The flourishing of late-antique studies in the last half-century has coincided with the rise of “world history” as an area of academic research. To an extent, some overlap has occurred, particularly with Sasanian Persia being considered alongside the late Roman Empire as constituting an essential component in what we think of in terms of the “shape” of late antiquity. Yet it is still the case that many approaches to late antiquity are bound up with conventional western narratives of historical progress, as defined in Jack Goody's The Theft of History (2006). Indeed, the debate about whether late antiquity was an age of dynamic transformation (as argued by Peter Brown and his disciples) or one of catastrophic disruption (as asserted, most recently, by Bryan Ward-Perkins) can be regarded as representing two different faces of an essentially evolutionary interpretation of western historical development. This article argues, however, that we can challenge such conventional narrative frameworks by taking a world historical perspective on late antiquity. It shows, first, that our interpretation of late antiquity depends on sources that themselves are representative of myriad local perspectives. Secondly, it argues that since Gibbon's time these sources have been made to serve an essentially western construct of and debate about history. The final section considers how taking a more global perspective allows us to challenge conventional approaches to and narratives of late antiquity.
This article offers an unashamedly personal set of challenges to conventional approaches to the study of late antiquity. In particular, it recommends that some of the impasses that currently bedevil debates in the discipline might be overcome by adopting a more world-historical approach to the subject. By that I mean not only seeing the history of late antiquity in a wider geographical perspective, but also a viewpoint that adopts an ethical stance that challenges the current paradigms within which late antiquity is debated: as I argue below, conventional accounts of the period focus their narratives around the experiences of the Roman Empire and, therefore, articulate an essentially western and Eurocentric interpretation of historical development. Of course, many specialists in the field are already making significant advances away from this western-dominated narrative; nevertheless, it strikes me as a worthwhile exercise to draw the strands of the debate together and to offer pointers to possible future directions.
Given the scope of the undertaking implicit in this recommendation, the enquiry presented here can only offer a brief overview of the themes and issues I want to contest: the examples cited below could be multiplied exponentially,1 and I aim to investigate many of the issues in more detail in the future. In other words, what is presented here is only the beginning of a larger project. I should also clarify that the outcomes of what I suggest here might take many forms. I have written this article mainly with an eye to research agendas; but there is no reason why some of the perspectives recommended here could not also be imported into a classroom setting, where they would surely provoke interesting discussions. But to begin with, and in order to demonstrate how ingrained the conventional approaches I wish to challenge have become, I present a narrative that will seem, at first, wholly familiar.
A victory had been won and the ruler wanted to celebrate it. The barbarians, true to form, had been duplicitous and had broken the treaty. Now a great hosting of them (Goths, Germans, and others) had invaded the empire, but they were no match for the empire's forces and had been utterly defeated. Many of the enemy had been slain in bloody vengeance for their treacherous behaviour in starting the war. More importantly, many of their leading men had been captured; best of all, their king had been captured alive. He would make a fine ornament for the ruler's victory celebrations at his capital, a living example of the ruler's indomitable power, a figure to be humiliated and put on public display. Such a great victory also deserved a permanent commemoration in text and image, so reliefs and inscriptions were set up showing the ruler in all his might lording it over his abject, cowering foe.
Such images are familiar to us from Roman imperial and late-antique monuments, like the reliefs from the now lost triumphal monument of Marcus Aurelius or from the extant arch of Septimius Severus in Rome, or those that decorate the obelisk base of Theodosius I in the hippodrome in Constantinople.2 But the set of victories and commemorations I have been describing so far do not come from that familiar context. Rather, the triumphant ruler was Shapur I, shahanshah of Sasanian Persia; the defeated barbarians were the Romans; and the captive king the emperor Valerian in 260 C.E. For humiliating display, I have in mind the tradition that Shapur used Valerian as a stool when mounting his horse or getting into his carriage, and that later, when the emperor died, his corpse was flayed and his skin tanned to provide a more permanent trophy.3 As for the epigraphic and visual commemorations, I mean the so-called Res Gestae Divi Saporis, the great trilingual inscription recording Shapur's victories, and the rock reliefs at Naqsh-i Rustam and Bishapur, showing his triumph over not only Valerian, but Gordian III and Philip the Arab too.4 If, however, anyone steeped in Roman imperial or late-antique history had assumed that my earlier description alluded to the victories of a Marcus, Severus, or Theodosius, their misapprehension would be wholly understandable, for they have been conditioned to think of a world centred on Rome, Constantinople, and the Mediterranean.
