Empires and Barbarians

by Peter Heather

  Mother Russia

  For the Slavicization of European Russia up to the Volga we have two main reference points. The first comes from historical sources. Thanks to Islamic geographers of the tenth century, we know that territories east of the Vistula that correspond to modern Byelorussia and Volhynia were under the control of so-called ‘eastern Slavs’ at this time. The most comprehensive picture of the area at the end of the early Middle Ages, however, is provided by a still later source, the Russian Primary Chronicle, whose text, as we now have it, was a product of the early twelfth century. According to its account, by about 900 AD a number of separate Slavic-speaking groups had come to occupy a truly vast area of eastern Europe. The text has most to say about Polyanians, the Slavic-speaking group settled around Kiev where the Chronicle was composed, but many other groups and their approximate locations are mentioned in passing. It is obviously retrospective, but there is no reason to think that it misrepresents to any significant degree the spread of Slavic-speakers across the East European Plain at the turn of the millennium. Byzantine sources, above all the De Administrando Imperio (but this time providing much less problematic, contemporary information), confirm the essential outlines. By the end of the first millennium, Slavic-speaking groups dominated a huge portion of the East European Plain, taking in territory well to the east of the River Dnieper and, in the case of the Slovenes, stretching their control as far north as Lake Ilmen (Map 19).36

  From a modern perspective, it’s no surprise to find Slavic-speakers distributed so widely across Mother Russia, but our second reference point shows that this had not always been the case. All the major rivers’ names across a massive tract of territory between the Vistula and the Volga, north of the confluence of the Pripet and the Dnieper, are in fact derived from Baltic rather than Slavic languages. The conclusion seems inescapable, therefore, that Baltic-speakers had at one point dominated this landscape. The situation observable in the tenth century, when Slavic-speakers were in more or less total control of it, must have been created at some point, therefore, by Slavic expansion. This sets up the fundamental conundrum of Russian prehistory.37 In the complete absence of historical sources, which of the succession of archaeological cultures observable in the Russian landscape across the eons of the first millennium represents the initial penetration of Slavic-speakers into zones originally dominated by Balts?

  This is another subject area that has benefited hugely from the immense archaeological enterprise that unfolded in Communist Europe after 1945, and here, too, many of the old Soviet-era agendas have been subsiding. It is now possible, on one level, to tell a fairly straightforward story, starting again from the certainly Slavic-speaking world of the Korchak Carpathians in c.500 AD. By the mid-sixth century, Korchak-type materials had spread not only westwards and southwards, but also eastwards further into Ukraine. In this period, the type-site – Korchak itself – was first occupied on the River Teterev near Zhitomir, and the spread continued subsequently. In the seventh century, Korchak materials were to be found further north, in Polesie: in the Pripet marsh zone of tree-name fame. At more or less the same time, a second cultural zone of importance to our story, the so-called Penkovka system, was coming into being, which between 550 and 650 took in large expanses of the forest steppe zone of Ukraine.

  In many ways, Penkovka materials are indistinguishable from their Korchak counterparts. Both systems generated small clusters of houses on river terraces that were highly convenient for subsistence agriculture. Penkovka houses, likewise, were partly sunk into the ground, and boasted stone-built corner ovens. The only things that distinguish Penkovka remains are the biconical shape of some the larger ceramics, together with the wider variety of iron tools and decorative metalwork often found in Penkovka contexts. To the outsider (and many insiders too, in fact) the similarities are thus massively more impressive than any differences, and most scholars are confident that if Korchak-type materials were generated by Slavic-speakers, then so was the Penkovka system. Indeed, on the basis of Jordanes’ report of the relative geographical distributions of the Sclavenes and Antae, Penkovka has often been thought to have been the product of the latter, and Korchak of the former. These precise identifications are questionable, but the basic similarities of the two systems, combined with the geographical coincidence between their spread and where we actually find Slavic-speakers in the sixth century, does make it reasonable to think of Penkovka, like Korchak, as at least Slavic-dominated.38

