The book contents and index, Editorial Introduction, and text files see the CONTENTS page.

The bottom line of the study: Neither the sources, nor archeology allow to equate “Sclavenes” and “Antes” with the origin of the Slavs. The work and the heart of the its Conclusion are confirming the fact that before 12th c., “no “Slavs” called themselves by this name, no group took on the label imposed by outsiders”, and “the first clear statement that “we are Slavs” comes from the twelfth-century Russian Primary Chronicle”. Behind the supra-ethnical names “Sclavenes” and “Antes” of the 6th-12th cc. stand numerous ethnically-assorted local horse pastoralist nomadic tribes, generically named Wends and Winidi (Wendeln “Wanderers” Vandals)  by the contemporaries, and in the case of the Sclavenes identified with Severeis, Cutrigurs, and Suvars, relatively small Türkic tribes of the Western Hunnic circle. The numerous multi-ethnic local sedentary agrarian tribes called Slavs were made in a Türkic cauldron, and are known from the distinctly Greek and Roman mirrors, with terminology largely formed from their perspective.

The greatest achievermernt of the F. Curta's Conclusion is the decoupling of the terms “Sclavenes” and “Antes” from the term “Slav”. Best of all it reflected in the depiction of the Sclavene society and its traits, which are contrasting with that of the Slavs. By ascribing to the ancestors of the later Slavs the traits typical to their masters, the Slavs are endowed with ephemeral characteristics that conflict with the well-known Slavic innate ethnology. This conflation follows the footsteps of the Russian Slavistics, rightly discarded by the author.

Posting clarifications, comments, and additional subheadings are (in blue italics) and blue boxes. Clipped details are replaced with [...] . Brief references are moved in-line. Page numbers are shown at the end of the page in blue. The adjective Türkic and the noun Türk are used to denote the global world for the Türkic community that includes Turkish and Turks as one of the constituents; Türk is a noun of which the Turkic and Türkic are adjectival derivatives, it is needed for rendering the Russian terms, with four distinct designations for four phenomena. The semantics of the above terminology in English vs. Russian is a result of their national histories.

112

CONCLUSION (Chapter 3. Slavs in early medieval sources)

I began this chapter with the statement that the nature of the Slavic settlement remains obscure to many modern historians. Several conclusions follow from the preceding discussion, but the most important is that, whether or not followed by actual settlement, there is no “infiltration” and no obscure progression. The evidence of written sources is quite explicit about this.

Could then “migration” be an appropriate term? Modern studies have shown that migration is a structured aspect of human behavior, involving a more or less permanent change of residence. Historians, however, generally treat migration as chaotic and inherently not explicable through general principles. Recent formulations of migration as a structured behavior have established that migrations are performed by defined subgroups (often kin-recruited) with specific goals, targeted on known destinations and likely to use familiar routes. Most migratory streams develop a counterstream moving back to the migrants' place of origin. [109. Lee 1966; Anthony 1990; Gnielch 198]

The problem with applying this concept of migration to the sixth- and seventh-century Slavs (i.e. Sclavene?) is that there is no pattern of an unique, continuous, and sudden invasion. Moreover, until the siege of Thessalonica during Heraclius' early regnal years, there is no evidence at all for outward migration, in the sense of a permanent change of residence. Almost all raids reported by Procopius in the mid-sixth century were followed by a return to the regions north of the Danube frontier (Typical nomadic punitive raids for being in the rears on payments). At times, the Sclavene warriors may have spent the winter on Roman territory, as in 550/1. However, Menander the Guardsman makes it clear that the wealth acquired during Sclavene raids was usually carried back home, across the Danube.

John of Ephesus, on the other hand, claims that in 584, after four years of raiding, the Sclavenes were still on Roman territory They had become “rich and possessed gold and silver, herds of horses and a lot of weapons, and learned to make war better than the Romans”. This, however, could hardly be interpreted as an indication of Slavic (i.e. Sclavene?) settlement. What John had in mind were warriors, not migrant farmers. Michael the Syrian, in a passage most likely taken from John, describes a Sclavene leader who took with him the ciborium (covered cup) of a church in Corinth, not a chief establishing himself in the conquered city. The only evidence for such a decision is that of the Sclavene tribes besieging Thessalonica in the early years of Heraclius' reign (610 – 641). They had brought their families with them, for they intended to establish themselves in the city following its conquest. This also indicates that they were not coming from afar, for the prisoners they had taken after the siege could return to Thessalonica carrying the booty taken by the Sclavenes from the inhabitants of the city.
113

Moreover, some of the tribes mentioned in the second homily of Book II are described in the fourth homily as living in the immediate vicinity of the city When did they settle there? Paul Lemerle argued that in the 610s a Slavic (i.e. Sclavene?) settlement around Thessalonica must have been a relatively recent phenomenon. How recent, however, is impossible to tell. The evidence regarding the mid-600s and the second half of that century suggests that the Sclavenes were by then already established at a short distance from the eastern frontier of the Lombard kingdom and from Constantinople. In 681, as the (Asparukh) Bulgars moved south of the Danube, there were already Slavic (i.e. Sclavene?) groups in the eastern Balkans and around Thessalonica. Judging from the existing evidence, therefore, a true migration could have taken place only during a relatively short period of time, namely not long after Heraclius' accession to power.¹¹⁰ To Theophylact Simocatta, writing about Maurice's reign on the basis of a late sixth- or early seventh-century source (the Feldzugsjournat), Sklavinia was still located north of the Danube frontier. In the mid-600s, the Sklaviniai moved to the outskirts of Constantinople and Thessalonica.