That this should be the case attests to the profound influence on modern perceptions of a supposedly “normative” world view underwritten by traditional, classical geographical divisions of the world into a civilized centre and barbaric periphery.5 In this traditional schema, Persians, like other non-Romans, inhabit the margins of the map. Such a world-view underpins classical and classicising historiography, and can be found, for instance, in the fourth-century Latin historian Ammianus Marcellinus’ celebrated description of the Huns as being “abnormally savage” and living “beyond the Maeotic sea, near the frozen ocean.” He goes on to catalogue their lifestyle in a form that reads like a negative checklist of the accoutrements of civilization as it was viewed by the Greeks and Romans: the Huns lack every marker of civilized life, from fire, to cities, to politics, and are only acknowledged begrudgingly as human. They are, therefore, doubly remote from civilization, in terms of both their geographical distance and their lack of cultural attainments.6
But the achievements of Shapur I, and his epigraphic and monumental commemoration of them, remind us that other perspectives, not centred on the classical Mediterranean, are possible. It is these perspectives that I want to explore here, and I organise my reflections as follows. First I offer a survey of late-antique perspectives on the world, showing their variety and complexity, and how they demonstrate that the Mediterranean-centred perspective of classical and classicising historiography is not the only view possible. Next, I discuss how the traditional shape of late antiquity has been made to fit into a customary western, and essentially Eurocentric, view of history – and how this might be regarded as deeply problematic. The final part of the article considers how that traditional view might be challenged by adopting an approach that is more sensitive both to the multiple local perspectives outlined in the first part of the discussion and to global contexts; this in turn shows how, by advocating a more world-historical perspective on events, we can challenge traditional narratives of the period, and see events in a new light.7
A VARIETY OF PERSPECTIVES
The ancients knew, of course, that there was a wider word beyond that in which they lived, and, moreover, that events in remote places could have profound implications for them. For example, in 357, negotiations took place on the frontier in Mesopotamia between Strategius Musonianus, the praetorian prefect of Oriens (speaking for the Roman emperor Constantius II), and Tamsapor, representative of the Sasanian shah Shapur II. Our source, Ammianus, reports that intelligence had reached Musonianus that Shapur was distracted by troubles “on the remotest frontiers of his realm, [and] was with difficulty and with great bloodshed of his troops driving back hostile tribesmen.” The prefect therefore instructed Tamsapor that
he should by letter advise the king finally to make peace with the Roman emperor, in order that by so doing he might be secure on his whole western frontier and could rush upon his persistent enemies [i.e. in central Asia]. Tamsapor consented and relying on this information, reported to the king that Constantius, being involved in very serious wars, entreated and begged for peace. But while these communications were being sent to the Chionitae and Euseni, in whose territories Sapor [Shapur] was passing the winter, a long time elapsed.8
A similar story, told this time from a Persian perspective, albeit preserved in the narrative of the Greek historian Menander Protector, recounts a speech delivered by the Persian envoy Andigan to the representatives of the emperor Tiberius II (578–82). On this occasion, the Persians enjoyed the strategic advantage, since Tiberius’ armies were fighting on several frontiers at once, but the Persians were at war only with Rome. In Andigan's view, this made the negotiation of peace a desirable outcome for the Romans; the alternative offered a grim prospect:
You are at war with many peoples, we with you alone: thus the necessity of the treaty. Just as the Romans would be certain to prevail if they were fighting either a number of tribes or the Persian kingdom alone, so we shall certainly conquer since we have a dispute with no one but the Romans and since we are committed to only one war.9
These two accounts, from two centuries apart, essentially tell the same story: the rivalry between Rome and Persia in the Middle East was determined not solely by events along the specific frontier that separated them, but by events across a wider geographical arena. Distractions of the Persians deep in central Asia, or of the Romans in Europe, could seriously compromise the ability of either power to wage war on the other.
It is clear, moreover, that this level of mutual understanding led Rome and Persia to develop some sort of begrudging respect for each other in late antiquity. One manifestation of this is a tantalising story related by Procopius and Agathias that when the emperor Arcadius died in 408, he entrusted his baby son Theodosius II to the protection of the Persian king Yazdegerd I.10 But the relationship reached its apogee in the later sixth century, when it is summarised in the words that Theophylact Simocatta places in the mouth of the shah Khusro II (590–628) in a letter to the Roman emperor Maurice (582–602):
God effected that the whole world should be illuminated from the very beginning by two eyes, namely by the most powerful kingdom of the Romans and by the most prudent sceptre of the Persian state. For by these greatest powers, the disobedient and bellicose tribes are winnowed and man's course is continually regulated and guided.11
This was not mere rhetoric: on occasion we can witness the Roman and Persian empires cooperating, such as in Maurice's efforts to establish Khusro II on his throne, or in collaborative Romano-Persian efforts to defend the Caucasus from raids mounded by central Asian nomads.12
While consideration of Romano-Persian interaction certainly extends our vision beyond a Mediterranean world centred on Rome and Constantinople, it risks simply replacing one vision of the world with another which this time has two centres—the two eyes spoken of by Theophylact. What we need to bear in mind is that beyond this simple bi-polar vision there are any number of multiple perspectives. For the world between Rome and Persia, we are singularly fortunate in having a wide range of contemporary accounts, composed in languages such as Armenian and Syriac, as well as later Arabic histories which incorporate earlier sources (including Persian ones).13 These accounts enable us to access the experiences of peoples living between the empires, and demonstrate that when Rome and Persia came to war, their conflicts could provide opportunities that communities living along the frontier could exploit to their own benefit. An episode narrated in the Chronicle of Pseudo-Joshua the Stylite shows how, during the Anastasian War of 502–506, the distraction of a major interregional conflict could open up opportunities for local Arab populations:
The Roman Tayyaye [Arabs] also crossed the Tigris in front of them, plundering capturing, and destroying all they could find in Persian territory. Since I know that you carefully examine everything, your holiness will well understand that this war was the cause of much enrichment for the Tayyaye [Arabs] on both sides, and that they did as they pleased in both empires.14
Such local perspectives are not limited to the frontier in the Middle East, but occur also in the West. The Spanish chronicler Hydatius of Lemica offers a striking example in his account of the events of the year 410 C.E. Now 410 is a significant watershed year in traditional narratives of the fall of Roman power in the West, since it saw the sack of Rome by Alaric's Goths. In his narrative, Hydatius does not fail to mention the sack, but what is remarkable is that his descriptions of contemporaneous happenings in his Spanish homeland overshadow this event. He devotes only a few short sentences to Rome, mentioning the sack and the capture of the imperial princess Galla Placidia. By contrast, he gives a more fulsome treatment to affairs in Spain, citing barbarian invasions of the Iberian peninsula and the disease that followed in the wake of these upheavals. He then provides a long discursive passage which analyses those events in apocalyptic terms. All told, the Spanish passage is almost three times as long as the entry on Alaric at Rome. 15 It is important to bear in mind, of course, that Hydatius’ narrative is coloured by his own particular perspective on events, in particular his apocalyptic outlook and sense that he was, in some way, writing a history of the end of the world.16
The narrative of Hydatius, by highlighting events in Spain over those at Rome, presents an unfamiliar perspective in which events like the sack of Rome—central to conventional grand narratives of late antiquity—are passed over relatively quickly in favour of detailed accounts of more local history. If, from a viewpoint centred on the Mediterranean, Spain might seem marginal, the effect of using a non-metropolitan source like Hydatius is to offer a different perspective opposed to that master narrative. The perspective offered by Pseudo-Joshua similarly presents a view from the periphery, but considered alongside a rich tradition of Syriac literature, makes even more emphatic than Hydatius the point that such “peripheries” can be considered as “centers” in their own right: it is possible to write a history of the frontier in Syria that sees it as central, and the Roman and Persian empires as marginal; a broadly similar case can be achieved for Armenia.17 These myriad local perspectives therefore present an opportunity to challenge the traditional account of late antiquity with its Mediterranean-focussed overarching grand narrative. It is to this theme that I now turn.