  The later seventh century, however, was an era of major change. Across previously Korchak areas and much of the Penkovka zone, together with a substantial area to the north on either side of the Dnieper which had previously fallen beyond the boundary of either system, there arose between 650 and 750 a new culture: Luka Raikovetskaia. Like its counterparts of the same era in the western reaches of Slavdom (the Tornow, Feldberg and Leipzig systems) the main difference between Luka Raikovetskaia and its predecessors was the fact that much of its pottery was now being finished on a slow wheel. Everything suggests that, in general terms, Luka Raikovetskaia represents the reincarnation of the Korchak and Penkovka systems in an era of technological advance, although debate continues over the extent to which it is correct to identify it as one unified system or to pick out instead a number of local variants.

  At the same time, some of the easternmost Penkovka areas underwent a further and very distinctive process of development. This zone, together with other territories, which had previously fallen outside the Penkovka system altogether, saw the development of the so-called Volyntsevo culture. Aside, again, from slightly different ceramics, it is distinguished from the Luka Raikovetskaia by a strikingly greater prevalence of both metalwork and strongholds. Its history began in the seventh century, again, but it continued to spread into the eighth, at which point its development is marked by the acquisition of a new name, Romny-Borshevo. The ceramics of this system stand in a direct line of development from the Volyntsevo, but spread over a much wider area (the main reason for the change of name), and particularly into the basins of the Upper Don and the Oka. Its settlement sites are also characterized by the still more extensive use of fortifications. After the period of initial formation, Luka Raikovetskaia and Romny-Borshevo both continued in unbroken sequences of development, with an ever-widening geographical distribution, through to the tenth century. By this point, they extended over the areas where most of the Slavic groups named in the Russian Primary Chronicle were located (Map 19), and there seems no doubt, therefore, that a direct association can be made between tenth-century Slavic-speakers and these two archaeological systems.39

  The rise and fall of the material-cultural systems of the East European Plain is now easy enough to follow in outline, and two sets of equations – between Korchak and Penkovka and known Slavic-speakers of the sixth century, and between Luka Raikovetskaia and Volyntsevo and Romny-Borshevo and known Slavic-speakers of the tenth – seem secure enough.

  But this is not a historical narrative, and should not be confused with one. What can be charted with some security now is the developing sequence of pottery traditions on the East European Plain in the second half of the first millennium. Historical sources also make it clear that the later phases of Volyntsevo and Romny-Borshevo coincide geographically and chronologically with the dominance of Slavic-speakers by the tenth century. But pots aren’t people, and trying to understand the human history that underlies these ceramic sequences, and their relationship to broader historical patterns of state formation and migration, raises many further questions.

  Two have particular force. First, the introduction of wheel-made pottery makes it difficult to be certain how direct was the line of evolution from Korchak and Penkovka populations to those of Luka Raikovetskaia and Volyntsevo. Did the new systems come into being just because Korchak and Penkovka potters adopted a new ceramic technology? If so, then given that the Korchak and Penkovka were in all probability dominated by Slavic speakers, presumably so too were the Luka Raikovet
skaia and Volyntsevo. This is the usual assumption, but the ceramic transformations could be hiding a much more complex human history, and the improved pottery – wheel-made pots are better than their handmade counterparts – might also have been adopted by non-Slavs. Second, what is the human history behind the subsequent spread of the patterns of the Luka Raikovetskaia and Volyntsevo cultural systems further north and east from the eighth to the tenth centuries? Was this an expansion of living human beings, or merely the spread of new habits among existing populations? These are both questions to which we will return.

