Empires and Barbarians

by Peter Heather

  Imparting yet another twist of intellectual intrigue to this web of argument was the natural desire of Slavic intellectuals to associate the early Slavs with the ‘best’ – in other words, technologically most advanced – sets of possibly relevant ancient remains. On once being shown some supposed early Germanic materials from the Baltic Bronze Age, Hitler became very agitated because, at the same date, Egyptians were already building pyramids. In his mind, this made the jumble of simple handmade pottery that had just been presented to him look just a touch unimpressive. The same kind of reflex has also distorted arguments over Slavic history, with many researchers wanting to associate their supposed ancient forebears with something a bit more presentable than crumbly handmade pottery. Rybakov’s identification of an original Slavic homeland in Ukraine, for instance, was based on associating the Slavs with one of the richest sets of remains from Iron Age eastern Europe: the Cernjachov culture. As we have seen, this boasted substantial settlements, iron weapons and tools, wheel-made pottery in a wide variety of designs and interesting jewellery: altogether a more satisfactory set of ancestral Slavs than other contemporary east Europeans living in sunken huts with handmade pottery of one drab design. Likewise Kostrzewski: measured in technological terms, the Przeworsk cultural system was one of the ‘best’ of Iron Age central Europe.

  In the aftermath of the Nazi era, Kostrzewski’s retorts instinctively command greater sympathy than Kossinna’s original arguments, but both were equally rooted in the demands of contemporary politics. Like so many of the alternative accounts of Slavic history produced before about 1970, constructing the best possible account of the pre-history of Central and Eastern Europe was firmly subordinated to political agendas. But in the last scholarly generation or so, and particularly since the Berlin Wall came down (although intellectual revolutions had then already been under way for a decade or more in parts of the Soviet bloc), these old political imperatives have lost much of their force. In the 1970s, Mark Shchukin demonstrated that the chronological coincidence between the rise and fall of the Cernjachov cultural system north of the Black Sea and that of Gothic power in the region was much too tight for it to be seen as anything other than Gothic-dominated. Slavic-speakers may well have lived within its boundaries, but it was the military power of the Goths that gave it shape. In Poland, too, Kostrzewski’s vision of Slavic continuity was being challenged, as profound discontinuities were shown to separate the pre-historic Lusatian and Pomeranian cultures of the first millennium BC from the Wielbark and Przeworsk cultural systems occupying the same landscape in the Roman period.5 Arguments in favour of a Slavic population between the Oder and the Vistula with a continuous history from at least c.1000 BC have lost much of their credibility, and the whole subject of early Slavic history is no longer marked by the same desperate scramble to fight off rival Slavs, minimize the role played by Germanic-speakers, and identify all the ‘best’ sets of remains as Slavic. This does not mean that the fights are going out all over Central and Eastern Europe, but current arguments are much better-tempered, and much more about the past for its own sake.

  With so much distracting superstructure stripped away, what do we now know about the Slavicization of Europe?

  Proto-Slavs

  The only possible place to begin is with the first documented appearance of Slavs in European history. Slavs – properly Sclavenes – make their debut just north of the Lower Danube frontier of the east Roman Empire in the first half of the sixth century. Writing around the year 550 AD, the east Roman historian Procopius records numerous raids by Sclavenes and Antae, whom he reports closely related, across the Danube and into Constantinople’s Balkan provinces. These attacks, or those of the Antae, to be precise, began in the reign of Justin I (518–27), although the Antae eventually became east Roman allies. By the 530s and 540s, in fact, the Sclavenes were posing the bigger problem, and Procopius’ narrative strongly implies that their attacks had steadily increased in frequency and ferocity. The recorded names of the leaders of these groups indicate that both were Slavic-speaking, and there seems no reason to doubt that Procopius’ account here is basically accurate.6 From around the year 500, then, we first find Slavic-speaking groups active in what is now Wallachia and southern Moldavia, the area between the Carpathians and the Danube.

  This region also throws up Korchak-type materials dating to the correct period. By themselves, Korchak remains cannot be dated with any accuracy, but among the Korchak materials of Wallachia and southern Moldavia archaeologists have found some datable imports. A late fifth-century brooch associated with some Korchak pottery was found at Dragosloveni in Wallachia, in a typically Korchak sunken building, and the same region has produced a famous cemetery, Sarata Monteoru, where the burials contain several brooches and belt buckles of the late fourth and early fifth century. In Moldavia, likewise, otherwise typically Korchak material has been found in several contexts alongside imported wheel-made pottery dating to the fifth century or the very beginning of the sixth, and one site near Kishinev produced a mid-fifth-century Hunnic-type mirror in a find of Korchak wares. You never know, of course, how long an item may have remained in circulation before being buried, but enough mid- to late fifth-century items have been found among the region’s Korchak materials to confirm that they spread across Moldavia and Wallachia in the later fifth and early sixth century, at the same time as Procopius above all, but also other east Roman historical sources, first record the appearance of Slavs in the same spot.7 This much is accepted by everyone. But was this Slavic presence south of the Carpathians generated by migration from elsewhere, or had Slavic groups been living here all along?