(i) The End of the World: Sources and Narrative Choices
Why have events at Rome in 410 traditionally commanded our attention more than those in Spain? In part, it reflects the perspectives of some contemporaries, who saw the Gothic sack in particularly portentous terms. A vivid account offered by Jerome recounts his emotional turmoil when he received news that “the city that had taken the whole world was itself taken.”18 For contemporaries, in the Latin West at any rate,19 such events seemed to signal a sense in which an established order was coming to an end. Another contemporary, the chronicler Prosper of Aquitaine, vividly attests to this view in his account of the capture of Carthage by the Vandals in 439. Having described the violence of the Vandals towards both Africa's citizens and its churches, Prosper commented: “Carthage suffered this captivity in the 585th year after it had become Roman.”20 In the Eastern Empire, where notions that Roman rule had been ordained by God persisted (an idea first formulated in the early fourth century by Eusebius of Caesarea in the heady atmosphere of the Christian emperor Constantine's reign), this conception of imperial endings took a little longer to become established.21 Moreover, when it came to be expressed by the chronicler Marcellinus Comes in the sixth century, it was motivated by different concerns from those that had motivated Jerome and Prosper. For Marcellinus, the deposition of the “last” western emperor Romulus Augustulus in 476 could be presented in symbolic terms:
With this Augustulus perished the western empire of the Roman people, which the first Augustus, Octavian, began to rule in the seven hundred and ninth year from the foundation of the city. This occurred in the five hundred and twenty-second year of the kingdom of the departed emperors, with Gothic kings thereafter holding Rome.22
As Brian Croke has shown, this presentation was calibrated to the concerns of an eastern establishment contemplating reconquest of the lost western provinces, and for whom it made sense to manufacture 476 as a turning point and to rewrite half a century's worth of diplomatic relations between Constantinople and Italy in order to provide a pretext for Justinian's western wars.23 By contrast, contemporary responses in 476 to the deposition of Romulus had been more sanguine, even at Constantinople. A fragment of Malchus of Philadelpia's history reports that when a delegation from the senate of Rome came to the eastern emperor Zeno to report the deposition of Romulus and claim that one emperor would be enough for them, Zeno dismissed them with the response that they still had an emperor: Julius Nepos, who had been sent to Italy in 473, deposed a year later, but was still claiming to rule “the West” from Dalmatia.24 So far as Zeno was concerned, the western empire was not dead yet.
I have spent some time rehearsing these familiar details because they have exercised a defining influence on modern perceptions of the grand narrative of late antiquity. Dates such as 410, 439, and 476—along with others, such as the Gothic immigration in 376 or the battle of Hadrianople in 378—have dominated our construction of the age. A restatement offered by Peter Heather makes clear not only what he sees as historically plausible, but also implies what he finds rhetorically pleasing:
[T]here is no serious historian who thinks that the western Empire fell entirely because of internal problems, or entirely because of exogenous shock. The emphasis of this book has been primarily on the latter, because in my view the growth of Hunnic power in Europe has been misunderstood and, with it, the intimate link between the arrival of the Huns [in c. 376] and the deposition of Romulus Augustulus [the last western emperor, in 476].25
Not everyone has agreed: in a pungent rejoinder, Michael Kulikowski has asserted that “Heather's idée fixe—that the Huns were responsible for the fall of the Roman empire and the end of the ancient world—is simple, elegant, and wrong.”26
The very debate about the significance of events such as those in 376, 378, 410, 476, and numerous others is instructive. In one sense it is depressingly so, in that it suggests that the major chronological nodes of the debate have moved on little since Edward Gibbon penned his History of the Decline and Fall of the Roman Empire more than two hundred years ago. I return to Gibbon presently, but for the moment let me also suggest that it is instructive of the point I made earlier in connection with my slightly devious presentation of the victories of Shapur I—namely, that such views reflect the dominance of a select number of sources, whose perspectives reflect what are regarded as ‘normative’, metropolitan views on our conception of the grand narrative of the period. Jerome, lamenting the capture of the city that had captured the world, was, memorably, more Ciceronian than Christian;27 Prosper, who passed judgement on the end of Roman Carthage, may have been from Aquitaine, but was an enthusiastic supporter of the Roman papacy and its claims to primacy;28 and Marcellinus, as we saw, is representative of coalescing opinion at sixth-century Constantinople. In other words, by prioritising their perspectives, and the sort of narrative they imply, we are relegating to the margins the sort of view on events presented by authors like Hydatius and Pseudo-Joshua. I would contend that the marginalisation of such views is deeply problematic, since it implies complicity in a particular view of history.29 Let me return now to Gibbon.