  Underlying both, however, is the still bigger issue that emerges when archaeology is confronted with linguistics. Does the documentable spread north and eastwards of the Korchak and Penkovka systems, a trajectory taken further by their Luka Raikovetskaia and Volyntsevo successors, represent the initial Slavicization of Russia and Ukraine, and the overall removal of these areas from a Baltic-speaking orbit In one view, this is perfectly possible. Some linguists, as we have seen, would date the initial separation of the Slavic and Baltic language families to the middle of the first millennium AD, making it natural to equate the appearance of Korchak-type cultures in the Carpathians rim with this moment of linguistic definition. If so, the subsequent spread north and eastwards of probably related archaeological cultures would in all likelihood represent the initial Slavicization of Mother Russia.

  But other linguists would date the Baltic/Slavic split rather earlier, possibly even to the second millennium BC. And, in line with this, other archaeologists would argue, on the basis of the general patterns of life that generated them, that some of the systems to be found on the Baltic side of the hydronym divide in the mid-first millennium – in particular the so-called Kolochin culture – are so similar to those that generated the Korchak system that it is arbitrary to suppose the latter was dominated by Slavic-speakers, and the former not. Again, this is, a priori, a perfectly possible argument. In that view, what we would be seeing in the spread of the Korchak system would be the ability of one particularly successful group of Slavic-speakers to spread their domination over an already broadly Slavic-speaking landscape. The first model – where the political dominance of Slavic-speakers and the arrival of the language go hand in hand – would resemble the broad pattern observed already in the Balkans and central Europe. But the less dramatic possibility – that much of Russia and Ukraine became Slavic-speaking sometime before our period – cannot be ruled out.40

  MIGRATION AND THE SLAVS

  From the outer foothills of the Carpathian system in the late fifth century, Slavic groups spread decisively south into the Balkans in the seventh, after a period of aggressive raiding that had lasted through most of the sixth. Other Slavic-speaking groups were at the same time spreading into southern Poland (the Mogilany group of the early sixth century) and westwards along the northern foothills of the Carpathians, reaching Moravia sometime in the first half of the sixth century, Bohemia in the second, and the confluences of the Havel, Saale and Elbe in the early seventh. A second line of advance also reached the Elbe perhaps only a little later, having spread north and west from the Vistula, but, as we have seen, the internal chronology of the Sukow-Dziedzice system remains vague. Dynamic sixth-century Slavic expansion towards the west was fully matched in Ukraine on the other side of the Carpathians, where both the Korchak and the closely related Penkovka systems spread over large areas in the sixth century. Even this much expansion, however, does not explain the dominance of Slavic-speaking groups across large areas of previously Baltic-speaking regions visible by the tenth century. The spread subsequently of the Luka Raikovetskaia and Volyntsevo systems over and beyond Korchak and Penkovka areas may reflect this, but, as we have seen, that story may be much more complicated than the simple linear progression of material cultures might suggest.

  Without historical evidence, it is impossible to know how dispersed Slavic-speakers already were in eastern Europe in c.500 AD. The geographical range of the action and the number of different forms it took does suggest, though, that there must have been substantially more Slavic-speakers at this point than just the Sclavenes and Antae of Moldavia and Wallachia, but this still makes Slavic expansion, particularly into European Russia, highly problematic. The spread of the Penkovka, Luka Raikovetskaia and Volyntsevo systems over different parts of the East European Plain really could represent either the initial Slavicization of these territories, or the triumph of one particular group of Slavic-speakers over their peers. Even more fundamentally, none of this does more than sketch in the barest outlines of Slavic expansion.

  By what means did Slavic-speakers spread their domination over such a large part of the European landscape, and what caused this fundamental revolution in European history?