The survey of Slavic (i.e. Sclavenian?) raiding activity during the sixth and the early seventh century points to another important conclusion. There seems to be a certain raiding pattern (Table 4). Independent Sclavene raids began in the 540s, with a long interruption after 551/2 (Justinian I, 527 – 565, resumed tribute payments to Bulgars in 551/2 after a loss of Egypt and in view of war with Arab Caliphate in Anatolia). They resumed in the late 570s (with Avars at the helm, Bulgars shoved to a dependent role, and Avars demanding payments to be made to them) and seem to have come to an end only after Maurice's (582 –  602) campaigns north of the Danube (591–595). A new phase opened with massive raids, both on land and on sea, during the early years of Heraclius' reign (610 – 641). One can hardly fail to notice that this pattern coincides with major engagements of Roman armies on other fronts: in Italy, in the 540s and 550s, as well as in Persia and on the eastern front in the 570s, the 580s, and the 610s. It has indeed been shown that the pattern of information-movement across the Danube frontier proves that northern peoples often seem to have known when sectors of the Empire's defense were weakened as a result of Roman problems elsewhere. The Sclavenes of 550, who were bent on capturing Thessalonica, quickly changed their plans as soon as they learned that Germanus was in Serdica.

110. John of Ephesus vi 25; Miracles of St Demetrius II.2.196. See Lemerle 1981:90. No evidence exists, however, that the Sclavenes established either on the frontier of the Lombard kingdom or near Constantinople came from regions located north of the Danube.
Sklavinia north of the Danube: Theophylact Simocatta viii 5.10. Whitby's unfortunate translation ("Peter prepared to move camp against the Sclavene horde") stands for [...]. See also Litavrin 1984:196.
114

The figures advanced by Menander the Guardsman and Archbishop John of Thessalonica for the Sclavene raids of the 580s were no doubt exaggerations. They suggest the efforts of these authors to explain why barbarians achieved success against the Empire in spite of being numerically and organizationally inferior to the Romans. In the 580s and the late 590s, the Sclavenes seem to have known remarkably well where to strike, in order to avoid major confrontations with Roman armies, and when to attack, in order to take advantage of the absence of troops.¹¹¹

I would stress, however, another important conclusion following from the preceding discussion. None of the Sclavene raids in the 540s or early 550s was organized under the leadership of a chief. Procopius could distinguish “armies” from “throngs,” but ignored any names of Sclavene chiefs or leaders. He claimed that the Sclavenes and the Antes “were not ruled by one man, but they [had] lived from old under a democracy, and consequently everything which, involved their welfare, whether for good or for ill, was referred to the people.” As the story of “phoney Chilbudius” suggests, the Antes did not even have a name for the Roman official, who was supposed to guide them into some sophisticated organization. They just called him. “Chilbudius”. ¹¹²

However, writing as he did in c. 560, Pseudo-Caesarius knew that, though living without the rule of anyone, the Sclavenes often killed their leaders “sometimes at feasts, sometimes on travels.” At the turn of the century, the picture radically changed, as the author of the Strategikon now recommended that Roman officers win over some of the Sclavene chiefs by persuasion or gifts, while attacking others, “so that their common hostility will not make them united or bring them together under one ruler.” As soon, as the Sclavene raids resumed in the late 570s, we learn of many Sclavene leaders, apparently different in status from each other. Names such, as Dauritas, Ardagastus, Musocius, and Peiragastus are in sharp contrast to the lack of any chief-names in Procopius' work. Other names, such as Chatzon, Samo, Dervanus, Walluc, or Perbundos, appear in seventh-century sources. Is the absence of names in Procopius' work just an illustration of his idea of “Slavic democracy” or does this reflect some aspect of Slavic society? This question is most difficult to answer. It is hard to understand, however, why Procopius should invent the “Slavic democracy” if nothing justified the use of this concept for contemporary Slavic society.

11 Menander the Guardsman, fr. 20,2; Miracles of St Demetrius 1 13.117. For the pattern of information-movement, see Lee 1993:141-2.
12 Procopius, Wars vn 14.22. See Benedicty 1963:46 and 1965:53; Evans 19189:63. There are many names of barbarian leaders in Procopius' Wars: Datios, Aordos, and Suartua, kings of the Herules (v 15.29 and 33); Torisind, king of the Gepids (vn 18.3); Auduin, king of the Lombards (vi 34.5); and Chinialon, the Cutrigur chief (vn 18.15).
115

See Table 4

It is interesting to note that, with the exception of the quasi-legendary King Boz of the Antes, Procopius' contemporary, Jordanes, also ignores any Slavic (i.e. Sclavene?) leaders. I am inclined, therefore, to take Procopius' evidence as a strong argumentutn ex silentio. Something had radically changed in Slavic (i.e. Sclavene?) society as the Slavic (i.e. Sclavene?) raiding activity resumed in the late 570s. A detailed discussion of this change is to be found in Chapter 7. For the moment, it is important to note that in terms of their social organization, the Sclavenes of the 580s were different from those of the 540s.¹¹³

Finally, there are important changes concerning the very name of the Slavs. Until the first decade of Heraclius' reign (610s), as Sclavene groups settled on Roman territory, all sources — Greek, Latin, or Syriac — spoke exclusively of Sclavenes and/or Antes. The author of Book 11 of the Miracles of St Demetrius was the first to introduce tribal names, such as the Drugubites, the Sagudates, the Belegezites, the Berzites, and the Rynchines. Fredegar spoke of Wends and Theophanes of Severeis. The evidence is too strong to be interpreted as mere accident.

The author of the Strategikon, a direct participant in Maurice's campaigns of the 590s, knew only of Sclavenes and Antes. The campaign diary later used by Theophylact Simocatta, but most likely written at about the same time as the Strategikon, also used only 'Sclavenes' and 'Antes.' In this particular case, 'Sclavenes' was an umbrella-term for various groups living beyond the frontier, in Sklavinia. As soon as Sklaviniai moved south of the Danube, the precise affiliation to any particular “tribe” became a key issue. Indeed, some “tribes” are described as hostile and bent on conquering Thessalonica, while others appear as friendly, willingly supplying food to the besieged city. The same may be true for Fredegar’s Wends. As they successfully fought the Avars and elected a king for themselves, the Sclavenes, in Fredegar's eyes, became “different” and required a new name, 'Wends.' A similar conclusion follows from Theophanes' account. According to him, after crossing the Danube in 681, the Bulgars did not encounter an undifferentiated mass of 'Slavs,' but (at least) two groups., the Severeis and the Seven Tribes. The newcomers approached and treated them as two separate entities.