(ii) Decline, Fall, and Transformation
The importance of the narrative constructed by Gibbon in his Decline and Fall and the reactions against it in, especially, the last half-century, are so well known as to be dispensed with relatively quickly. In his significant recharacterisation of the age, The World of Late Antiquity (1971), Peter Brown remarked: “It is only too easy to write about the Late Antique world as if it were merely a melancholy tale of ‘Decline and Fall’.” Such a picture, he argued, could not account for the largely positive developments of the era: “why Europe became Christian and why the Near East became Muslim,” for example. In his view, “Looking at the Late Antique world, we are caught between the regretful contemplation of ancient ruins and the excited acclamation of new growth.”30 Under Brown's influence, the balance now shifted towards a consideration of cultural transformation and an emphasis on continuities over disruption;31 moreover, the geographical horizons were now extended to encompass the Near East and Islam as well as Europe and Christendom.
In the decades that followed, it is clear that excited acclamation overtook regretful contemplation. In 1996, for example, Glen Bowersock remarked that “it is probably fair to say that no responsible historian of the ancient or medieval world would want to address or acknowledge the fall of Rome as either fact or paradigm”; and that “[t]he fall of Rome is no longer needed, and like the writing on a faded papyrus, it no longer speaks to us.”32 More recently, Edward James remarked with trenchant optimism that “‘Decline’ has been banned from the … vocabulary” of scholars working in the field.33 Such assertions might be taken as indicating that Gibbon's views had been consigned to the historiographical scrap heap, and that any restatement of them is at best old-fashioned (perhaps even irresponsible). But there are good reasons for regarding that optimism as having been misplaced. In fact, the last fifteen years have seen renewed interest in decline and fall as hermeneutical tools with which to interpret late antiquity.
Particular expressions of disquiet began to be expressed by scholars working on late-antique urbanism. Thus in 2001, the spirit of Gibbon was unmistakably present in the title of Wolf Liebeschuetz's Decline and Fall of the Roman City.34 A more forceful statement of misgivings about the whole post-Brown late antiquity project was to be found in the editors’ prefatory remarks to one of the volumes dealing with cities that emerged from the European Science Foundation's Transformation of the Roman World project. What is also striking is that their conclusion is clearly influenced by particular geographical and chronological preconceptions:
From the perspective of any research into urbanism that starts with the Roman period, it is very difficult to view developments in the sixth and seventh centuries, except for the late antique christianization of the city, as part of some neutral (or even positive) “transformation”. The changes that occurred in urban life generally look more like the dissolution of a sophisticated and impressive experiment in how to order society – an experiment developed by the Greeks and Romans and centred on the Mediterranean.35
Indeed, it is one of the authors of this observation, Bryan Ward-Perkins, who has contributed the most sustained critique of the post-Brownian vision of late antiquity as an age of neutral or positive transformation. In his 2005 study, The Fall of Rome and the End of Civilization, he asserts that late antiquity needs to be considered anew as a period of disruption and collapse.36
Works like that of Ward-Perkins and Peter Heather (mentioned earlier) seem to suggest a resurgence of an essentially Gibbonian view of history, and indeed they have been characterised as regressive.37 At one level, this might mean little more than that the pendulum of historical opinion is swinging back against Brown and his disciples.38 But I think there is something rather more than that at stake. In short, it seems to me insistence on decline and fall reflects a particularly prejudiced perspective on the end of the ancient world and emergence of the middle ages.39 We need to return to Gibbon to see what I mean.
To Gibbon, and despite all his efforts to encompass Byzantium and Islam in his narrative,40 the demise of the Roman empire was first and foremost an event significant in western European history. That significance is implicit throughout much of his narrative, but in his “General considerations on the decline of the empire in the west” that concluded volume 3 of his history, Gibbon made this European dimension explicit by considering how a similar chain of events might affect the Europe of his own day.41 When the final volume of Decline and Fall was published in 1788, Gibbon was probably content that his image of a civilization brought low by the irruption of external factors—alien peoples, the barbarians, and an alien creed, Christianity—was a reasoned one. But almost immediately events unfolded that were to cause Gibbon to consider other factors that he had omitted from consideration. As France was thrown into the convulsive horrors of revolution, Gibbon was forced to the realisation that some central tenets of his History were seriously mistaken. In particular, the events in France made him wonder if he had seriously underestimated the significance of internal revolts and civil wars in the undermining of the empire. In papers prepared for an unrealised seventh volume of Decline and Fall he noted: “Should I not have deduced the decline of the Empire from the Civil Wars, that ensued after the fall of Nero or even from the tyranny which succeeded the reign of Augustus? Alas! I should: but of what avail is this tardy knowledge? Where error is irretrievable, repentance is useless.”42
(iii) Geographies, Eurocentricism, and the Theft of History
Gibbon's misgivings, it seems to me, have been too little appreciated in recent years. Moreover, the geographical focus on the Mediterranean and Europe that dominates much work (but not all: see below) on late antiquity and the early middle ages, and which has had an enduring impact on the very structures we impose on the past is deeply problematic (for example, through periodization, in which “the fall of Rome” is a hinge moment). It belongs to a style of analysing the past that has been characterised by the anthropologist Jack Goody as amounting to a European theft of history, where the experiences of Europe provide the central framework within which the totality of history is interpreted, and where the central concern of historians is largely focused on Europe (particularly north-western Europe).43 Goody's characterisation of how western history “works” can be traced in various ways in modern histories of the late and post-Roman world, and have, I think, been signally important in the recent enthusiasm for decline and fall.