  Trying to understand the human history behind the creation of Slavic Europe is even more difficult than exploring the migratory activities of the Germanic groups caught up in the fall of the Roman Empire, for two main reasons. First and foremost, the historical evidence has yet more holes in it. The second reason would apply anyway, even if the literary evidence were fuller. The appearance of Sclavenes and Antae in the sub-Carpathian region by c.500 AD represented one huge revolution in itself, in that Slavic-speakers are not previously documented in this area. But other Slavic-speakers then spread over a vast range of times and places to create Slavic Europe as it stood at the end of the first millennium: Moravia, Bohemia and Ukraine in the sixth century, the Balkans in the seventh, the Russian forest zone as late as the eighth and ninth, and north-central Europe sometime in between. It is inconceivable that expansionary activities that were dispersed so widely in time and space could all have taken a single form.

  Flows of Migration

  In the absence of historical sources of good quality, the scale of the population units involved in Slavic migration is particularly difficult to estimate. The only decent-looking figures refer to sixth-century raiding parties, who consistently came in groups of a few thousand. On one occasion, a mixed group of 1,600 Huns, Antae and Sclavenes burst on to east Roman territory; on another, 3,000 Slavic raiders had to pay the Gepids a gold coin apiece to be ferried to safety. Hildegesius’ mixed warband of Gepids and Slavs comprised 6,000 men, and a reportedly ‘elite’ force of 5,000 Slavs made a surprise attempt to storm the defences of Thessalonica in 598.41 These figures are reasonably consistent, but they refer to a different kind of activity from the expansionary migration that affected the Balkans in the seventh century and the Carpathians and central European uplands in the sixth. It is not likely that the same kinds of social group were responsible for both activities.

  As its various contexts indicate – ranging from central Europe in the sixth century to Lake Ilmen in the ninth – there were so many different processes involved in the creation of Slavic Europe that it is worth confining the discussion initially to sixth- and seventh-century central Europe and the Balkans, where we have at least some historical documentation. But even just within these spheres, two completely different kinds of outcome are visible in archaeological terms. In some contexts – particularly in the foothills of the Carpathians, Moravia, Bohemia, the Elbe–Saale region, and western and southern Ukraine in the sixth century, together with parts of Thrace shortly after 600 AD – the upshot was the more or less complete transfer to a new area of a Korchak-type material culture in all its measurable expressions: lifestyle, technology and social patterns. The only thing that varies between these areas is the shape of some of the pottery. A very different archaeological result is found in many other areas that we know to have been Slavic-dominated, but where recovered archaeological assemblages have produced only isolated Korchak elements in what is overall a much more varied body of material. This pattern prevailed over much of the former Roman Balkans and much of the North European Plain west of the Vistula from the seventh century, where investigation has thrown up only a few Korchak-type ceramics, not the whole system transferred to a new area. Any account of Slavic expansion must account f
or both of these consequences.

  Despite the lack of narrative sources, the translation of whole Korchak-type socioeconomic systems into entirely new landscapes is suggestive of a particular kind of migration process. The standard settlement unit – therefore, presumably, also the socioeconomic one – prevailing in these areas was small. Korchak hamlets typically consist of groupings of no more than ten to twenty small dwellings, each clearly designed for nuclear families. On reflection, these hamlets also provide an indication of the maximum size of the basic migration unit involved in areas that have produced a ‘pure’ Korchak result. Korchak Europe was clearly created by the spread of such units outwards from the foothills of the Carpathians, and they were moving either as ready-made communities of this kind (the maximum view) or possibly in even smaller groups that came together only at their destination. Korchak dwellings look large enough for about five people, so we’re looking here at migration units of no more than fifty to a hundred. Comparing this phenomenon with the different migration forms we have so far encountered, the likeliest process to have produced it was something along the lines of expansion by wave of advance (see page 22). In about one hundred and fifty years, as we have seen, Korchak remains spread from the fringes of the Carpathians to the Lower Elbe, while retaining much of their basic cultural form. The extended chronology of these remains makes clear that this group of Europe’s Slavic-speakers was more conservative than once thought. Older chronologies confined Korchak settlements to the fifth and the earlier sixth century, but we now know that some groups maintained this lifestyle for two centuries or more, spreading slowly in small groups across the European landscape.42