What all this suggests, in my opinion, is that the name 'Sclavene' was a purely Byzantine construct, designed to make sense of a complicated configuration of ethnies on the other side of the northern frontier of the Empire.

113 For independent [...] Sclavenes killing their leaders, see Riedinger 1969:302. Sclavene chiefs united under one ruler: Strategikon xi 4.30 [...]. For Boz, see Jordanes, Getica 247. Paul the Deacon also avoids mentioning any Sclavene leaders, though at the time he wrote the History of the Lombards, the Carantani were already organized as a polity under the “dynasty” of dux Boruth. The Life of St Hrodbert, bishop of Salzburg, indicates that the Carantani had a rex not long after Arnefrit, the son of the Friulan duke Lupus, fled ad Sclavorum gentem in Carnuntum, quod corrupte vocitant Carantanum (to the people Sclavorum Carnuntum, which is distorted to Carantanum) (Historia Langobardorum v 22; Vita Hrodberti, p. 159).
116

Byzantme criteria for classifying ethnic groups were substantially different from ours. In spite of their common language, “an utterly barbarous tongue,” the Sclavenes and the Antes were often at war with each other.

On the other hand, the author of the Strategikon knew that there was more than one Sclavene king, and that Sclavene “kingdoms” were always at odds with one another. Despite obvious differences in status, the name 'Slavs' applies to both those attacking Constantinople in 626 as allies of the Avars and those who were the subjects of the Kagan. It might be that 'Sclavene' was initially the self-designation of a particular ethnic group. In its most strictly defined sense, however, the “Sclavene ethnicity” is a Byzantine invention.¹¹⁴

114 Procopius, Wars vn 14.26; Strategikon xi 4.30. For the name 'Sclavene,' see Pckkancn 1971; Schelesniker 1973:11; Schramm 1995:165.
117

222

CONCLUSION (Chapter 5. Barbarians on the frontier)

There are at least three important conclusions to be drawn from this sweeping survey of the archaeology of the Carpathian basin and the steppe north of the Black Sea. First, in all cases discussed in this chapter, material culture may be and was indeed used for the construction of social identities. Despite interaction across the buffer zone between the Danube and the Tisza rivers, clear material culture distinctions were maintained in a wide range of artifacts. The nature and function of these distinctions is very similar to those identified by Ian Hodder in the Baringo district of Kenya. As in East Africa, material culture contrasts were maintained in order to justify between-group competition and negative reciprocity. Displays of emblemic styles were particularly important at the time of the Lombard—Gepid wars in the mid-500s. More often than not, such styles were associated with the status of aristocratic women, wives, daughters, or mothers of “kings”.  This may be a result of the special emphasis laid on public representation of group identity, but may also be an indication of the intricate relationship between ethnic and gender identity.

The examination of hoards of silver and bronze also shows that women were symbolic vehicles for the construction of social identities. In this case, however, it is more difficult to decide precisely what kind of identity was constructed through displays of female dress accessories. Unlike the Carpathian basin, the specific way in which identity was expressed was not funerary assemblage but lavish offering of silver and bronze artifacts, which may have represented a particular form of potlatch.

55 Artifacts in the Khacki hoard were wrapped in silk (Bobrinskii 1901:147). It is tempting to associate the burial of earlier hoards, for which a date may be tentatively assigned to с. 630, with the beginning of the civil war within the western division of the Gok Turk Empire and the subsequent rise of the Bulgar Kaganate. However, both were located in Left Bank Ukraine and the steppe north of the Caucasus mountains, at a considerable distance from the main concentration of hoards in the Middle Dnieper area. Moreover, the hoarding phenomenon clearly continued through the second half of the seventh century, as hoards were buried in Volhynia and Slovakia.
224

However, just as with displays of wealth in rich female graves, deposition of hoards may have served as “tournaments of value”. [56. The phrase is that of Bradley 1990:139] Like funerals, hoards were used for social display mainly during periods of instability when the status of the individual needed underlining. An important route to social advancement was most likely access to foreign goods, such as Byzantine silver plate. Within the Empire, the social status which silver plate conferred or reflected was often seen in terms of wealth and power. The donation of family silver to be recycled into liturgical silver or given to the poor became a literary topos. In “barbarian” contexts, transactions in which silver plate was symbolically displayed were certainly different. To claim that acquisition, imitation, and use of Roman silver plate reflects the degree to which barbarians were Romanized [57. Mango 1995:81] is simply to ignore that the symbolic system changed with the changing contexts in which imported objects were employed. There can be no doubt, however, that Byzantine silver plate was viewed as “exotic” and “precious” for an image of power, for stamped objects were only produced for imperial distribution. The ability to acquire fine largitio objects carried a considerable premium. The same is true for objects of possibly Sassanian (224–651 AD) origin, such as the Zemiansky Vrbovok bowl. On the other hand, that hoards of silver conveyed an image of power much stronger than grave-goods may be deduced from the fact that some contain several sets of ornaments, which suggests that such collections were the property of more than one person. In other words, hoards of silver and bronze may have permitted a more “extravagant” display of metalwork than the provision of grave-goods.