Consider, by way of example, two standard overviews of the late-antique/early medieval period published in related, if distinct, series: volume 14 of the second edition of the Cambridge Ancient History (CAH 14), covering the period 425–600, and the first volume New Cambridge Medieval History (NCMH 1), covering circa 500 to circa 700.44 Their chronological overlap is not matched, however, by precise geographical concurrence. CAH 14 is altogether more concerned with the world of the eastern Mediterranean and territories such as Armenia and Persia; NCMH 1 does not neglect these (early Byzantium and Islam receive detailed treatment), but even so the geographical focus has unmistakably moved west and north to be centred on Europe, and includes territories like Ireland that lay outside the purview of CAH 14. In both cases, the geographical focus of these compendia is shaped by their narrative concerns: the CAH is clearly shaped by the world that has gone before, and reflects a world view that encompasses, and saw as central, the Near East and the southern shores of the Mediterranean; the NCMH, by contrast, is clearly shaped by the world that is to come: a world focused on the medieval civilization of, above all, Europe.45 The medieval histories of Near Eastern territories and North Africa belong elsewhere.46
In this respect, we can see one influence of Goody's western theft of history, in that as we move from antiquity to the middle ages we also move inexorably westwards (and northwards) to an emphatically European history. That is not to say that such a vision has not been challenged. Peter Sarris’ recent overview of the period 500–700, published somewhat incongruously in the “Oxford History of Medieval Europe,” is altogether more wide-ranging in its geographical vision than many comparable histories (including NCMH 1).47 As Christopher Kelly pointed out in his review of Sarris, “It is easy to forget that this is hard-fought territory, disputed by fractious experts and partitioned between different university departments.”48 Here we gain a glimpse of how fraught a task is any effort to challenge, for late antiquity, the western theft of history identified by Goody: anyone seeking to issue such a challenge would need to overcome deep-rooted academic and institutional divisions.
(iv) Narratives, Progress, and Analogy
Another feature of this western construct of history noted by Goody is its insistence on progress as an essentially upward curve from “the dawn of civilization” through to modern industrial capitalism. In any such narrative, the end of the ancient world seems to represent a disruption of that upward curve, and the recent restatements of decline and fall have noted this. For instance, in 1996, the Italian scholar Aldo Schiavone, in an account of the lack of a seamless progression from the ancient to modern world, noted that the study of late antiquity (particular the Brownian late antiquity with its emphasis on transformation) was a valuable enterprise: it presented the historian with “an entirely new universe … in which simplistic and teleological explanations have no part.” At the same time, however, he suggested that it “tends to overshadow an essential point” by “downplaying the disruptive and catastrophic aspects of the changeover” between antiquity and the Middle Ages, a process that he described in a memorable phrase as a “historical thrombosis.”49 In Schiavone's view, the roots of Rome's fall were to be sought in the structure of its economy, in particular the reliance on slave labour for productivity. Such an economy, he argued, had no inherent growth mechanism; as a result, it could only expand so far and no further, and its ultimate collapse was in many respects inevitable.50
Ward-Perkins’ The Fall of Rome and the End of Civilization is, I think, even more embroiled in this western narrative of progress and its disruption, and it is precisely this disruption to progress that allows him to present the late-antique/early-medieval transition in such bleak terms. He makes frequent use of analogy to draw attention to the complexity of the Roman system in the period before the “Germanic invasions.” Such analogies contribute to a picture of Roman imperial civilization that risks distortion by stressing its inherent modernity. We are repeatedly told that the use of some commodity, such as metals, roof tiles, pottery, bricks, and even literacy, was not to be paralleled in Europe until the late-medieval or modern period.51 In other words, the fall of Rome is presented as a disruption of that essentially progressive model of western history.52
(v) The barbarians and the wider world
In addition to presenting the fall of Rome as a disruption of progress, this decline and fall model subscribes to the western theft of history in a third way: how it presents the barbarian peoples who transformed the political landscape of the Roman world by relegating them to the position of outsiders who crashed in towards the centre, wreaking destruction in their wake. This neatly reflects the perspective of the late antique Greek and Latin literary sources upon which historians customarily rely. For them, the world beyond the empire's frontiers was not well little known and barely understood beyond standard geographical and ethnographic topoi. When Ammianus mentioned the Chionitae and Euseni in the context of Musonianus’ negotiations with Tamsapor in 357, he gives little indication of their precise geographical location, apart from their remoteness. Occasional glimpses of distant lands and peoples can be had, for example, from Priscus of Panium's account of his visit to the court of Attila in central Europe in the fifth century, or Menander Protector's references to east Roman contacts with the Turkish tribes around the Caspian Sea in the sixth century.53 In spite of this, the essential picture presented by the sources is one that, as anthropologist Thomas J. Barfield has remarked of the relationship between China and the nomads of the central Asian steppe, consists “of seemingly random events presented chronologically, with one obscure tribe following another.”54 Yet, to comprehend the events of late antiquity, it is necessary to take a wider view. After all, the barbarian invasions represent (depending on one's viewpoint) an overlapping or collision of the histories of the Mediterranean region and the wider world, and I do not think it is possible to comprehend one without consideration of the other.55 In other words, the traditional narrative plot of barbarian invasions and the late-antique transition demands consideration of that wider world. In much of what follows, I use “Eurasia” as a shorthand for that broader perspective, but it should be remembered that the geographical horizons went further still: for instance, interactions between Rome and Persia also encompassed parts of Africa (Axum in Ethiopia) and Arabia (Himyar in Yemen) to the south, as well as Armenia and the Caucasus to the north.
LATE ANTIQUITY AND WORLD HISTORY
In this last section, I want to examine the potential of considering late antiquity against such a wider world-historical background. To begin with, I need to revisit the question of what we mean by the term “late antiquity,” show how subjective it is, and demonstrate how it has often excluded a world-historical perspective. Only then can we posit a wider, global perspective and explore the avenues for enquiry that it might open up.
(i) What (When, and Where) is Late Antiquity?
