not appreciate this tentative anticipation that Liebeschuetz claimed to find additions to book 2, advancing arguments similar to those of Barnes for book 1. He supposed that 124C was added after Aurelian's eventual restoration to the prefecture in 414.[48] But, quite apart from the fact that Synesius was probably dead by 412, Liebeschuetz misunderstands Synesius's method and purpose.
In the first place, the hypothesis of interpolation implies an earlier version that either lacked this section or originally contained something different. Liebeschuetz was led to his hypothesis by the apparent conflict between references to restoration and excuses for delay within the same chapter (124A–125C). But the alleged reference to restoration in 124C cannot be neatly disengaged from the context of delay. The entire second half of book 2 is devoted to vague talk about Osiris's popularity and Typhos's imminent fall, with copious metaphysical obfuscation concerning holy mysteries not to be revealed. Even its excessive length suggests an overburdened attempt to gloss over and make the best of the unwelcome reality that Aurelian was out of power. If Synesius had wanted to add a reference to Aurelian's eventual restoration, he would have done better to delete or rewrite the whole second half of the book. It is pervaded with increasingly wan excuses, and the book concludes by counseling patience yet again: "It makes no sense for a man who will soon receive his just deserts to distress himself" (129B). One sentence alluding to Aurelian's restoration does nothing to redeem the many pages that continue to imply the opposite.
For all its philosophical trimmings, De providentia is a piece of journalism: not reportage, but a partisan broadside with a very short effective life span. Journalism of the hour cannot be brought up-to-date by inserting one sentence. All Synesius did, when at some later date republishing the two parts together, was to add a brief preface containing two pieces of information for future readers: a clue to the identity of Typhos and Osiris, and the fact that book 1 was originally an independent work.[49]
Güldenpenning claimed that the reference to Aurelian "becoming more glorious as he grew older" (124C) has to imply the passage of a
[48] Liebeschuetz 1983, 40. This suggestion was in fact first made by Sievers (1870, 368) and firmly embraced by Güldenpenning (1885, 219 n. 4).
[49] The clue to the identity of Typhos and Osiris was the name of their father; even this would hardly have sufficed after a few years, especially since Synesius spent the rest of his days far from court.
decade or so.[50] But Synesius's myth cannot be translated so simplistically into contemporary terms. Aurelian was back in Constantinople, but not in office. He must therefore be portrayed as the sage elder statesman, not king, on the excuse that "a king's ears are often deceived" (125C). There is no reference in book 2 to anything that happened later than Aurelian's return. Indeed, with that one exception, the narrative stops with the massacre. It is only too obvious that Synesius wrote book 2 in some haste and embarrassment immediately after Aurelian's return, when Gaïnas was gone but Caesarius was still in power and in other respects there was little more cause for optimism than when he wrote book 1.
III—
Caesarius and the Exculpation of Gaïnas
Synesius's apparently precise sequence of events for Typhos's coup misrepresents the reality of Gaïnas's coup in at least four major respects. On the identifications of Typhos as Caesarius and Osiris as Aurelian that Synesius obviously expected and encouraged his readers to make, his myth (1) ascribes to Caesarius the initiative in Gaïnas's dealings with him, (2) exaggerates and antedates these dealings, (3) suppresses Aurelian's dealings with Gaïnas, and, most important of all, (4) entirely suppresses the role of Eutropius in the political intrigues of the age.
Indeed, if the struggle for power between the two brothers really went back, as Synesius explicitly claims, before Aurelian's appointment to the prefecture in the summer of 399, then the main obstacle to the ambitions of both men was not yet Gaïnas, but Eutropius. It was only on Eutropius's fall that the praetorian prefecture, robbed of much of its power by Eutropius's palace government, became once more the supreme magistracy of state, the "kingship of Egypt" in Synesius's myth. So paradoxically the myth presupposes and yet systematically ignores Eutropius.
It is instructive to observe how skillfully Synesius contrives to gloss over the omission of what might have seemed an unavoidable element in his story, the previous occupant of the throne for which Osiris and Typhos were in such keen competition. In the first part of book 1 he gives a long account of Osiris's preparation for the kingship and the advice of
[50] A man represented elsewhere "in der Blüte seiner Jahre . . . konnte unmöglich schon 'Greis' genannt werden, das pab t nur auf die Jahre 415 und 416" (Güldenpenning 1885, 219 n. 4).
the old king, his father. At 96A Osiris is initiated into the kingship by his father, who then "departed, taking the same road as the gods" (102D), much as in Claudian's description Theodosius ascends to heaven, leaving behind Arcadius and Honorius as twin masters of the Roman world (III Cons. Hon . 162f.). So too Synesius implies that Osiris is succeeding to the throne on his father's death. That does not quite fit the myth, but is perfectly logical:[51] if two princes are competing for a throne, it follows that the previous king, their father, must either have just died or be on the point of death.
Most scholars have assumed that Aurelian's father, Taurus, really did live till 399,[52] by a remarkable coincidence dying just before his son became praetorian prefect. The truth is that he had been dead for more than thirty years, as proved by the inscription on the statue erected to his memory in the forum of Trajan during the joint reign of Valentinian and Valens, that is to say, between 364 and 367.[53] Even if he had still been alive in 399, it would have been almost forty years since he had held the office he was supposed to be passing on to his son: Taurus had held his prefecture in 355–61. Moreover, he had been praetorian prefect of Italy, not the East. And in any case, it was of course the emperor, not Taurus, who appointed his sons to the offices alluded to by Synesius.
The old king is a purely symbolic figure, the incarnation of the best of the Greco-Roman tradition of government. By presenting Osiris as the spiritual heir of such a father's wisdom and experience, Synesius is able to gloss over both the identity of Aurelian's real-life predecessor and the circumstances of the succession. The omission is clearly deliberate,
[51] Succession to the throne is a crucial element of native Egyptian traditions, but it is the succession of each new pharaoh as Horus incarnate to a predecessor who is assimilated to Osiris (Griffiths 1980). How Osiris came to rule is not at issue. Greek versions assimilate Hesiodic theogonies to fill in this vacancy, but still essentially take for granted Osiris's right to rule.
[52] Demougeot 1951, 236f.; PLRE 1.880 ("still alive in the East in the 390s"); von Haehling 1978, 294 ("um 390 lebte er im Osten").