Finally, the survey of the archaeological evidence from the Carpathian basin and the steppe north of the Black Sea strongly suggests that in order for material culture to participate in the construction of social identities, artifacts need to be given meaning in social context. Swords with P-shaped sheath attachments or stirrups with elongated attachment loops were not “Avar” because of being of Central Asian origin, but because of being used in a specific way in specific transactions (such as display of grave-goods) in the new social milieu in which Avar warriors found themselves after с 570. Similarly, there are no specific “Lombard” or “Gepid” brooches, for many fibulae found in female burials on both sides of the “no man's land” either were “imported” from distant locations or imitated such exotic imports. There is, therefore, no “phenotypic” expression of a preformed ethnic identity, though identity is constructed by manipulating certain artifact-categories. The value of each of these artifact-categories depended less on questions of supply and origin than on the social strategies employed by those who used them.
225

On the other hand, objects that are the prerogative of an elite may be imitated by lower-ranking groups. “Citations” from the material culture discourse which can be identified in rich burials or hoards may be found in completely different contexts, such as settlements. As I will argue in the last chapter, just as in the case of “Lombard” and “Gepid” identities, Slavic ethnicity may have been communicated through displays of objects whose use was restricted to local elites. In such cases, artifacts similar to those found in Ukrainian hoards are not mere analogies. They have become metaphors.

CONCLUSION (Chapter 6. Elites and group identity)

The archaeological study of identity and status is often based on the analysis of burial assemblages, notably of the nature and symbolism of grave goods. The extensive use of cremation, rather than inhumation, as well as the possible use of funerary rites that may have left no trace in the archaeological record, prevented the use of such data for Slavic archaeology.⁸⁶ The data on which this chapter is based derive primarily from settlement excavations. Despite this bias, there are some important conclusions to be drawn for the reconstruction of social organization and ethnic identity.

First, there is already enough evidence to move away from the migrationist model which has dominated the discipline of Slavic archaeology ever since its inception (Chapter 1). A retreat from migrationism is necessary simply because the available data do not fit any of the current models for the study of (pre)historic migration. Cultural correspondences were too often explained in terms of long-distance migration, despite lack of any clear concept of migration to guide such explanations. Recent research in anthropology and other social sciences laid a strong emphasis on discriminating between such diverse phenomena as seasonal population movements, “scouting,” and outward migration. It has become increasingly evident that migrations across ecological or cultural boundaries would require considerable planning on the part of the migrants, and should leave substantial and clear archaeological evidence. '"Cultures, as one archaeologist noted, “do not migrate. It is often only a very narrowly defined, goal-oriented subgroup that migrates”. ⁸⁷ To speak of the Prague culture as the culture of the migrating Slavs is, therefore, a nonsense.

86 Zoll-Adamikova 1980:948 and 1983. The long awaited publication of the large cemetery at Sarata Monteoru may bring significant changes to current views on status and identity expressed through mortuary displays. For an interesting study based exclusively on cemeteries, see Losert 1991.
87 Anthony 1990:909. For the archaeological model of (pre)histonc migrations, see Rouse 1986; Stark, Clark, and Elson 1995.
307

Furthermore, the archaeological evidence discussed in this chapter does not match any long-distance migratory pattern. Assemblages in the Lower Danube area, both east and south of the Carpathian mountains, antedate those of the alleged Slavic Urheimat in the Zhitomir Polesie, on which Irina Rusanova based her theory of the Prague-Korchak-Zhitomir type. More recent attempts to move the Urheimat to Podolia and northern Bukovina are ultimately based on the dating of crossbow brooches found at Kodyn and some other places. These brooches, however, are not the only late fifth- or early sixth-century artifacts in the area. Despite lack of closed finds comparable to those at Kodyn, there are good reasons to believe that at least some archaeological assemblages in south and east Romania go back as early as с 500.

The evidence is certainly too meager to draw any firm conclusions, but from what we have it appears that instead of a “Slavic culture” originating in a homeland and then spreading to surrounding areas, we should envisage a much broader area of common economic and cultural traditions. The implementation of an agricultural economic profile, which is so evident on later sites, is very likely to have involved some short-distance movement of people. The dominant type of economy seems to have been some form of “itinerant agriculture” which encouraged settlement mobility. Suzana Dolinescu-Ferche's research at Dulceanca brilliantly illustrates this model. Such population movements, however, cannot be defined as migration. There is simply no evidence for the idea that the inhabitants of the sixth-and early seventh-century settlements in Romania, Moldova, and Ukraine were colonists from the North.⁸⁸

Nor does the idea of a “Slavic tide” covering the Balkans in the early 600s fit the existing archaeological data. South of the Danube river, no archaeological assemblage comparable to those found north of that river produced any clear evidence for a date earlier than с 600. By contrast, there is no doubt that many early Byzantine forts in the Balkans were abandoned only during Heraclius' early regnal years (Chapter 4). The ceramic assemblages found at Argos and Olympia have nothing to do with these developments, for there are good reasons to believe they are of a much later date. It is unlikely that either the small settlement at Musici or the cremation cemetery at Olympia existed at the time of the final withdrawal of Roman armies from the Balkans. The archaeological assemblages at Garvan may also be of a much later date than assumed by the archaeologist who led excavations there. Though both Greece and Albania produced clear evidence of seventh-century burial assemblages, they have nothing in common with the “Slavic culture” north of the Danube river.

88 Slavic Urheimat in Podolia and Bukovina: Baran 1978, 1981, 1991, and 1994. See also Godkowski 1979. For Dulceana, see n. 57. For settlement mobility, see also Shtefan 1968.
308

The analysis of a considerable number of settlement features found in Romania, Moldova, and Ukraine has shown, on the other hand, that the second half of the sixth century and the early seventh century was a period of crucial change in the culture history of communities leaving north of the Danube river. While the existence of many settlements may have begun at a much earlier date, it is precisely during this period that they came to share a number of stylistic traits which may have been associated with emblemic styles. Pots ornamented with finger impressions or notches on the lip, clay pans, and Grubenhauser with stone or clay oven are just a few examples of regional styles which became the norm in the late 500s and early 600s. Not all represented ethnicity, as suggested by Christian symbols incised on pots. Others may have represented cross-regional identities, as in the case of “Slavic” bow brooches with their ornamental patterns pointing to long-distance social contacts. Symbols drawn from “exotic” milieus may have been culturally authenticated and transformed into “native” symbols. The production of local series of bow fibulae, some imitating larger or more sophisticated specimens, may indicate this process. As such “imports” were “internalized,” emulation of elite styles may have contributed to the dissemination of ornamental patterns.