Although the current debate on late antiquity takes its impetus from, above all, the publications of Peter Brown and his disciples (and detractors), the origins of the concept go back further, particularly to a cluster of art historians, notably Alois Riegl and Josef Strzygowski, working in the years around 1900.56 It was Riegl who first deployed Spätantike in his Spätrömische Kunstindustrie (Vienna, 1901) to describe the arts and crafts of the transitional period between antiquity and the middle ages, and to see a metamorphosis of classical forms occurring at this point. As for Strzygowski, his Orient oder Rom (Leipzig, 1901), by looking to Iran as a source of artistic inspiration for new forms, perhaps hinted at an early attempt to understand this transitional period in world-historical terms; sadly, his venture was compromised by his adherence to notions of Aryan and Nordic supremacy that mirrored his political allegiances.57
Subsequent developments in the study of late antiquity have largely been concerned with an overarching western narrative, squarely focused on a story that has Mediterranean origins and a largely European outcome. Thus the term antiquité tardive came into vogue with French scholars interested in the development of the Roman Catholic tradition.58 In Anglophone scholarship, the Eurocentric tendency can be seen among both those scholars who approach late antiquity by moving forward from classical, Graeco-Roman antiquity into the fourth century and beyond;59 and those who take their starting point with the developed institutions of medieval Christendom and then move backwards in search of their origins.60
In both of these cases, late antiquity is defined in relation to other periods as part of a particular narrative plot—as either a sequel during which the heritage of classical antiquity was transformed into something else (mainly conceived of in terms of western culture), or as a prequel during which the forms of medieval culture coalesced from the raw materials of an earlier age. Consequently it is unsurprising that there has been some difficulty in arriving at an agreed definition of the chronological parameters of late antiquity, beyond a vague consensus that it falls somewhere in the middle of the first millennium. Even Brown's programmatic World of Late Antiquity is vague on this, in spite of his advertised limits of Marcus Aurelius and Muhammad.61 One result of this has been a resistance to what Andrea Giardina has diagnosed as an ‘elephantiasis’ of late antiquity, in which the label is applied to a period that stretches ever further backwards into the Roman imperial period and forwards into the middle ages.62 Objections to this “long” late antiquity include those who are interested in different narrative plots, and who feel that these are obscured by the focus on the late antique. Averil Cameron has recently raised precisely this sort of concern from the perspective of Byzantinists.63 But this highlights an important issue: in large measure, chronological definitions (which traditionally have derived from political history, and so are difficult to delineate precisely for processes of cultural change) will always involve an element of subjectivity, depending on the narrative being constructed and the actors involved.64
Related to this is another point, and that relates not to when late antiquity was, but where it was. Geographical considerations will often depend on the particular emphases of a narrative. For instance, Garth Fowden proposed over twenty years ago that the stage for Romano-Persian conflicts in late antiquity, and their ideological (particularly religious) consequences, encompassed a vast “mountain arena” stretching from Axum in Ethiopia to the mountain ranges of Afghanistan, from Yemen and Arabia to the Caucasus and the Caspian; his recent argument in favour of considering the first millennium as a chronological unit has restated, refined, and elaborated this geographical context, now cast as the “Eurasian hinge.”65 But this geographical focus depends on the story he wishes to tell, which in his recent restatement clusters around the development and maturation of monotheistic religious traditions in Judaism, Christianity, and Islam; and it has its costs, as can be seen in his characterisation of Latin Christendom in western Europe as backward and underdeveloped.66 This means that cultural definitions of particular late-antique narratives can result in the exclusion from them of particular regions. At one level, this is understandable, since a particular argument requires a particular focus. But it results in a situation pithily described by Guy Halsall as one in which “Late Antiquity was just something that happened to other people,” for example by excluding from the narrative the inhabitants of Ireland or barbarian Europe.67 Just as much as periodization and narrative plots, the definitions of “central” regions (be they Brogiolo and Ward-Perkins’ “Graeco-Roman Mediterranean” or Fowden's “Eurasian hinge”) involve a series of subjective choices.
There are various ways in which other regions can be integrated into a wider world history of late antiquity. One approach is to acknowledge interconnections between the various parts of the world in the late antique period (however we define it). By this I mean more than just traceable connections (to which I return below), but also ones that determined the development of cultures and societies. Halsall argues that the development of societies abutting Rome's European frontiers is such that “Germanic-speaking barbaricum was, perhaps paradoxically, more integrally a part of the Roman Empire than many of the imperial provinces.”68 But examining states bordering on the Empire through a Roman lens can bring its own risks, for example by highlighting features of non-Roman societies that resemble those found within the Empire to the exclusion of others. An important paper by Michael Morony on Sasanian Iran might seem to have done precisely that, by considering the extent to which the Persian empire resembled the Roman across a range of categories, such as political structure, religious profile, and military factors. But his intent was vigorous advocacy of the inclusion of the Sasanian polity as an integral part of the late antique world, and was conceived of in response to overly Romanocentric notions of what constitutes late antiquity.69 While this might seem to subsume the Sasanians into an explicitly western view of history, it has its uses beyond advocacy, since highlighting similarities between cultures and societies can allow for meaningful comparisons to be made.70 But as I hope to show now, we can go rather further than that.