[53] The text as preserved reads: Jadque constantia aeq(ue) probato v.c. Tauro comiti ordinis primi, quaestori sacri palatii, patricia dignitate praef(ecto) praet(orio) per Italiam atq(ue) Africam, D.N. Valentinianus et Valens victores ac triumphatores semper Augusti statuam sub auro quam adprobante amplissimo senatu iamdudum meruerat ad perpetuam laudabilis viri memoriam reddi iusserunt (Ann. Epigr. 1934, 159, omitting the viri ; correctly, R. Paribeni, Not.d.scavi 11 [1933]: 492, pl. xv; and PLRE 1.879). Taurus was exiled by Julian (Amm. Marc. 22.3.4), at which time his statue in the forum of Trajan was presumably removed. It was replaced by Valentinian and Valens, evidently after Taurus's death (Vogler 1979, 128, rightly refers to a "dédicace posthume"). On any hypothesis it is odd that there is no reference to Taurus's consulate of 361. Even if he had been stripped of his consulate by Julian (for which there is no evidence), it was not removed from official lists retroactively (CLRE 257).
because it was in effect to Eutropius's power that Aurelian succeeded— and with the aid of none other than Gaïnas.
At 108B Synesius introduces "the commander of the foreign troops," away on an unsuccessful campaign, evidently Gaïnas's campaign against Tribigild in 399. In his absence Typhos's wife warns Gaïnas's wife that Osiris "is preparing an accusation of treachery against him, charging that it is a prearranged war he is fighting, with the barbarians dividing their troops but sharing a common purpose" (108C). There can be no doubt that it is to the allegation of collusion between Gaïnas and Tribigild that Synesius is alluding. But it is also clear from the detailed narrative of Zosimus that this collusion, or at any rate Gaïnas's refusal to engage Tribigild, took place before Eutropius's fall.[54] Indeed it was by threatening that he could not defeat Tribigild or come to terms with him so long as Eutropius remained in power that Gaïnas prevailed upon Arcadius to depose him. It was now that Aurelian stepped in to fill the vacuum behind Arcadius's throne. It was not Caesarius but Aurelian who was the first beneficiary of Gaïnas's intervention. Furthermore, a specific illustration of collaboration between Aurelian and Gaïnas can be adduced.
However unsuccessful Eutropius's policies were judged to be, the exile in Cyprus to which Arcadius originally condemned him should have been sufficient, especially since he had been guaranteed safe conduct after taking asylum in a church. But according to Zosimus (5.18),
at Gaïnas's urgent pressure upon Arcadius to have the man killed, the emperor's ministers made a mockery of the oath that they had tendered Eutropius when he was dragged out of the church. They recalled him from Cyprus; then, as if they had sworn only that they would not kill him while he was in Constantinople, they dispatched him to Chalcedon and there had him murdered.
Philostorgius names "Aurelian the prefect" as president of the tribunal that condemned Eutropius to death at a suburb called Panteichion, a few miles down the Asiatic coast from Chalcedon.[55] Here is Aurelian playing the unworthy role of Gaïnas's executioner. No doubt the alliance did not last long, but once it had broken down, naturally Synesius preferred not to mention it. Instead he retrojects to the moment of Eutropius's fall the allegation that Typhos had the support of barbarians to succeed to the "throne"; indeed he claims they were his only support.
[54] As established above, chapter 6, section II.
[55] HE 11.6.
Synesius's claim that it was Typhos who took the initiative in his dealings with the barbarian general has never troubled those who believe implicitly in warring pro- and antibarbarian factions. According to Demougeot, "Synesius, dont le témoignage est le plus sûr et quil[*] faut suivre de préférence à tout autre, raconte que linitiative[*] du complot vint de Caesarius et que ce dernier dut recourir à la ruse pour entraîner le chef des mercenaires."[56] But unlike his younger brother Aurelian, Caesarius had already held both the praetorian prefecture and the consulate and had previously served Theodosius with distinction as magister officiorum . He was already a distinguished elder statesman. Even granting that he wanted another spell in the prefecture, he could scarcely improve his position by plotting with a disgruntled barbarian. For if there was a struggle for power between civilians at Arcadius's court, there was a much more serious struggle for power between court and generals. None of the civilian leaders can have wanted to become the puppet of an Eastern Stilicho. Had Caesarius taken the risk of using Gaïnas to regain the prefecture, he could not be sure that Gaïnas would not end up controlling him. After all, he had destroyed all three of Caesarius's predecessors when they refused to meet his demands.[57] Why should Caesarius have thought that he, a civilian like them, could control Gaïnas any longer or better than Rufinus, Eutropius, and Aurelian?
In any case, the details of Synesius's scenario do not fit the circumstances of A.D. 400. According to Synesius, Typhos's wife stirs up the barbarians against Osiris from jealousy of his power. This represents a double displacement of responsibility for the coup: from Gaïnas onto Typhos/Caesarius and thence onto Typhos's wife, whose energy and competence demean her husband. She initiates the plot by telling the barbarian general's wife that Osiris is planning to dismiss her husband and rid the army of barbarians. Her speech, though rhetorically arresting, is far from being a statement of pro- and antibarbarian ideologies, for from context it is clear that she is lying about Osiris and that she is manipulating the unwitting and (Synesius claims) reluctant barbarians in order to win the kingship for her husband. It hardly needs to be said that nothing of the sort happened in 400. Gaïnas was no innocent tricked into rebelling against Aurelian by a civilian. He had already done almost exactly
[56] Demougeot 1951, 249.
[57] It is usually assumed that Gaïnas's role in the killing of Rufinus was undertaken solely in Stilicho's interests (Demougeot 1951, 154–57). It is in fact more than doubtful whether Gaïnas was at any time Stilicho's creature (Cameron 1970a, 146–48), and it seems most likely that before executing his plan he attempted to extract a command from Eutropius.
the same thing the previous year when he forced Arcadius to depose Eutropius in the summer of 399. Nor did any civilian inspire him on either occasion.
Why was Synesius so anxious to exculpate Gaïnas at Caesarius's expense? A recent study by Rita Lizzi has argued that Synesius's thoughts about barbarians "matured" between De regno and De providentia:[ 58] on closer study he came to realize that barbarians were in themselves a neutral force, who could be of service to the state if correctly handled. It is true that book 1 of De providentia is less openly hostile toward barbarians than De regno . But what Lizzi overlooked is that book 2 treats them as savages once more; indeed, the old beggarwoman's speech at 118A–B paraphrases De regno 23A–B and 25C. There is no mystery here—and certainly no change of heart by Synesius. It is all a matter of placing each book in its context.