Second, the analysis of intrasite distributions of artifacts suggests that with the agricultural economy established as a dominant subsistence pattern, processing and consumption of special, cereal-based foods, such as flat loaves of bread, became an essential ingredient of communal activities. The principal locus for these activities was now the communal front region of the settlement. Finds of tools and clay pans cluster around this region. This may have also been an arena for ceremonies orchestrated to convey complex messages of group identity (The communal feasts have more to do with religious rites and social events that purely ethnic affiliation; multi-ethnic congregation is rather a norm than an exception).

It is against this background that the relative status of those who wore “Slavic” brooches becomes visible in both social and archaeological terms. Since fibulae were primarily female dress accessories, it is likely that, as with contemporary hoards of silver and bronze in Ukraine, women were symbolic vehicles for the construction of social identity. Just what kind of identity was symbolized is a matter of how “Slavic” bow fibulae are to be interpreted. Wearing a Mazurian or a Crimean brooch may have given the wearer a social locus associated with images of power. Wearing a local reproduction of such a fibula was, no doubt, a very different statement, though still related to status. Beyond emulation, therefore, “Slavic” bow fibulae, particularly much cruder specimens, without complicated scrollwork ornaments, may have conveyed a message pertaining to group identity. Whether living within the same region or widely scattered, adherence to a brooch style helped to integrate isolated individuals within a group whose social boundaries crisscrossed those of local communities.
3099

At the same time, brooches articulated a hierarchy of identities both within and between those communities. Production of bow fibulae involved knowledge of complicated technological processes and access to them was certainly restricted by the ability either to procure such items from distant locations or to employ a craftsman with enough experience and skill to replicate ornamental patterns and brooch-forms. Just as with “Lombard” and “Gepid” brooches, “Slavic” bow fibulae were not “phenotypic” expressions of a preformed ethnic identity. There were no Slavic fibulae per se. Access to and manipulation of such artifacts, however, may have been strategies for gaining admission into a group of people known to Byzantine authors as “Slavs”.

CONCLUSION (Chapter 7. “Kings” and “democracy”: power in early Slavic society) (Slavic should be enclosed in the quotation marks: “Slavic”)

There is no indication of Slavic chiefs before с. 560. Notwithstanding his detailed description of Slavic society, Procopius knew nothing about them. He carefully recorded, however, the names of several other barbarian leaders in the area, especially kings of the Gepids, Herules, and Lombards, or Cutrigur chieftains. That this is no accident is shown by Procopius' claim that both Sclavene and Antes “are not ruled by one man, but they lived from old under a democracy” [43. Procopius, Wars vii 14.22] No Slavic raid recorded in the Wars seems to have been organized by military leaders and the story of the “phoney Chilbudius,” with its emphasis on the false identity of the would-be chief of the Antes resonates with Procopius' notion of Slavic “democracy”. The Slavic ethnographic excursus, which is probably based on his interviews with Sclavene and Antian mercenaries in Italy, is the longest in all of Procopius' work. As a consequence, the absence of Slavic leaders cannot be explained by either Procopius' lack of interest or his hostility towards those whom he viewed as nomads (see Chapter 2). His image of the Slavs is much more favorable than that of their neighbors in Procopius' oikumene. But he seems to have denied political leadership only to the Slavs. There is no reason to believe, however, that Procopius deliberately omitted the names of Slavic military leaders, when he was so attentive in distinguishing Sclavene “throngs” from Sclavene “armies” (see Chapter 3).

332

The first political leaders appear in Pseudo-Caesarius' Eratopokriseis, which was written in the 560s. The largest number and the widest variety of leadership forms, however, occur in sources regarding the last quarter of the sixth century. Names of individual chiefs suddenly appear in Menander the Guardsman, Strategikon, and Theophylact Simocatta. In sharp contrast to the picture given by Procopius, the author of the Strategikon even suggests that Sclavene chiefs may at times unite and accept, albeit temporarily, being “ruled by one man”. This is also the period in which chiefs emerged, who spoke in the name of their respective groups, boldly proclaiming their independence. It is also during this period that chiefs, often mentioned by name, were leading more or less successful raids across the Danube. These were the raids which most strikingly coincided with major engagements of the Roman armies in the east. The chiefs knew where and when to strike, in order to avoid major concentrations of Roman troops. This strongly suggests that among all three categories of leaders discussed in this chapter, which may have possibly existed at that time, warrior leaders (great-men) were the most common.

The end of the sixth century is also the period in which we can see increasing competition between chiefs. The author of the Strategikon knew that there were many Sclavene “kings, always at odds with each other,” a useful political detail for any Roman general who happened to wage war against any one of them. What were the stakes of this competition, we can only guess.

As shown in Chapter 6, the second half of the sixth century was a period of dramatic change in the material culture of communities living north of the Danube river. Shortly before and after AD 600, symbols of personal identity were in higher demand. The greatest number of links between ornamental patterns displayed by bow fibulae found in Romania, Crimea, and Mazuria is that of specimens dated to this period. Long-distance connections, as well as the display of different patterns on various groups of “Slavic” bow fibulae point to social competition. If the intrasite distribution of artifacts in the common front region of the sites analyzed in Chapter 6 can, in any way, be associated with competitive feasting, which is a typical feature for big-man leadership, we may be able to visualize some aspects of this competition. War, however, was the overwhelming concern of those who, though unable to fight in ordered battle, were nevertheless extremely skillful in ambushing Roman troops.