(ii) Global Perspectives on Late Antiquity
I want to close by suggesting that a broader, global vista might allow us to put the late-antique experience of the Roman world in a more meaningful perspective than the traditional Mediterranean focus allows, and that such a perspective might allow us to ask useful questions about what we do when we study late antiquity. I should state very clearly that I am not pretending that no-one has attempted such an approach. An instructive example is provided by Giusto Traina's study of the year 428, which takes its reader on a grand tour of the world, from Persia around the provinces of the Roman world (and, in some cases, the territories of a post-imperial west) and back again.71 The approach adopted in Traina's book bears comparison with John Wills’ study of the year 1688, traditionally associated in Anglocentric historiography with the “Glorious Revolution.” By expanding his horizons, Wills was able to encompass other, less familiar narratives, presenting familiar events (at least to European historians) in the light of unfamiliar juxtapositions.72 Such a comparison, however, also highlights the very different resources upon which scholars of late antiquity and early modern history depend. While Wills’ account is truly global and encompasses regions such as the Americas and sub-Saharan Africa, Traina's is still very much the world of Khusro II's “two eyes of the world” fixed on the Roman and Sasanian empires and adjacent territories.73
We might extend our geographical horizons much further than this. That connections existed across a wide sweep of Eurasia, from the Atlantic in the West to the Pacific in the East (and north to central Europe and the Baltic, or south into Africa), is hardly in doubt.74 A few examples will need to stand here for a much wider body of work. Touraj Daryaee has shown that the tentacles of Persian influence, as witnessed by evidence for trade, extended not only West towards the Roman empire and south to Arabia, Yemen, and Ethiopia, but also East to the Indian Ocean and the archipelagos of south-east Asia.75 Christopher Beckwith has meticulously traced the reach and dynamics of trade and other interactions across central Asia between Europe and T'ang China in the early middle ages.76 Meanwhile the work of the “Late Iron Age and ‘Roman’ Ireland” project has revealed the vigour of Irish contacts with Roman Britain that goes beyond trade to encompass conflict and cultural change.77
Given that such connections are traceable, there are various ways in which they might be incorporated into our understanding of late antiquity. It should be stated immediately that we should not exaggerate them and make them into a totalising discourse that obscures the nuances of local diversity.78 Nor should we forget that many such contacts were often indirect and involved several stages of interaction between intermediary groups.79 On the whole, we are presented with a series of overlapping and intersecting zones of interactions and perceptions. This could happen even across comparatively small distances. A neat example emerges from Priscus of Panium's account of his embassy to the court of Attila on behalf of the eastern emperor Theodosius II in 449. Priscus mentions the presence at the Hunnic court also of plenipotentiaries sent from the western emperor independently of the diplomatic initiatives emanating from Constantinople.80 On a much larger scale, we have seen that Persians clearly had some view of the western reaches of the Roman Empire, just as Romans had some inkling of the Central Asian territories ruled by Persia. But both Persians and Romans had areas of interaction in other directions too, and for reasons unrelated to the ebb and flow of war and peace between them. In the case of the Persians, we can see that their interactions could reach as far east as China: al-Tabari, the early historian of Islam, relates in his History of Kings and Prophets that as Sasanian Iran was being overrun by the Muslim Arabs, its last shah, Yazdegerd III, appealed to the emperor of China for help.81 The world was certainly interconnected, but those interconnections could be haphazard and piecemeal.
We need, moreover, to allow that individual interactions, while they might be linked to wider geopolitical contacts between polities, often had a logic of their own. Consider, for instance, Sasanian contacts with Arabia. Most often, these have been interpreted in a narrative of Romano-Persian rivalry that sees Roman and Persian influences in a string of territories from the Caucasus to Himyar in Yemen and Axum in Ethiopia as part of a grand strategy of imperial rivalry. In an earlier part of his History, al-Tabari recounts how when Shapur II (one of Rome's most redoubtable enemies in the fourth century) came to the throne as a child in 309, his minority placed the Persian realm in a condition of weakness that was exploited by its neighbours. Arabs crossed the Persian Gulf and commandeered lands in southern Iran, and it was only when Shapur reached his majority that Persian power reasserted itself. This involved not only driving the Arabs back across the Gulf, but also a punitive military expedition deep into the Arabian peninsula that was, at least as al-Tabari describes it, violent, bloody, and disruptive.82 Individual details of al-Tabari's account might be questioned, but there is enough archaeological evidence to suggest that Sasanian influence on Arabia, particularly in the East and South, was indeed profound.83 More importantly for my purposes here, this is a campaign that leaves no trace whatsoever in the Greek and Latin sources of the period, and as such opens our eyes to a history in which Rome plays no part.84
What utility does this have for the historian of late antiquity? Above all, we should see it as offering opportunities to interrogate our conventional accounts of the late-antique and early medieval centuries. In her preface to the English translation of Traina's 428 AD, Averil Cameron has drawn attention to “the unexpected juxtapositions of history” and to extraordinary instances of “simultaneity.”85 A similar point was made by Lester K. Little in his presidential address to the Medieval Academy of America in 2004. He noted that significant comparanda and juxtapositions can prompt us to ask questions that might never have occurred to us before, and that a global, comparative account of history might enable us to get away from what he identified as an essentially teleological and western-dominated narrative of history.86 Above all, while individual political histories might be traced across significant geographical distances, perhaps the greatest utility of adapting a world/global perspective on late antiquity would be in terms of thematic comparisons.
Such comparanda and juxtapositions are not hard to find in late antiquity. Consider for example the activities of Christians like Patrick in Ireland, Ulfilas in Gothia, or Nestorian missionaries (not to mention Manichaean ones) travelling through the Near East to Central Asia and China. Each of them was involved in various ways with religious institutions within the Roman oikoumene, but they also extended their faiths outwards in various ways, and with far reaching consequences. Moreover, the last mentioned groups, Nestorians and Manichaeans, intersected with a world rich with other religious traditions, such as Buddhism.87 Comparing their various stories might highlight telling parallels, but equally, and just as importantly, it might pull into focus distinctive features of each narrative. To religious comparanda we might add other developments such as contemporaneous (both with each other, and with the religious histories outlined above) linguistic evolutions: the development of ogham script in Ireland depended on the influence of Latin;88 among the Goths in the Danubian territories, the missionary Ulfilas developed a Gothic alphabet, based on Greek, so that he could write down Christian scripture in their language;89 and in Bactria in central Asia, Greek letters also formed the basis for a written form of a local language used between the fourth and eighth centuries.90
Moreover, comparisons of political history across a broader geographical sweep allows us to see the experiences of the late Roman empire as less unique than the Eurocentric insistence of the “fall of Rome” paradigm might suggest. Barbarian invasions also afflicted the Persians, spilling through the Caspian gates into the Mesopotamian heartland of the empire. Further east, Eurasian nomads even passed through the mountains of the Hindu Kush and, for a time, established a kingdom in northern India.91 What has customarily been considered in predominantly western terms emerges instead, from a more global perspective, to be one of a number of concurrent (and possibly interconnected) processes affecting a wide sweep of territory and a great number of polities.