Synesius wrote De regno a full year before Tribigild's rebellion, when Gaïnas was an unemployed comes rei militaris with no prospects, and Alaric was far away in Illyricum. It was therefore safe for Synesius to say what he thought about barbarians. Book 2 of De providentia he wrote after Gaïnas had withdrawn to Thrace,[59] when the immediate crisis was over and he could once more say what he thought. Book 1, however, he wrote in the immediate aftermath of the massacre of 12 July, when the crisis was still far from over. Gaïnas was encamped outside the city with the greater part of his army intact. He might easily decide on vengeance. It was at this moment that Synesius formulated the plan of his work. Gaïnas was still too much of a threat to be the villain of the piece, so Synesius cast the new prefect Caesarius in that role. Personal motives may have played some part; Caesarius apparently reduced or delayed the tax remissions Synesius had won for Cyrene from Aurelian.[60] In addition, Caesarius had supported Gaïnas's demand for an Arian church inside
[58] Lizzi 1981, 60–62.
[59] 122A refers to embassies sent to Gaïnas when he had already "withdrawn a great distance from Thebes," clearly alluding to Chrysostom's embassy to Thrace (Theod. HE 5.33.1).
[60] 112C, 114B; since Synesius three times elsewhere claims to have secured great benefits for Cyrene (Hymn 1.496f., Ep . 154 fin., De insomniis 148D), it does not look as if Caesarius can have actually revoked Aurelian's concessions. In general it is probably safe to disregard the accusations of wholesale misgovernment Synesius levels against Typhos, partly because they are all traditional clichés, but perhaps more because he had been in office only three months when Synesius wrote. There had hardly been time for him to undo all Aurelian's work, as Synesius claims he did (112B). On the older interpretation and chronology, it could have been argued that Aurelian undid his brother's work in turn when reappointed to the prefecture, but on the new chronology this option is no longer open. Synesius left Constantinople shortly after Aurelian's return in the fall of 400, with Caesarius still prefect. Ep . 129 shows him setting out on another voyage to Constantinople the following year (Garzya's date of 403 presupposes the old departure date of 402), but beingfoiled by unfavorable winds. What the purpose of this visit was and whether it was ever completed we do not know.
the city walls (115B). The circles in which Synesius moved, shocked at what they saw as Caesarius's collaboration with Gaïnas, were no doubt urging Arcadius to appoint someone more loyal to deal with the crisis. In the event, of course, with the able support of Fravitta, Caesarius was able to retrieve the situation triumphantly. But at the time he wrote book 1, Synesius seems to have been convinced that Caesarius would soon fall from power. This is made clear by 114C. In the course of listing the "prophecies" that are supposed to foretell the downfall of Typhos and his barbarians, Synesius adds: "If Typhos himself should remain in his tyrant's palace, even so do not despair of the gods." This is the first time Synesius sounds the note that becomes so insistent in book 2: it is only a matter of time before Typhos falls. In book 2 his conviction seems increasingly to fail, but there is no reason to doubt that when writing book 1 Synesius genuinely believed that Caesarius could not hold on to the prefecture much longer. This conviction is the cornerstone of De providentia .
It explains, for example, why Osiris does not himself take any action against the barbarians. Not only does he not put into effect the antibarbarian measures proclaimed in De regno ; he is apparently taken quite unawares by the barbarian coup. This is the less surprising because, the way Synesius sets the scene in De providentia , it is not the barbarians who are the danger, but Typhos, secretly plotting behind Osiris's back. One of the strangest passages in book 1 is 108C–D, where it is Typhos's wife who accuses Osiris of planning to rid the state of barbarians. After De regno , this is just what we should expect any hero of Synesius to do, ruthlessly and proudly, but in the context we are bound to interpret the accusation as a malicious calumny along with the rest of her self-seeking lies. Indeed there is a clear indication in the text. At 108B Typhos's wife "adds horror on horror" in her attempt to win over the wife of the barbarian general, "announcing supposed secret plots (
illustration from a contemporary writer, Philostorgius:
The solution lies in his anxiety not to offend Gaïnas. There is some truth in the words of Typhos's wife. If Gaïnas had been imprudent enough to walk back into Constantinople in April 400 without a bodyguard, Aurelian almost certainly would have "taken away his command."[63] That is why Gaïnas made his preemptive strike. That much but no more is true of the barbarian coup in Synesius's allegory. The barbarian deposes Osiris to protect himself; the difference is that, according to De providentia , his belief is mistaken. He has been taken in by the lies of Typhos's wife. Synesius was writing in the fear that Gaïnas would reassert his control of Constantinople. If that happened, his allegory purported to show that Gaïnas had acted under a misapprehension. Aurelian had never wished him any harm; that was all the doing of his crafty brother. In De providentia Synesius performs the remarkable feat of writing an account of the coup of 400 that exculpates both Aurelian and Gaïnas; all the blame is put on the shoulders of a man who was not even in office when it took place!
There are other passages that bear out this interpretation. For example, at 111A Synesius claims that Typhos wanted Osiris "put to death as soon and as violently as possible," while the barbarians, "although they believed they had been wronged, were indignant at this and respected his virtue." According to Zosimus (5.18.9), in complete contrast, Gaïnas threatened the hostages with drawn sword before agreeing to settle for exile. The picturesque detail of the sword may be an embellishment to enhance the picture of the ruthless barbarian, but Chrysostom's Homilia cum Saturninus et Aurelianus acti essent in exilium confirms the danger in which the hostages at one stage stood. Yet according to Synesius, the barbarians not only refused to execute Osiris but were "embarrassed" even to exile him and "decided that it should not be exile but retirement (
It should be remembered that it was Gaïnas who had insisted on the execution of Eutropius.[64] It was also Typhos, Synesius alleges, who sug-
[62] HE 11.8 (p. 138.22 Bidez-Winkelmann; cf. pp. 16.25, 52.14). Cf. Hdt. 7.211.3, just cited.
[63] "Osiris has decided to bring him back with all the compulsion and trickery he can manage. As soon as he is away from his troops, he will take away his command and cruelly destroy him and you and your children" (108C).
[64] Zos. 5.18.2. Compare too the evidently ironic frag. 75.6 (= 67.10 Blockley) of Eunapius: "But when Gaïnas had destroyed the enemy (for he waged an extremely vigorous campaign against the eunuch, so very noble a man was he)."
gested that the barbarians sack "Thebes," a suggestion that, once more, the barbarians repudiated (110C). For their leader "said he marched against Osiris not as a volunteer but under a compulsion that Osiris himself had created" (110D).
Coming from Synesius of all people, this astonishing apology for a barbarian unanimously condemned in the rest of our tradition is the clearest possible proof that he was writing while Gaïnas was still alive and still a threat. In book 2, by contrast, written when Gaïnas evidently no longer threatened Constantinople, we are assured that the barbarians were planning "utterly to destroy Egyptian society and govern it in Scythian fashion" (121B). So much for the boldness with which Synesius has so often been credited.