333

That Slavic society was geared up for warfare is evident from the significant quantity of weapons, especially arrow and spear-heads, that were found on sixth- and seventh-century sites.⁴⁴ It is therefore possible that at least some of the evidence for destruction by fire, which sixth- to seventh-century sites in Romania, Moldova, and Ukraine occasionally produced, is the result of inter-group conflicts. After all, as the author of the Strategikon observed, in the Slavic “democracy", “nobody is willing to yield to another”. ⁴⁵

44 Strategikon xi 4.30. It was often noted that “Slavic” settlements produced no weapons (e.g., Dolinescu-Ferche 1984:145). The archaeological evidence, however, gives a different picture. Arrows: Dolinescu-Ferche 1979:205 fig. 22/15; Rosetti 1934:212 fig. 7/4; Dolinescu-Ferche and Constantiniu 1981:322 fig. 18/9; Turcu and Ciuceanu 1992:200; Teodor I98o:fig. 31/8; Mitrea 1974—6:figs. 16/1 and 15/5 and 1994:328 fig. 27/5; Teodor 198413:29 fig. 6/7; Dolinescu-Ferche I986: fig. 22/21, 22; Szekely 1974—6:pl. x/24 and 25; Toropu 1976:211; Rafalovich and Lapushnian 1974:133 fig. 10/15; Rafalovich 1968:96 fig. 29/8. Spears: Zirra and Cazimir 1963:60; Constantiniu 19651x182; Szekely I992: pl. X/8; Vakulenko and Prikhodniuk 1984:68 fig. 38/1. For a battle-axe, see Vakulenko and Prikhodniuk 1984:68 fig. 39/9.
45 Strategikon xi 4.14. Heavy destruction by fire of numerous buildings is evident at Bucharest-Maicaneshti and Kavetchina. By contrast, only two buildings of the large settlement at Davideni were destroyed by fire. The same is true for Selishte and Dulceanca III.
334

CONCLUSION: THE MAKING OF THE SLAVS

As its title suggests, the subject matter of this book is not the Slavs, but the process leading to what is now known as “the Slavs”. This process was a function of both ethnic formation and ethnic identification. In both cases, the “Slavs” were the object, not the subject. [...] Though in agreement with those who maintain that the history of the Slavs began in the sixth century, I argue that the Slavs were an invention of the sixth century. Inventing, however, presupposed both imagining or labeling by outsiders and self-identification.[1 Ivanov 1991c and 1993]
335

MIGRATION

A brief examination of the historiography of the “Slavic problem” yields an important conclusion: the dominant discourse in Slavic studies, that of “expert” linguists and archaeologists, profoundly influenced the study of the early Slavs. Though the evidence, both historical and archaeological, presented itself in a historical light, historians were expected merely to comb the written sources for evidence to match what was already known from the linguistic-archaeological model. Because this model was based on widely spread ideas about such critical concepts as culture, migration, and language, the basic assumptions on which the model was based were rarely, if ever, questioned. One such assumption was that ethnies, like languages, originate in an Urheimat and then expand over large areas through migration. Migration was defined in the terms of the Kulturkreis school, as the relatively rapid spread of racial and cultural elements. This led many scholars to abandon a serious consideration of the historical evidence and to postulate instead a Slavic Urheimat located in the marshes of the Pripet river. Chased from their homeland in the North by the rigors of the harsh climate, the Slavs then inundated Eastern Europe.

A Slavic homeland implied, however, that the history of the Slavs was older than the first Slavic raids known from historical sources. The cornerstone of all theories attempting to project the Slavs into prehistory was Jordanes' Getica. Jordanes equated the Sclavenes and Antes with the (Venethi, Venedi, Wends) also known from much earlier sources, such as Pliny the Elder, Tacitus, and Ptolemy. This made it possible to claim the Venedi of Tacitus, Pliny, and Ptolemy for the Slavic history. It also provided a meaning to archaeological research of “Slavic antiquity.” A Polish linguist, Tadeusz Lehr-Spbwinski, first suggested that the archaeological culture of the Vistula basin during the first century вс to the first century AD, which was known as the Przeworsk culture, was that of Tacitus' Venedi. Soviet archaeologists argued that the Slavic Venethi were the majority of the population in the area covered by the Chernyakhov culture of the fourth century AD. They claimed that by AD 300, the Antes separated themselves from the linguistic and archaeological block of the Venedi, and were soon followed by the Sclavenes. More often than not, therefore, the task of the archaeologist was to illustrate conclusions already drawn from Jordanes' account of the Slavic Venethi.

336

[...] My argument is that instead of being an eyewitness account, Jordanes' description of Sclavenes and Antes was based on two or more maps with different geographical projections, the imaginary space of which he filled with both sixth-century and much earlier ethnic names he found in various sources. This seriously diminishes the value of the most important piece of evidence invoked by advocates of both a considerable antiquity of the Slavs and their migration from the North. Moreover, no source dated before Justinian's reign (527-65) refers to Slavs or Slavic Venethi. Despite some overlap in timespans covered by Procopius' Wars and the chronicle of Marcellinus Comes (including the continuation to 548 added by another author), there is no mention of Slavs in the chronicle. Procopius, on the other hand, made it very clear that a “Slavic problem” arose, along with others, only during Justinian's reign.[2 Procopius, Secret History 18.20—1]

The Slavs did not migrate from the Pripet marshes because of hostile environmental conditions. Nor did they develop forms of social organization enabling them to cope with such conditions and presumably based on cooperation and social equality (zadruga). Niederle's thesis does not stand against the existing evidence and has at its basis an outdated concept of migration. That the migrationist model should be abandoned is also suggested by the archaeological evidence examined in Chapter 6. No class of evidence matches current models for the archaeological study of (pre)historic migration. More important, assemblages of the Lower Danube area, where, according to the migrationist model, the Slavs migrated from the Pripet marshes, long antedate the earliest evidence available from assemblages in the alleged Urheimat. Short-distance population movements, but not migration, must have accompanied the implementation of a form of “itinerant agriculture,” which, though not based on the slash-and-burn method, may have encouraged settlement mobility.

That the Slavs were present on the northern bank of the Danube before the implementation of Justinian's building program in the mid-500s is demonstrated by their raids known from Procopius. It will probably remain unknown whether or not any of the groups arguably living in contemporary settlements excavated by Romanian archaeologists called themselves Sclavenes or Antes. This, however, was the region from which Romans recruited mercenaries for the war in Italy This is also the region that produced the largest number of coins struck under Emperors Anastasius and Justin I, as well as during Justinian's early regnal year. [...] It is hard to judge from the existing evidence, but from what we have it appears that the Slavic raids mentioned by Procopius originated in this same region. This may also explain why Chilbudius' campaigns of the early 530s targeted against Sclavenes, Antes, and Cutrigurs were directed to a region not far from the Danube river.