  The traditional answer has always been migration. For one thing, being so close to the Roman frontier and encompassing some territories that had even been part of it in the second and third centuries, the sub-Carpathian region is relatively well documented during the first half of the first millennium. It is not, therefore, your average run-of-the-mill argument from silence, that no source mentions Slavs here before the year 500. Other, non-Slavic-speaking groups occupied it in the Roman period. Equally striking, there is no evidence that Slavs played any significant role in the mid-fifth-century Hunnic Empire of Attila, whose remit certainly encompassed this zone. Many different subject peoples figure at different points in narratives of its rise and fall, as we have seen, but Slavs are notable only for their absence. The best that anyone has ever done in arguing for a Slavic presence in Attila’s world is to claim that the word strava – which the sixth-century Jordanes says was the term used by the Huns for funeral eulogies of their dead leader – derives from Slavic. It may do, but we know nothing about the language of the Huns, so it may, alternatively, have had its own echt Hunnic origin. It is certainly a very slender peg upon which to hang the claim that otherwise undocumented Slavs played a major role in Attila’s Empire.

  The argument in favour of migration also has its more positive dimensions. Writing in Constantinople at more or less the same time as Procopius, Jordanes gives the following famous account of the Slavs in the mid-sixth century:

  Within these rivers lies Dacia, encircled by the lofty Alps [Carpathians] as by a crown. Near their left ridge, which inclines towards the north, and beginning at the source of the Vistula, the populous race of the Venethi dwell, occupying a great expanse of land. Though their names are now dispersed amid various clans and places, yet they are chiefly called Sclaveni and Antes. The abode of the Sclaveni extends from the city of Noviodunum . . . to the Dniester, and northward as far as the Vistula . . . The Antes, who are the bravest of these peoples . . . spread from the Dniester to the Dnieper, rivers that are many days’ journey apart.

  Much of this coincides with what Procopius reports, with the extra piece of information that the Sclavenes and Antae had emerged from an earlier group called the Venedi. This is potentially of great importance because, unlike Slavs, the Venedi are mentioned in sources of the Roman period. As we have seen, Tacitus places them geographically east
of the Vistula in a broad belt of territory in between the Fenni (Finns) of the arctic north and the Carpathian Mountains. Pliny, a little earlier, had also heard of Venedae, as he named them, but gave no further information. The second-century geographer Ptolemy knew no more about them than a few extra group names. There’s no doubt that the Venedi existed and that they lived in eastern Europe in the first half of the millennium, but more than this the Romans didn’t know. This part of Europe was slightly less mysterious for them than what lay beyond, where people had ‘human faces and features, but the bodies and limbs of beasts’, but only just. The key point, of course, is that if you compare these earlier reports with Jordanes’, it is only natural to suppose that the appearance of Venedi-derived Slavs in the sub-Carpathian region around or just after the year 500 was the result of migration from the north.8

  Jordanes’ evidence also partly coincides with one of the most famous arguments of them all in early Slavic studies, deriving from the study of linguistics, and with some of the more respectable archaeological ones. All modern Slavic languages have in common an old Slavic name for the hornbeam, whereas the terms for beech, larch and yew are all Germanic loanwords. This must be, so it was argued influentially at the beginning of the twentieth century, because the hornbeam dominated the vegetation in the original ‘Slavic homeland’. On investigation, the only suitable geographical locality turned out to be the Pripet marsh area of Polesie, a rather soggy zone some 350 kilometres north of the Carpathians (Map 17). Not surprisingly, this led to much subsequent archaeological effort being expended in the Pripet region, with Irina Rusanova arguing, on the basis of extensive research there in the 1950s and 1960s, that she had unearthed the very earliest Korchak materials. The type-site of Korchak itself was one of her excavations, and led her to change Borkovsky’s original ‘Prague’ label to ‘Korchak’ for the characteristic combination of sunken huts and handmade cooking pots on the basis of its claimed anteriority.9

  This once standard vision of early Slavic history has recently been challenged, however, by Florin Curta, who argues that, on the contrary, historical Slavs emerged precisely where they are first mentioned: the south-eastern fringes of the Carpathian system. His reasoning is based on a mixture of history and archaeology. To start with, he denies the veracity of Jordanes’ report that the Slavs derived from the Venedi. Jordanes’ history can be shown to depend at certain points upon Tacitus’ Germania, and Curta argues that the Venedi–Slav linkage was Jordanes’ own invention on the basis of what Tacitus has to say – a further example of a documented tendency for Roman writers to claim that there were no ‘new barbarians’, merely old ones by new names. On the archaeological front, Curta also attacks Rusanova’s conclusions, arguing that the Korchak materials of the sub-Carpathian region are older than their equivalents in Polesie and hence could not derive from them. More positively, and this is the main focus of his broader study, Curta draws attention to the substantial body of both historical and archaeological evidence showing that those Slavs in contact with the east Roman world in the sixth century were caught up in a dynamic process of sociopolitical and economic transformation. It was, he argues, precisely this process which ‘created’ the ‘first’ or Proto-Slavs.10