Such audacious manipulation of the facts must appear all the more piquant when it is borne in mind that Synesius had chosen the wrong side. It should now be clear why book 2, written after Gaïnas had withdrawn to Thrace, is such an evasive and unsatisfactory work. Still clearer, Synesius had nothing to gain by adding to his book in later years. So disastrously misconceived a work could not possibly be brought up to date. Before long Gaïnas was killed, and far from being replaced by Aurelian Caesarius went from strength to strength. It was in wholly different circumstances that Aurelian was eventually recalled to the prefecture in 414.
IV—
Anti-Germanism in Action
Racial prejudice is always at least latent in any society with a sizable minority of a different ethnic origin. And the problem is always worse when, as with the Goths in the Roman world, there is a religious, as well as a racial, difference. The question is, Was anti-Germanism a political issue in East Roman society? Simplistic answers have held the field hitherto. With barbarian rebellions on every side, it will surprise no one that there was an antibarbarian party. The question is, Was there a pro -barbarian party?
The thesis of an antibarbarian party committed to "de-Germanize" the army was inferred from Synesius's De regno and De providentia . We have now seen that, read properly, neither work supports the idea. But what of the evidence outside Synesius? Seeck and Demougeot set forth most clearly as an explanation of various events in 399–401 the hypothesis that once in power in 399 Aurelian ruthlessly put antibarbarian policies into effect and purged the army of Gothic soldiers and generals
alike. For example, Seeck claimed that it was because of this new policy that Tribigild broke off negotiations with Arcadius and Gaïnas joined him in open rebellion. It was also because Aurelian dismissed him from the command in Illyricum that Alaric resumed the life of a freebooter and turned to Italy; Aurelian thus bears indirect responsibility for the sack of Rome. Closer inspection will show that these are highly dubious connections. To begin with Gaïnas and Tribigild, did the advent of Aurelian in August 399 really affect the behavior of the two Gothic commanders in Phrygia?
We have already seen reason to doubt whether, as Zosimus alleges, Gaïnas and Tribigild planned their rebellion in advance. Indeed, there was no one rebellion in which both collaborated. First came Tribigild's rebellion, which Gaïnas exploited to his own ends; and nearly a year later the coup of April 400 at Chalcedon, in which Tribigild played no part. A number of intermediate stages may be distinguished. We should not imagine that Gaïnas had the coup at Chalcedon in mind when he left Constantinople on Tribigild's trail the year before. A review of the successive stages suggests less a preconcerted plan than a series of ad hoc responses, each more drastic than the last. Even after Chalcedon, Gaïnas still seems to have had no clear plan of campaign; whence the rapid disintegration of his position.
Gaïnas's first move, not taken till after Leo's defeat, was to tell Arcadius that he was not strong enough to inflict a military defeat on Tribigild and that the only solution was to give way to his demands. Chief among these demands was the surrender of Eutropius, who he claimed was responsible for the entire situation.[65] The best that can be said for Gaïnas's conduct at this stage is that after the destruction of his colleague's army he may have been genuinely dubious about committing his own Gothic troops against Tribigild's Goths.[66] But even so, it was not for Gaïnas in such a situation to tell the emperor to surrender his chief minister to the enemy. Tribigild might in fact have been satisfied with the Roman command he had asked for before rebelling. But at no point do we hear what Tribigild's own terms were; Gaïnas claimed to speak for him. There can be little doubt that Gaïnas was using Tribigild to satisfy both his own ambition and his long-standing grudge against Eutropius.[67]
Arcadius responded by dismissing Eutropius from office and send-
[65] Zos. 5.17.5.
[66] So, for example, Albert 1984, 121f.
[67] According to Zosimus, he felt that he had not been "considered worthy of the appropriate rank" (5.13.1) and was especially disgruntled by the twin honors Eutropius had recently acquired, the consulate and patriciate (5.17.4).
ing him to exile in Cyprus. This was not enough for Gaïnas. He wrote again, insisting that Tribigild would not be satisfied till Eutropius was "got rid of."[68] Once more Arcadius complied; Eutropius was recalled from Cyprus and executed. Again, there seems no reason to doubt this account. Book 2 of Claudian's In Eutropium , together with its long "stop press" preface, confirms that Eutropius's fall did indeed happen in two stages: first, exile to Cyprus and, then, recall and execution.[69]
For the next stage of the story it may be helpful to quote Zosimus 5.18.4 entire:
Although it was obvious to everyone that Gaïnas was now moving toward revolution, he thought he was undetected; having Tribigild under his thumb because of his superiority in power and rank, he made a truce with the emperor and exchanged oaths on behalf of Tribigild. He then returned through Phrygia and Lydia, followed by Tribigild.
Unfortunately Zosimus does not date this encounter, but it must by now have been fairly late in the year. The reference to oaths suggests that Gaïnas and Arcadius met face-to-face; and if Gaïnas concluded a truce on behalf of Tribigild, the implication is that he was present and Tribigild not. It is interesting to note that however "obvious" it may have seemed after the event that Gaïnas and Tribigild were in collusion at the time, Gaïnas himself was still behaving—and being treated by the imperial government—as an independent agent.
Aurelian must now have been in power for at least a month, and there is no suggestion that Gaïnas was reacting drastically to a drastic new antibarbarian policy. On the contrary, he came to Constantinople to announce that the trouble with Tribigild was over. Zosimus shows that in retrospect at least he was thought to be lying, but there seems no reason to doubt that at this stage Gaïnas hoped that with Eutropius finally out of the way both he and Tribigild would receive the rank and honor they deserved. So both Tribigild's revolt and Gaïnas's prevarication antedate Aurelian's prefecture. Not only did Aurelian not provoke the two Goths to revolt. The first month of his prefecture saw a temporary solution to the grievances of both. Not only did he not immediately take a firm antibarbarian stand. He began, as we have seen, by collaborating with Gaïnas, superintending the treacherous execution of Eutropius.
[68] Zos. 5.18.1.
[69] Zosimus also twice mentions Eutropius's asylum, confirmed not only by Claudian but by the homily Chrysostom delivered while the unfortunate eunuch was cowering behind the altar. Zosimus's further claim that Eutropius was executed at Chalcedon so as not to violate the oath given him in church is also borne out in part by Philostorgius's statement that the execution took place at Panteichion, a suburb of Chalcedon (HE 11.6).
