It is all too easy to be beguiled by the careers of men like Dexter and the evidence they provide for the apparent unity of the vast territory that Rome held under its sway. Of course, the empire was held together by networks other than political ones. Interregional trade was a powerful contributor to creating a single Roman world (Greene 1986; Ward-Perkins 2005: 87-104). But by concentrating above all on the unity of the empire, we miss the essential diversity of the various elements of which it was composed. Indeed, the apt remark has been made that ‘‘the Roman Empire is too often seen as a whole, too seldom as a collection of provinces’’ (Wickham 2005: 3). Such factors often recede from view when emphasis is placed on the shared political and cultural experiences of the empire’s social elite, based mostly in cities (Garnsey and Saller 1987: 178-95).
Examination of the individual territories that made up the Roman Empire presents a picture of astonishing diversity. That was in part the product of geography. The empire comprised various ecological zones, ranging from the areas of arid semi-desert in the Near East and Africa to the climatic extremes of hot summers and bitterly cold winters found, for example, in much of Continental Europe and central Asia Minor. Even this distinction is quite crude. Within individual regions and provinces there could be stark differences of landscape and ecology, which in turn could give rise to divergent economic structures. In Italy, for example, the upland areas in the Apennines and the Alps were best suited to pastoral economies, although the scale of activity varied in different parts of the peninsula (Frayn 1984: 11-27). More lowland regions were better suited to arable crops, although again regional variation existed: the volcanic soils of Campania were ideal for growing vines and olive trees; the latter, however, did not grow in the Po valley, where crops such as grain and rice flourished (Wickham 2005: 33-4).
Such variety - which can be mirrored in provinces throughout the empire - had ramifications for the imperial economy, a feature often trumpeted as symbolic of the unity and sophistication of the Roman world (see Schiavone 2000). There can be little dispute that there had been a considerable volume of long-distance trade under the early empire, and that this suffered disruption and retraction during Late Antiquity (McCormick 2001: 25-119; Ward-Perkins 2005: 123-36). But we must also recognize that, even during the early imperial centuries, such interregional exchanges were liable to seasonal variation (Duncan-Jones 1990: 7-29; see McCormick 2001: 444-68), were limited by quite basic sailing technologies (Horden and Purcell 2000: 123-72), and existed side by side with more local patterns of production and redistribution that were, perhaps, more characteristic of the patterns of ancient trade generally (Woolf 1992; see Morony 2004: 175-6).
In addition to differences of ecology and economy, there was considerable cultural diversity within the empire. At a basic level, this was related to ecological factors, and was manifested, for example, in the consumption of different animal products in different regions: zooarchaeological remains imply a predilection for pork in Italy and southern Gaul, beefin northern Gaul, and lamb and goat in Libya (Dyson 2003: 66). Diversity was visible too in linguistic terms, with a division between the use of Latin as the major public language (used, for instance, in inscriptions) in the western provinces and Greek in the east - although Latin was the official language of Roman law throughout the empire (Millar 1999: 105-8). It should be noted, however, that Latin was phased out as the language of governance and law in the east under Justinian. By then, the east constituted the core of imperial territory, and Latin would likely have been incomprehensible to the larger part of its inhabitants; thus Justinian’s decision was a tardy recognition of everyday realities by the slow-moving engines of the administration (Croke 2005: 73-4).
Those distinctions become more complex when considered at a more local level. Various languages had survived in spoken form throughout earlier centuries of Roman rule (Mitchell 2000: 120-1), and some - such as Punic and Libyan in North Africa and the Palmyrene dialect of Aramaic in the Near East - had also been used in public documents (W. V. Harris 1989: 175-90; Millar 1968). Late Antiquity, however, saw the emergence of a variety of vernaculars - such as Armenian, Syriac, and, to a lesser extent, Coptic - as literary languages, particularly in the service of Christian discourse (Jones 1964: 991-7; Bowersock 1990: 29-35, 57-8; Thomson 2000: 667-71). Those linguistic divergences mirrored other cultural peculiarities that might be manifested, for example, in religious terms. In Syria and Armenia, hostility to the authority of Constantinople perhaps accounts for the adherence of both areas to Monophysite Christianity (Fowden 1993: 104-9; Hugh N. Kennedy 2000: 599; but see Jones 1959).
Such diversity should not occasion a great deal of surprise, given that the empire had been assembled from a variety of cultural zones, each of which interacted with Roman culture in its own peculiar ways (useful syntheses in Ball 2000 and Wells 1999 on east and west respectively). As the example of local vernaculars shows, such diversity persisted into Late Antiquity, and whatever elements contributed toward unity occurred within that context of variety. It should be remembered, however, that those various manifestations of diversity were not necessarily coterminous. Nor were they continuous throughout the early and late Roman periods, as culture, society, and economy were constantly developing and reshaping. In particular, the empire’s Levantine provinces seem to have undergone a process of considerable economic expansion during the fifth and sixth centuries that contrasted markedly with retraction elsewhere (Morony 2004: 168-83; Ward-Perkins 2005: 123-6).