337

We are fortunate to have first-hand sources of information for the late 500s and the early 600s, such as the Strategikon, and the campaign diary used by Theophylact Simocatta's Books vi—viii. In both cases, our knowledge, however restricted, of what was going on north of the Danube river is based, almost certainly, on eyewitness accounts. Neither Theophylact nor the author of the Strategikon knew any other area of Slavic settlements except that located north of the Danube frontier. Furthermore, no clear evidence exists of an outright migration of the Slavs (Sclavenes) to the regions south of the Danube until the early years of Heraclius' reign. Phocas' revolt of 602 was not followed by an irresistible flood of Sclavenes submerging the Balkans. In fact, there are no raids recorded during Phocas' reign, either by Sclavenes or by Avars. By contrast, large-scale raiding activities resumed during Heraclius' early regnal years. This is also confirmed by the archaeological evidence discussed in Chapter 4. Some forts along the Danube or in the interior were destroyed by fire at some point between Justinian's and Maurice's reigns. In many cases, however, restoration followed destruction and forts were abandoned at various dates without signs of violence. After Maurice's assassination, Phocas' army returned to the Danube and remained there at least until 605, if not 620. This is clearly attested by Sebeos and does not contradict in any way what we know from the archaeological and numismatic evidence. The earliest archaeological evidence of settlement assemblages postdating the general withdrawal of Roman armies from the Balkans is that of the 700s. This suggests that there was no “Slavic tide” in the Balkans following the presumed collapse of the Danube frontier. In addition, the archaeological evidence confirms the picture drawn from the analysis of written sources, namely that the “Slavs” were isolated pockets of population in various areas of the Balkans, which seem to have experienced serious demographic decline in the seventh century.

[...] the disintegration of the military system in the Balkans, which Justinian implemented in the mid-500s, was the result not so much of the destruction inflicted by barbarian invasions, as of serious economic and financial problems caused both by the emperor's policies elsewhere and by the impossibility of providing sufficient economic support to his gigantic building program of defense. This conclusion is substantiated by the analysis of sixth-century Byzantine coin hoards, which suggest that inflation, not barbarian raids, was responsible for high rates of non-retrieval.
338

ETHNICITY AND ETHNIE: THE VIEW FROM THE INSIDE

After Chilbudius' death in 533, there was a drastic change in Justinian's agenda in the Balkans. From this moment until Maurice's campaigns of the 590s, no offensive strategy underpinned imperial policies in the area. Instead, Justinian began an impressive plan of fortification, of a size and quality the Balkans had never witnessed before. [...] These measures were not taken in response to any major threat, for Roman troops were still in control of the left bank of the Danube, possibly through bridge-heads such as those of Turnu Severin (Drobeta) and Celei (Sucidava). This is shown by the edict 13, issued in 538, which clearly stated that troops were still sent (if only as a form of punishment) north of the Danube river, “in order to watch at the frontier of that place.

In addition to military and administrative measures, Justinian offered his alliance to the Antes (foedus of 545) and began to recruit mercenaries from among both Sclavenes and Antes for his war in Italy. All this suggests that Chilbudius' (Roman general) campaigns of the early 530s opened a series of very aggressive measures on the Danube frontier, which were meant to consolidate the Roman military infrastructure in the Balkans. It is during this period of aggressive intrusion into affairs north of the Danube frontier that Sclavenes and Antes entered the orbit of Roman interests. Justinian's measures were meant to stabilize the situation in barbaricum, which is why the foedus with the Antes was only signed after the end of the war between Antes and Sclavenes. Whether or not he intended to create a buffer zone between the Danube frontier and the steppe corridor to the northeast, Justinian's goal was only partially fulfilled.

Two devastating invasions of the Cutrigurs, in 539/40 and 558/9, respectively, broke through both Justinian's system of alliances and his fortified frontier. None of the subsequent Sclavene raids can be compared in either size or consequences to the Cutrigur invasions. However, knowing that the first recorded raid of the Sclavenes is in 545, it is possible that Sclavene raiding was a response to Justinian's aggressive policies, with both the fortified frontier and his barbarian allies. The Sclavenes may have felt encouraged by the Cutrigur breakthrough of 540, but it is no accident that their first raid coincided with Justinian's alliance with the Antes.

339

The interruption of Sclavene raids coincides with the completion of the building program. With the exception of Zabergan's invasion of 558/9, there were no raids across the Danube for twenty-five years. This is an indication of the efficiency of the defensive system, consisting of three interrelated fortification lines, the strongest of which was not along the Danube, but along the Stara Planina. Later, this grandiose program was extended to the northwestern Balkans, following the defeat of the Ostrogoths and the conquest of Dalmatia. Along the Danube and in the immediate hinterland, forts were relatively small (less than 1 hectare of enclosed area). Each one may have been garrisoned by a numerus (tagma), the minimal unit of the early Byzantine army, with up to 500 men. This may explain why small armies of Sclavenes (such as those responsible for the raids in the late 540s and early 550s) had no problems taking a relatively large number of forts. It also explains why Sclavene or Avar armies, no matter how large, moved with remarkable speed after crossing the Danube, without encountering any major resistance. The excavation of forts and the estimation of the number of soldiers who may have manned these forts in the Iron Gates area indicate that the entire sector may have relied for its defense on forces amounting to some 5,000 men, the equivalent of a Roman legion (3. This figure may have been even smaller if, as suggested in Chapter 4, some forts were inhabited by soldiers with their families). If, as argued in Chapter 7, the population of a Sclavene χωρίον was somewhat inferior in size to one or two bandons (400 to 800 men), we may be able to visualize the effort of mobilizing warriors for a successful raid across the Danube, which a great-man like Ardagastus (Sclavene chieftain/great-man) may have faced. It is hard to believe that any chief was able to raise an army of 100,000, as maintained by Menander the Guardsman. (4 Menander the Guardsman, fr. 20,2) The 5,000 warriors who attacked Thessalonica at some point before 586, nevertheless, is a likely figure. In any case, there is no reason to doubt the ability of Archbishop John, who may have been an eyewitness, to give a gross estimate of the enemy's force. If so, then this indicates that raids strong enough to reach distant targets, such as Thessalonica, usually aimed at mobilizing a military force roughly equivalent to a Roman legion. Furthermore, there is no evidence, until the early regnal years of Heraclius, of an outright migration of the Slavs (Sclavenes) to the region south of the Danube river.⁵ No evidence exists that Romans ever tried to prevent the crossing, despite the existence of a Danube military fleet. Moreover, all major confrontations with Sclavene armies or “throngs” took place south of the Stara Planina mountains.