The next stage was Gaïnas's return in April, when he demanded the surrender of Aurelian, Saturninus, and John. Unfortunately Zosimus gives no explanation for this demand, but we may surely infer that Aurelian did not keep some agreement made with Gaïnas at the earlier meeting. It is of course open to believers in the nationalist program to infer that it was during this interval that Aurelian finally proclaimed it.[70] Yet is it conceivable that he did so at a moment when all three Eastern magistri militum were Goths and one of them suspected of collusion with a Gothic rebel?[71] Whether or not Aurelian agreed with Synesius about the desirability of a national army, there were obviously serious practical problems in eliminating the current system. Seeck conceded that Aurelian chose an "unhappy moment" to make his announcement, but saw him as a fanatic whose intransigence provoked the rebellions of both Gaïnas and Alaric.
The weakest element in this highly influential characterization of Aurelian is that not even Synesius so portrays him. Osiris is presented throughout as a weak ruler: well-meaning and noble, but culpably weak. When Demougeot claimed that Aurelian "attempted to rid himself of Gaïnas with a prosecution for high treason,"[72] citing the speech of Typhos's wife in De providentia 109A, she entirely ignored the fact that these are the words of a character who is patently represented as lying. Even so, she suppressed the one element in the context that made the suggestion, however false, at any rate not absurd: namely that Osiris's plan was secret . It is only because the secret was betrayed to him by Typhos's wife that Gaïnas anticipated Osiris's move by rebellion. So the only source directly contradicts what it is alleged to prove: that Gaïnas rebelled in response to the new anti-German policy.
Aurelian's crime was not some new policy toward Gaïnas, but precisely his reversion to the old. This explains why Gaïnas reacted in April 400 exactly as he had done in July 399. It was the continuation of Eutropius's policy toward Gaïnas that brought about Aurelian's fall. Zosimus does not tell us Arcadius's side of the truce with Gaïnas, but we may surely infer that Gaïnas was concerned then with the same issues that concerned him after the deposition of Aurelian.
For all the talk in our sources about Gaïnas's outrageous demands,
[70] "Proclaimed" is Seeck's own word ("verkündigte," 1913, 318). At the very least one would expect such a potentially explosive scheme to be put into effect secretly over a long period—which, significantly enough, is just what Synesius's fiction envisages (108D–109A).
[71] The three: Alaric, Fravitta, and Gaïnas.
[72] Demougeot 1951, 246.
the only two that can be identified are, first, an Arian church in Constantinople and, second, the consulate. When arguing against the Arian church, Chrysostom is represented by Theodoret as begging Gaïnas to be satisfied with his wealth and honors, remarking specifically: "You have been deemed worthy of the consular robe" (
Arcadius considered the other request seriously, but after consulting Chrysostom he finally turned it down.[77] It is significant of Gaïnas's aims that even at this stage he accepted refusal and submitted to the indignity of attending a church outside the city walls. Here too Caesarius may have supported Gaïnas's request. In book 2 Synesius accuses Typhos of being "Scythian" in his religion (121B), but in book 1 he employs the tactic (analyzed in section III, above) of transferring responsibility from Gaïnas to Typhos. There he claims that Typhos tried to introduce a false temple into the city, "not in his own person, for fear of the people of Egypt, but through the barbarians" (115B). That is to say, Synesius claims, preposterously enough, that it was really Caesarius who wanted the Arian church, but to protect himself he tried to use Gaïnas to achieve his own ends. It will be noted that in effect Synesius unwittingly concedes that there was no direct evidence to associate Caesarius with the
[73] As quoted by Theodoret (HE 5.32.6).
[74] PLRE I.380.
[75] CLRE , pp. 9–10.
[76] Theodoret evidently did considerable research for the lost five-book Life of Chrysostom summarized by Photius (Cod. 273, p. 507b Bekker); see too above, p. 98, for unique firsthand source material on Chrysostom cited by Theodoret. Baur (1959, 1:xxix–xxx) unnecessarily belittles Theodoret's value for the study of Chrysostom.
[77] Theod. HE 5.32.
incident at all. It is almost inconceivable that a man who had enjoyed a distinguished career under Theodosius should actually have been an Arian. But a man whose father had been an Arian might at any rate be tolerant.[78] A law addressed to Caesarius as prefect and probably to be assigned to December 395 allows the extremist Arian sect of the Eunomians the right to make wills,[79] a right taken away from them by his intolerant predecessor Rufinus earlier the same year.[80] Not too much should be made of this one law since another, dated to April 396, orders that those responsible for the "crime of the Eunomians," whose "madness induced them to such false doctrine, shall be tracked down and expelled as exiles from the municipalities."[81] But it was enough for a bigot like Synesius to tar Caesarius with the same brush as his "darling Goths" (122B).
Seeck's influential thesis of a "victory of anti-Germanism" following Gaïnas's defeat is entirely dependent on the assumption (now shown to be false) that Aurelian was restored to the prefecture on Gaïnas's fall and carried out the rest of his reforms over a period of several years (401–5). If there were independent evidence that the prefect who collaborated with the Goths had been disgraced, if the Gothic general had soon been executed and the nationalist patriot restored to office, there might have been some basis for such speculations. But Caesarius was no more of a collaborator than Aurelian, and he was not disgraced or deposed; Aurelian was no more of a patriot than Caesarius, and he was not restored; and Fravitta remained the ranking Roman magister militum for several years. Indeed Fravitta's consulate in 401 must have been construed, as it was surely intended, as an eirenicon to the Gothic community of the Eastern empire, a sign that not all Goths were rebels and traitors: those who served the state could still rise to the pinnacle of fame.
This brings us to Alaric. We have seen that up till the sudden outbreak of Tribigild's rebellion, the main barbarian threat within the Eastern empire had been Alaric. He had plundered the Balkans for three
[78] Taurus was Constantius's representative at the Council of Ariminum: Seeck 1894, 444; PLRE I.879–80; T. D. Barnes, Entretiens . . . Fondation Hardt 34 (1989): 313–14.
[79] Cod. Theod. 16.5.27; 24 June (MSS), when Rufinus was still alive; 25 December (Seeck 1919, 287).
[80] Cod. Theod. 16.5.23.