In such circumstances, we might think of the political unity of the empire as being maintained only by keeping in check various tendencies that might lead it to spring apart. That was, indeed, apparent in the very foundations of the empire, which rested on the incorporation of communities, mainly cities, that were largely left to run their own affairs. So long as various dues were paid - either tangibly as taxes or compulsory services, or symbolically in terms of allegiance and imperial service - direct interference from the imperial authorities occurred only to insure the smooth operation of the state’s administrative and legal apparatus or to settle problems whose resolution confounded the abilities of local elites (Garnsey and Saller 1987; Moralee 2004). Such a policy suited elites across the length and breadth of the empire since, through their participation in imperial government, they were able to maintain their social preeminence within their communities (Veyne 1976; Heather 1994a; Lendon 1997). Intervention by the imperial authorities was probably more intrusive after the provincial reforms of Diocletian, which saw an increase in the number of governors and other bureaucrats, and a gradual erosion of the administrative autonomy formerly enjoyed by civic elites (Jones 1964: 737-66; Liebeschuetz 2001a: 104-36). Even then, there was still a degree of tension between the aspirations of central government and the limitations imposed by time, distance, and noncompliance (Kelly 2004: 114-29).
Furthermore, the empire encompassed within its boundaries societies whose integration was, in many respects, quite minimal. Beyond the cities, there was often considerable continuity of patterns of land use and settlement from pre-Roman times (Dyson 2003: 42-5, 75-6), suggesting that, in some places (although certainly not all), integration into the empire had a minimal impact on the everyday experiences of peasant farmers. The backwardness of country folk was a common literary trope throughout antiquity; but the assertion of Bishop Syne-sius of Ptolemais that a peasant in his native Cyrenaica thought that the ruler of the world in the early fifth century was the Homeric king Agamemnon implies a world outside the cities where the high politics of the Roman Empire mattered little (Synesius, Ep. 148; Bregman 1982: 19). The limits of Roman power were not only cultural, but political also. For example, the tribes living in the mountains of Cilicia in southeastern Asia Minor constituted a society over which Roman rule had little direct influence, and they are recorded as launching attacks - variously categorized as banditry and rebellion - on the settlements of the coastal plains from the first century until the fifth (Hopwood 1989). Outbreaks of disorder in other provinces occurred throughout the empire’s existence, suggesting that the image of the Roman Empire as all-powerful needs to be tempered by frank acknowledgment of its many failures even within its territorial limits (Shaw 1984a; Nippel 1995: 100-12; Grunewald 2004). It ought to be borne in mind, for instance, that the evidence for the workings of imperial administration in the provinces - whether inscriptions or law codes - represents its response to failings within the system and, as such, might be better regarded as a guide to the aspirations of government rather than its actual achievements (Harries 1999: 77-98).
From the foregoing, it could be argued that what was truly remarkable about the Roman Empire was not that it fell, but that it should have lasted so long in an age when communications, through which its unity might be secured, were limited (Kelly 2004: 115-17). In fact, before the collapse that overcame the west in the fifth century, there were hints that different regions might pursue their own destinies, if strong, central authority were undermined. Thus the political upheavals of the third century yielded clear signs that the empire might fracture, with a breakaway empire in Britain, Gaul, and Spain (ad 259-74) and another in the Middle East ruled from the Syrian city of Palmyra (ad 260-72) (Drinkwater 1987; Millar 1993: 159-73). Other instances can be found of those tendencies, notably the various usurpers who emerged during the late third, fourth, and early fifth centuries in, particularly, the northwestern provinces (Casey 1994; Paschoud and Szidat 1997; Kulikowski 2000). In each of those cases, however, the secessionist regimes conceived of their authority in legitimate imperial terms, as can be seen from coins and inscriptions (Long 1996). That was also the case, albeit in different ways, for the various successor kingdoms that replaced the Roman Empire in the west in the fifth century. Thus in ad 500, the Ostrogothic king of Italy, Theoderic, visited Rome in the time-honored manner of a Roman emperor, performing a ceremonial arrival ( adventus), addressing the senate, hosting games in the circus, and staying in the old imperial palace on the Palatine hill (Vitiello 2004). Similarly, the Frankish kings of Gaul hosted chariot races in circuses at Paris and Soissons, apparently in imitation of practice at Constantinople (McCormick 1986: 332-4). For all its limitations, the image of imperial authority was so potent that it was the most natural means of expressing legitimacy even for elites that sought to secede from or replace it.
World History