By looking across this much broader geographical canvas, we might consider other factors that afflicted not only the Roman world, but adjacent territories too. Recent work on the palaeoenvironmental history of Eurasia is beginning to suggest that factors completely separate from invasion and war were having an impact on the transformation of society.92 It is relatively uncontroversial to state that the end of the ancient world witnessed a contraction of economic activity. In large measure, the dismemberment of the Roman Empire played a key role in this process, since there was simply no longer a need for a sophisticated market economy whose chief raison d’être had been the maintenance of Rome's fiscal and military apparatus. But recent research is beginning to suggest some types of economic decline were to be found not only within the Empire, but also further afield. For instance, studies of pollen samples from archaeological excavations in Poland suggest a contraction of cereal production there in the fifth century that is directly comparable to situations in territories that were (or recently had been) in the Roman Empire. That strongly suggests that war and upheaval were not the only factors underlying change, and new archaeological material is beginning to hint at other possible answers. Analysis of alluvial deposits in the Rhône valley suggests a period of considerable climatic upheaval between the fourth century and the seventh: rainfall increased dramatically, and it seems likely that agricultural productivity will have declined drastically over an extended period. Similar evidence is beginning to be yielded by study of the advances and retreats of Alpine glaciers. Between the late-fifth century and until the seventh, a number of these seem to have expanded considerably.93 Taken together with the evidence of alluvial deposits in the Rhône valley, and numerous references in historical documentation for the period, not least the widespread references to the ‘years without summer’ in the mid-sixth century (presumably the consequence of some sort of volcanic event),94 the seemingly inescapable conclusion is that the end of antiquity and the dawn of the middle ages was accompanied by much colder and wetter climatic conditions across much of Eurasia. Given the vulnerability of pre-modern agricultures to small shifts in temperature, there is reason to think that these changed climatic conditions could have had a devastating impact.95
To such ecological factors we might also add disease. Just as ancient agriculture was fragile, so too was ancient demography, meaning that major epidemics were likely to have significant, far-reaching consequences. And the end of the ancient world was accompanied by just such an epidemic: the so-called Justinianic plague, first observed at Pelusium in the Nile Delta in 541, and which recurred throughout the Mediterranean world, the Near East, and northern Europe for some two centuries. The plague had long been known from grim accounts in sources ranging from Arabia to Ireland that stressed almost incomprehensible levels of suffering and mortality. Historians have justifiably been wary of taking such accounts at face value, but recent work on this pandemic is beginning to suggest that its impact was indeed profound. Comparison of the late ancient accounts of the disease with modern studies of epidemiology would appear to confirm that the epidemic is our first attested incidence of bubonic plague. Its impact can be seen across a range of phenomena, including contractions of economy and settlement, and demographic decline. Such considerations suggest that plague may have been a more decisive factor in the fortunes of the eastern Roman Empire than the Arab invasions of its territory in the seventh century.96 Moreover, plague was not the only disease to afflict late ancient populations. Throughout antiquity and the middle ages, malaria was a major threat to the health of Mediterranean populations; when it occurred concurrently with plague, the impact could be devastating.97 In combination, these ecological and demographic factors likely had a profound impact on society, not just in the Roman and post-Roman Mediterranean, but also more widely across Eurasia.98 Plainly, the barbarian invasions do not explain everything, nor can events in the Roman world be understood in isolation.
CONCLUSIONS: RISKS AND PROSPECTS
This article has argued that any evaluation of the transformations of late antiquity needs to resist temptations to see the experience of the West in late antiquity as somehow exceptional or unique. As I have suggested here, a variety of phenomena traditionally seen as central to the late Roman narrative—from barbarian invasion, to religious expansion—and some that are only beginning to be explored—such as the history of disease and environmental change—can be observed across a wide area of Eurasia between the third and the eighth century. If we simply sit tight on a Mediterranean-focused perspective, we might as well admit that our horizons have expanded little since Socrates told his interlocutors in the Phaedo that they sat around their sea like frogs or ants around a pond.
This is not to say that there are not difficulties associated with the sort of global late antiquity project outlined here. One obvious problem—and one that I freely admit I am in no position to surmount—is that such an undertaking would require a mastery of the histories of a wide range of cultures, not to mention their multiplicity of languages;99 as such, a project of the sort recommended here might best be undertaken as a collaborative effort.100 Other difficulties in adopting a global perspective are that more local histories might get neglected,101 although I hope I have made it sufficiently clear that my vision here of world history is made up of myriad local narratives, and that we should not lose sight of one in pursuit of the other. Yet, for all these challenges, I still think that the task of seeing late antiquity in world historical perspective is one worth pursuing. By considering that wider perspective, we might compensate in two ways for what Jack Goody identified as the western theft of history: one that sees the experience of late-antiquity freed from the shackles of narratives of progress or disruption, and one that sees the experience of the Mediterranean and adjacent regions as part of a wider history of humanity, sometimes shared, sometimes remarkably distinctive.102