[81] Cod. Theod. 16.5.31; cf. 32; 28, addressed to an Aurelian who was proconsul of Asia on 3 September 395, lays down sanctions for all heretics "who deviate in even a minor point of doctrine from the path of the Catholic faith." If both date and office are correct, this cannot be the famous Aurelian, who could not have been proconsul of Asia after already holding the higher post of city prefect.
years, laid siege to Constantinople, and indirectly been the ruin of both Rufinus and Eutropius. It looks as though it was at one time feared that he might assist Tribigild, since Eutropius's original plan was to send Leo against Tribigild himself, at the time in Phrygia, and Gaïnas "to meet the enemy in Thrace and the Hellespont if he found them disturbing those regions."[82] Tribigild might have marched down to the Hellespont,[83] but it is also the place Alaric would have crossed if he had wished to do so.[84]
Unfortunately, though so loquacious about Tribigild, Zosimus tells us nothing at all about the activities of Alaric between 396 and 402. And though he gives an account of Stilicho's expedition of 396 that forms a useful corrective to Claudian,[85] he dated it to 395, having confused it with the earlier expedition during the lifetime of Rufinus.[86] It is from Claudian, not Zosimus, that we learn the key to Synesius's antibarbarian tirade in De regno : Eutropius's appointment of Alaric to the Roman command in Illyricum.[87] This presumably kept him quiet for a while, though in 398 the "skirmishings" mentioned by Synesius seemed to some to augur a renewal of hostilities. But they never came. The real puzzle of 400–401 is that Alaric suddenly decided to leave Illyricum and invade Italy, arriving there on 18 November 401.[88]
On the traditional interpretation, this has never seemed much of a problem. According to Seeck,[89] in the summer of 399 the new prefect Aurelian at once dismissed Alaric from his command and withdrew his subsidies, whereupon Alaric began to devastate Illyricum again. Within a year he had bled Illyricum white and so turned to the West. Yet is it credible that Aurelian would have wantonly provoked aggression from Alaric when he was already faced with the problem of Gaïnas and Tribigild in Asia? For Seeck he was a fanatic, but we have shown that there is
[82] Zos. 5.14.1.
[83] At 5.14.3 Zosimus alleges that Gaïnas had planned to meet up with Tribigild at the Hellespont, so that they could take over all of Asia; but that is surely no more than an inference—and a very improbable inference—from Gaïnas's location there.
[84] So Ridley 1982 ad loc., but there seems no justification for his further inference that Gaïnas was already "distrusted by Eutropius." After all, serious as Gaïnas's subsequent collusion with Tribigild proved to be, collusion with Alaric would have been even more disastrous. If Eutropius had the slightest reason to suspect Gaïnas of harboring treasonable designs, he would never have appointed him in the first place.
[85] Cameron 1970a, 86f., 169f.
[86] Cameron (1970a, 474–77) shows that "the telescoping of the two campaigns was already present in Eunapius." See too forthcoming article of Cameron on date 396.
[87] According to Zosimus, it was a Roman command that Alaric really wanted, "being angry [in 395/96] that he did not command an army but had only the barbarians Theodosius had given him when he helped to put down the tyrant Eugenius" (5.5.4).
[88] Chron.Min. 1.299.
[89] Seeck 1913, 318 and 328.
no support for this view even in Synesius. As for the renewed devastation of Illyricum, the only source cited surely refers to Alaric's return to the East in 403, after his failure to break into Italy.[90] Faced with the withdrawal of his subsidies, we should expect Alaric to have brought direct pressure to bear on Arcadius, desperately vulnerable during most of 399–400 because of Tribigild's rebellion and Gaïnas's coup. The invasion of Italy was not just a campaign: the Goths took their women and children and booty with them.[91] There must have been some major reason for such a mass emigration. Mazzarino remarked casually that "the anti-barbarian party purged Illyricum of barbarians."[92] But it is difficult to believe the Eastern government capable of such a feat in 400–401, and it is certainly remarkable that such a victory should not have been recorded. At a time when imperial victories were not as common or decisive as in happier days, more was made of every success.[93] The local chronicle of Constantinople that is partially preserved in the Paschal Chronicle and the Chronicle of Marcellinus routinely records the announcement of imperial victories at Constantinople,[94] and we might expect to have found some mention of a victory over Alaric here if not in the erratic Zosimus.
The end of 400 saw the arrival of a new factor on the scene, one whose significance has perhaps been underestimated: Uldin the Hun. A week or two after the massacre Gaïnas withdrew to Thrace, but, finding it hard to provision his troops there,[95] he decided to make for the Chersonnese and cross the Hellespont into Asia. Not having any ships, he improvised rafts from freshly cut trees. But Fravitta had ships with bronze prows and easily cut Gaïnas's force to pieces. Gaïnas himself, who had remained in the Chersonnese, withdrew into Thrace again, where Fravitta (as we have seen) did not pursue him. Gaïnas first killed his remaining Roman troops, whose loyalty he suspected,[96] and then
[90] Collectio Avellana 38.
[91] Claudian describes how the Romans captured Gothic women, and the plunder of Greece after Pollentia: Bell.Get. 84–85, 615–26; VI Cons.Hon. 130, 243–45, 282, 297.
[92] Mazzarino 1942, 68.
[93] See McCormick 1986, 59–60 ("There appears to be a correlation between severe and widely perceived blows to imperial prestige and intensification in the rhythm of imperial victory celebrations"). But his statistical argument is misleading, since he makes insufficient allowance for the different nature of the sources from the late fourth century on: the chronicles list victory announcements that might not have been mentioned at all in earlier sources.
[94] CLRE 54–56; McCormick 1986, 56f.
[95] "He found the cities defended by walls and guarded by their magistrates and inhabitants: as a result of previous attacks, they were not only prepared for war, but ready to fight with all their strength" (Zos. 5.19.6). It is optimistic to attribute this, with Demougeot (1951, 259), to the implementation of Aurelian's national program.
[96] Zos. 5.21.6.
"crossed the Danube with his barbarians, intending to return to his native land and there spend the rest of his days." This is most unlikely. We may compare the case of the Frankish magister militum Silvanus falsely accused of treason in 355: after considering rejoining the Franks beyond the frontier he decided that they would "either kill him or betray him for a bribe."[97] A Romanized barbarian could not return to the old life. Gaïnas had the additional disadvantage of not coming from a Gothic royal house. He had deserted from the Gothic army after Adrianople and enlisted in the Roman army as a common soldier.[98] Paradoxically, here was a barbarian leader whose authority was limited to the Roman world.
Did he then actually cross the Danube, and was it there that he was killed by Uldin? So it has usually been supposed in modern times. But as Gibbon observed long ago, "the narrative of Zosimus . . . must be corrected by the testimony of Socrates and Sozomen, that he was killed in Thrace; and by the precise and authentic dates of the . . . Paschal Chronicle."[99] According to the Paschal Chronicle , the battle with Fravitta took place on 23 December 400, and Gaïnas's pickled head arrived at Constantinople on 3 January 401. The interval is far too short for the several encounters Gaïnas is said to have had with Uldin and the trip from wherever the final battle took place to Constantinople,[100] though one of the dates may be wrong.[101] But Socrates and Sozomen both place Gaïnas's death in Thrace, even if they do make the blunder of saying that he was killed by a Roman army. And according to Philostorgius, perhaps present in Constantinople at the time and writing ca. 430, it was after retiring to "upper Thrace" that "some Huns soon after attacked and killed him."[102] Even if Gaïnas briefly crossed the Danube to evade pursuit and recruit reinforcements, he will surely have crossed back again when Uldin began his pursuit. It is likely enough that the final battle took place in Thrace.