  Many of Curta’s points are well taken. His demolition of Rusanova’s chronology is entirely convincing. The Polesian Korchak materials certainly postdate their equivalents south of the Carpathians. It is also very likely that somewhere in the Carpathian system is the correct zone in which to place the origins of at least those Slavs who ended up in an east Roman orbit in the sixth century. Curta himself argues for their origin in its south-eastern approaches. Another recent view, proposed first by Volodymyr Baran and taken further by Polish archaeologists of the so-called Cracow School, suggests that we should perhaps be looking more to the north-east. Here, in modern Podolia, large quantities of early-vintage Korchak materials (much earlier than those of Polesie still further to the north-east) have been unearthed. The fundamental Korchak dating problem remains, but there do seem to be more, slightly earlier, datable imports in Podolia than in Curta’s favoured spot to the south-east. Curta’s sub-Carpathian Korchak settlements also came into being after a century of sparse settlement in that region. For these reasons, it looks likely that the first Slavs explicitly to appear in the historical record had their immediate origins in a population group from north-east of the Carpathians. If so, they spread quickly. The Podolian Korchak materials can predate the Wallachian and Moldavian by at most an archaeological generation (about twenty-five years) or two.11

  I am not convinced, either, that Curta is right to be so dismissive of Jordanes. By definition, since we don’t have Jordanes’ own account of his working methods or any real means of cross-checking, the idea that he invented the link to the Venedi on the basis of Tacitus can only be hypothesis. It is not demonstrable fact, and there are some telling points against it. Jordanes started life as the military secretary of an east Roman commander stationed on the Danube frontier at the very same time that Slavic attacks were intensifying. He also provides – again, presumably, drawing on his own knowledge – very precise information on the resettlement south of the Lower Danube frontier line of various population fragments from the wreck of Attila’s Empire. This underlines his knowledge of the region, and makes it far from implausible that he had authentic information on what the Slavs of this region themselves understood of their origins. In fact, Jordanes may well also provide the earliest historical reference to a Slavic group in action, making passing reference to a war of the Gothic king Vinitharius against the Antae. Jordanes is chronologically confused here. He thought that Vinitharius lived in the later fourth century, but he was in fact one of those warband leaders that Valamer defeated to create the Amal-led Ostrogoths, probably after the death of Attila. But Jordanes certainly places this war before the Amal-led Goths moved west of the Carpathians to the Great Hungarian Plain, so this does fit with a picture of Slavic-speaking Antae operating on the eastern fringes of the Carpathians in the fifth century. It is also the case that the term ‘Wends’ – deriving from the old Roman label Venedi – was used from the seventh century onwards by many early medieval western European populations of their new Slavic neighbours; and migration, as we shall see, was a characteristic of documented Slavic-speaking populations from the sixth century onwards. Large numbers of Sclavenes and Antae of Moldavia and Wallachia would end up in the Balkans from the early seventh century, and already in the sixth, other Slavs can be documented moving west through the central European uplands. Given that, as we have seen so often before, real migration habits build up in population groups, this does add further weight to the suggestion that the first Slavic groups found north of the Lower Danube frontier in the early sixth century had moved into the region in the recent past. All this is certainly enough to make Curta’s dismissal of Jordanes at best inconclusive, and to my mind it is likely that, for once, the historian actually knew what he was talking about.12

  But there is also a bigger point here. From about the year 500, as we shall examine more closely in a moment, a huge expansion process unfolded that would make Slavic-speakers the dominant force across vast tracts of the European landscape from the Elbe to the Volga. This process is not well documented, but it does seem inconceivable that all this can have derived from an original population group confined just to Moldavia and Wallachia. Even if you do discount Jordanes on the Venedi and accept that Sclavenes and Antae emerged where they are first mentioned, this still leaves you with the overall phenomenon of Slavicization to explain. Putting tree names to one side, the linguistic evidence provides two broader points of reference.

  First, the modern Slavic language groups (east, west and south) are remarkable for how close they have remained to one another: so close in fact that they are mutually comprehensible. Everything suggests that this closeness results not from processes of linguistic convergence in the recent past, but from the fact that they split apart from one another at a comp
aratively late date. Second, Slavic languages as a whole are most closely related to those of Europe’s Baltic-speakers, whom hydronyms show to have been much more widely distributed over eastern Europe in the past than is the case today (Map 16). As we saw with the Anglo-Saxons in England, the names of larger rivers seem to have a fossil-like capacity to survive major cultural transformations. Not so long before the split between the various branches of Slavic, therefore, we must envisage a previous split between Slavic-speakers and Baltic-speakers. They had previously shared the one, older set of closely related Indo-European dialects.13 Seen in equally broad terms, the archaeological evidence paints a very similar picture. The only possible progenitors of the still extremely simple farming lifestyles visible in Korchak and closely related remains of c.500 are the subsistence-farming communities of Europe east of the Vistula and north of the Carpathians in the Roman period. Both sets of evidence tell the same story. The broader Slavic-speaking population of Europe clearly emerged from somewhere among eastern Europe’s non-Germanic-speaking population. Even if Jordanes was making up the link between the Sclavenes and the Antae (which I strongly doubt), the likelihood is that he was correct.

  THE SLAVICIZATION OF EUROPE