So it may be that 401 saw a major Hunnic incursion into Thrace.
[97] Amm. Marc. 15.5.16.
[98] Soz. HE 8.4.1.
[99] Decline and Fall, ed. Bury (1909), 3:394 n. 39.
[100] This is true even if we accept Maenchen-Helfen's ingenious suggestion (1973, 59) that the final battle took place near Novae, "the place on the Danube nearest the capital, connected with it by a first-rate road."
[101] It is possible that 23 December could be the date of the battle with Uldin rather than Fravitta, which might have fallen much earlier (December would be rather late in the year for an experienced commander to try crossing the Hellespont on rafts). The usually more reliable Marcellinus places Gaïnas's death in February. For both texts, Mommsen, Chron. Min. II.66.
[102] HE 11.8, p. 139.17–18 Bidez-Winkelmann.
Zosimus goes on to describe how Uldin received a reward for Gaïnas's head and concluded a treaty with Arcadius. No terms are mentioned, but the Huns presumably offered to fight as federates on the usual terms. The obvious guess is that they offered to fight the Goths in Illyricum. Since it was to escape the Huns that the Goths had made their fateful crossing of the Danube in 376, this cannot have been welcome news to Alaric. If it was now that the Eastern government withdrew Alaric's subsidies, he might well have felt that this was an opportune moment to move on to fresh pastures.
A contemporary document makes an interesting link between the fall of Gaïnas and Alaric's departure for the West. It is a homily falsely ascribed to Chrysostom that was delivered at the tomb of St. Thomas on his festal day, that is to say, as Tillemont saw, at Edessa on 22 August.[103] The homily is an attack on Arianism, which the preacher associated particularly with barbarians, that is to say, Goths. St. Thomas is represented as reproaching Arius for taking holy plunder from the hands of a barbarian and for despoiling the apostles. "Fall with your own allies," he cries. "I have cast down one of your citadels; I have freed Thrace of your tyranny."[104] Thomas promises that he will soon chase him from the rest of the world; soon he will free the West from Arius's folly. "Blessed Thomas, as you have freed Thrace, so may you free the West. I shall hear that the bandit has been destroyed like the tyrant before him." The only explanation that fits all these clues, as again Tillemont saw, is that the tyrant is Gaïnas, the bandit Alaric,[105] and the year 402.[106] The preacher clearly seems to think of Alaric as having been driven from Thrace rather than leaving it of his own accord.
His talk of freeing the West as Thrace had been freed is particularly hard to reconcile with the popular modern notion that it was with the blessing of the Eastern government that Alaric was sent off to be a thorn
[103] PG 59.497–500, with Montfaucon's introduction; cf. too J. A. De Aldama, Repertorium pseudochrysostomianum (Paris 1965), 193, no. 517. The sarcophagus of Thomas had been brought to Edessa on 22 August 394: H. Delehaye, Les engines du culte des martyrs , 2d ed. (Brussels 1933), 59 and 212–13.
[104] PG 59.500.
[105] Tillemont knew that Gaïnas is not said to have stolen any holy vessels, but made the unhappy conjecture that Arcadius had been forced to give him vessels from the Holy Apostles in Constantinople. This is most improbable. It must have been Alaric's men who robbed churches during his three years of plundering. It is clear even from Orosius's apologetic (Hist. adv. pag. 7.39) that the Goths plundered the churches of Rome in 410.
[106] Alaric is still thought of as a threat to the West, suggesting that the preacher was speaking before he returned to Illyricum in 403, driven out of Italy by Stilicho. Demougeot (1951, 269 n. 194) strangely thought that the bandit was Stilicho (who was not even an Arian) and that the homily reflected the propaganda of the "antibarbarians" that Stilicho would soon fall from power.
in Stilicho's side.[107] There is in fact a decisive argument against this hypothesis. In 401 Stilicho recognized the Eastern consul for the first time since 398, and in 402 Honorius publicly signaled the restoration of concord between East and West by taking the consulate jointly with his brother. It is not till the end of 403 that relations began to deteriorate again, with a series of formal protests from Honorius that Arcadius ignored and the nonrecognition of the Eastern consul for 404. It would be impossible to explain this temporary thaw in the cold war if there had been the slightest suspicion that Alaric had been paid or encouraged to invade the West.[108]
Whatever the reasons for Alaric's departure, the Eastern government surely took the credit for it. And the head of the Eastern government at this time was undoubtedly Caesarius. There was no need of antibarbarian fanatics to take what steps were required to save the East from Gaïnas and Alaric. Indeed it may have been a providential intervention by the Huns that helped speed Alaric on his way, as it had previously cut off Gaïnas's escape. There is no reason to believe that Caesarius was any more of a friend to the Goths than his brother was. He was just a shrewder politician. It is not that there were no antibarbarians. But since there were no pro-barbarians, it makes little sense to talk of an antibarbarian party. The East may well have cut down so far as possible on the use of federates in the years to come, but by enrolling barbarians individually in the regular army rather than building up a genuinely national army. And there is certainly no evidence that there was a barbarian purge. The massacre was a response to the tyranny of Gaïnas rather than the expression of a deep hatred of all Goths. Many Goths, both Catholic and Arian, survived the massacre. There was a thriving community of Arian Goths in fifth-century Constantinople, whose theological differences with a splinter group known as the Psathyrians were reconciled by the consul Plinta in 419.[109]
V—
Conclusions
Predictably enough, the antibarbarian straitjacket was never a comfortable fit. Inevitably, the founding members of the coalition (apart from Synesius himself) had to be Aurelian, John, Saturninus, and Eu-
[107] E.g., Demougeot 1951, 269; Grumel 1951, 39; Stein 1959, 248 ("Le gouvernement oriental . . . avait ainsi le double avantage de se débarasser de Barbares dangereux et du même coup de tenir, grâce à eux, Stilicon en échec").
[108] As rightly emphasized by W. N. Bayless, Classical Journal 72 (1976): 65–67.
[109] E. A. Thompson, The Visigoths in the Time of Ulfila (Oxford 1966), 135–38.
doxia. But Saturninus, as Holum has already remarked, "had earlier been an architect of the Theodosian Gothic settlement in 382, whence the barbarian 'problem' took its roots."[110] And Eudoxia was the daughter of a Frankish general.[111] Of course, it is always possible to argue that Saturninus came to repent the Theodosian policies he had been obliged to execute, and that to gloss over her own German origins Eudoxia did her best to be more Roman than the Romans. But there is no evidence at all that even John and Aurelian were antibarbarians, much less fanatics.[112] Not one of Arcadius's ministers pursued a consistently antibarbarian policy.
Hardly more convincing than this "anti-German coalition" is Dagron's thesis that it was a "senatorial party" that took over from Eutropius: "Ce qui est certain, en revanche, cest[*] que toutes les analyses politiques de cette époque tiennent le plus grand compte de lexistence[*] dun[*] sénat à Constantinople et de la personnalité de ce sénat."[113] But it is not easy at any time to identify the Constantinopolitan senate as a separate body with esprit de corps and policies of its own distinct from those of the court.[114] And of Aurelian's two identifiable associates, John was clearly a court favorite, in 400 perhaps not yet even a member of the senate. And Saturninus had been a professional soldier all his active life (from ca. 350 to at least 383). As perhaps the oldest surviving Eastern consul, he could no doubt be described as a prominent senator, but his continuing importance was undoubtedly due to his abilities as a courtier. He is described by Zosimus as a flatterer, "adept at pandering to the whims and schemes of imperial favorites."[115] So far was he from being a staunch senatorial champion that he even curried favor with Eutropius, betraying his former colleague Timasius.[116] It was not on ideological grounds that Gaïnas singled out Saturninus as one of his victims in April 400, but because he saw him as a creature of Eutropius, a turncoat against his class. As for Caesarius and Aurelian, while they were certainly prominent members of the Byzantine senate, at the same time they were men who, like their father before them, had spent all their lives in the imperial service. It is not obvious that such men represented senatorial traditions as distinct from those of the imperial bureaucracy.
[110] Holum 1982, 68 n. 80.
[111] It was for this reason that Güldenpenning (1885, 97) put her in the pro-German camp; more hesitantly, Bury 1889, 79.
[112] "Nationalistes farouches," in Demougeot's favorite formulation (1951, 265).
[113] Dagron 1974, 204.
[114] Cameron 1978, 276.
[115] Zos. 5.9.3.
[116] Zos. 5.9.3.
Moreover, they were rivals. And their rivalry stemmed from the fact that each had attached himself to a different court favorite.
The even less plausible recent appeal to a party of "traditionalists" or "Hellenists" has been discussed in chapter 3. There is no identifiable correlation between classical tastes in literature and philosophy and either religious or political views.
Byzantine politics are not to be explained in terms of these anachronistic party labels. All political rivalry was overshadowed and conditioned by the paramount need to gain and maintain influence with the emperor. We must accept the obvious explanation of the infighting at court ca. 400: the inevitable struggle for power behind the throne of a weak emperor. In 399 the issue was not so much hostility toward Goths in general as the very real threat from one particular Goth: Gaïnas. The ascendancy at which Gaïnas aimed would have threatened the independence of all the civilian rivals for the power behind Arcadius's throne. Eutropius made use of Gaïnas to rid himself of Rufinus and was then unwise enough not to fulfill his part of the bargain. The ambitious Eudoxia wanted Eutropius out of the palace for her own reasons. There is no need to fabricate an ideological basis for Aurelian's ambition—and certainly no evidence on which to do so. It was the combined efforts of Eudoxia, Gaïnas, and Aurelian, each with different aims, that brought Eutropius down. There is no new anti-Germanism evident in Aurelian's policies once he was prefect. Indeed in the only areas where we might have looked for it, that is to say, in his dealings with Gaïnas and Stilicho, he pursued exactly the same policies as Eutropius. He even repeated Eutropius's fatal mistake of underestimating Gaïnas's ambition and determination.
We do not need to credit Caesarius, waiting in the wings, with a pro-barbarian ideology. If he supported Gaïnas's perfectly reasonable demand for an Arian church, he showed himself more of a traditionalist than his brother. Synesius describes Typhos as "overturning ancestral laws" in championing the barbarian's demand (115B), the Hellene unexpectedly here showing himself a prophet of the new intolerance of his age. In fact it was less than twenty years since Theodosius had forbidden Arians to worship within the city walls.[117] If Caesarius negotiated with Gaïnas in the immediate aftermath of the massacre, as Synesius alleges, this was hardly "to invite him back, as though no irreparable damage had been done" (121B), but to implore him not to take the terrible revenge all must have been fearing. It was, after all, the Goths who were
[117] Cod.Theod. 16.5.6 (381), 12 (383), 13 (384); cf. Socr. HE 5.20.
the injured party, with thousands of civilians butchered in the streets, and women and children burned alive in their own church. So long as it was prudent, Caesarius negotiated; once the situation permitted, he declared Gaïnas a public enemy and saw to the rapid mobilization of Fravitta's army.[118] Despite Synesius's repeated allegations of plotting and collusion with the Goths, within a year of taking office Caesarius had achieved a more satisfactory solution of the Gothic problem than any of his predecessors.
The reluctance of Eutropius and Aurelian after him to give Gaïnas what he wanted is surely to be interpreted less in terms of racial prejudice than as just one aspect of a (largely successful) attempt to keep the generals subordinate to the civil authorities. Eutropius, a palace eunuch with no independent power base, was naturally anxious to protect himself against all rivals. But he was most concerned about the military. It is no coincidence that there had been no more military consuls since Theodosius's death. Eutropius had actually treated Gaïnas better than he had treated such Roman generals as Abundantius and Timasius. Gaïnas he refused to promote, but Abundantius and Timasius he accused of treason and exiled. Yet however personal his motives, perhaps the main achievement of Eutropius's much-maligned administration was, at the critical moment, to have prevented a military takeover of the government by an Eastern Stilicho, whether barbarian or Roman.
Caesarius relaxed the unwise confrontational policy toward Stilicho pursued by his three predecessors without making any real concessions. His chief general Fravitta was a man of barbarian birth, but thoroughly Romanized, loyal, and above all successful. It was thanks to Eutropius, Caesarius, and their longer-lived successor Anthemius that power at the Eastern court continued to remain firmly in civilian hands, despite the incapacity of Arcadius and the minority of Theodosius II. This is a phenomenon of greater importance and more lasting significance than the transient, if terrifying, wave of anti-Germanism that racked Constantinople on 12 July 400.
[118] It was the praetorian prefect who supervised conscription and acted as quartermaster general for the army: R. Grosse, Römische Militärgeschichte (Berlin 1920), 158–60.