5. Forts, at least those of medium and large size, were permanently occupied, but the number of soldiers actually manning these forts may have considerably varied in time. Judging from the general picture of military operations in the 500s, it is likely, however, that this number was often too small. By contrast, during the fourth century no less than 15,000 milites ripenses were in charge with the Danube frontier in Dobrudja. See Aricescu 1977:189.

З40

Nevertheless, the efficiency of the fortified frontier, at least in its initial phase, cannot be doubted. During the last fifteen years of Justinian's reign, no Slavic raid crossed the Danube. The implementation of the fortified frontier seems to have been accompanied by its economic “closure.” This is shown by the absence of both copper and gold coins dated between 545 and 565 in both stray finds and hoards found in Romania. The economic “closure” was not deliberate, for it is likely that the strain on coin circulation, which is also visible in hoards found south of the Danube frontier, was caused by the very execution of Justinian's gigantic plan.

[Follows review of archeological finds, pages 341-342]

[...] The earliest changes in material culture which can be associated with emblemic styles and arguably represent some form of group identity postdate by a few decades the first mention of Sclavenes and Antes in historical sources. Can we call (Slavic) ethnicity this identity constructed by material culture means? The analysis presented in Chapter 5 shows that material culture may have been and indeed was used for the construction of ethnicity. Despite intensive interaction across the “no man's land” between the Tisza and the Danube, clear material culture distinctions were maintained in a wide range of artifacts. Material culture contrasts were created and maintained in order to justify between-group competition. As a consequence, emblemic styles were particularly visible during the Lombard-Gepid wars of the mid-500s. Because group identity, and especially ethnicity, necessitated public displays of such styles, artifacts used for the construction of ethnicity were, more often than not, associated with the female apparel, in particular with that of aristocratic women. The same is true for hoards of silver and bronze in the Middle Dnieper area. In addition, hoards emphasize that an important route to social advancement was access to foreign goods, such as Byzantine silver plate. Finally, ethnicity, as defined in the first chapter, presupposes an orientation to the past, determined by charismatic entrepreneurs, who gather adherents by using familiar amalgamative metaphors. The inspiration for many ornamental patterns on “Slavic” brooches were fifth-century decorative patterns, such as the Gava-Domolospuszta scrollwork, brooch forms of the Aquileia class, or pairs of bird heads. At least bird heads can be viewed as “citations” from the “heroic” past, for this decoration was typically associated with artifacts dated to the times of Attila's Hunnic Empire. [7. See Werner 1956:72—3; Bierbrauer 1994:147; Kazanski 1993]
343

To judge from the existing evidence, the rise of the local (i.e. Slavic) elites was coincidental with the dissemination of emblemic styles which may have represented some form of group identity. It is very likely that this is more than simple coincidence. Big-men and chiefs became prominent especially in contexts in which they embodied collective interest and responsibility. Chiefs like Dauritas (Daurentius/Dauritas, Sclavene chief) and Samo “created” groups by speaking and taking action in the name of their respective communities. Political and military mobilization was the response to the historical conditions created by the implementation of the fortified frontier on the Danube. In this sense, the group identity represented by emblemic styles was a goal-oriented identity, formed by internal organization and stimulated by external pressure. The politicization of cultural differences is, no doubt, one of the most important features of ethnicity Repeated production and consumption of distinctive styles of material culture may have represented ethnic identity. The construction of ethnicity was, however, linked to the signification of social differentiation. Changing social relations impelled displays of group identity. The adoption of the dress with bow fibulae was a means by which individuals could both claim their membership of the new group and proclaim the achievement and consolidation of elite status. (8. For a slightly different interpretation of the “Slavic culture,” see Gorman 1998:109)

Can we put the name “Slavic” to this (or these) ethnic identity(-ies)? As suggested in Chapter 3, the Sclavene ethnicity is likely to have been an invention of Byzantine authors, despite the possibility, which is often stressed by linguistically minded historians, that the name itself was derived from the self-designation of an ethnic group. It is interesting to note that this ethnic name (slovene) appeared much later and only on the periphery of the Slavic linguistic area, at the interface with linguistically different groups. Was language, then, as Soviet ethnographers had it, the “precondition for the rise of ethnic communities"? ⁹ In the case of the Slavic ethnie, the answer must be negative, for a variety of reasons.

First, contemporary sources attest the use of more than one language by individuals whom their authors viewed as Antes or Sclavenes. The “phoney Chilbudius” (Antian POW) was able to claim successfully a false identity, that of a Roman general, because he spoke Latin fluently, and Perbundos, the “king” of the Rynchines (“Slavic” tribe), had a thorough command of Greek. In fact, language shifts were inextricably tied to shifts in the political economy in which speech situations were located. (10. See Urciuoli 1995) Just how complicated this political economy may have been is shown by the episode of the Gepid taken prisoner by Priscus' army, during the 593 campaign. He was close to the Sclavene “king” Musocius and communicated with him in the “king's language”. Formerly a Christian, he betrayed his leader and cooperated with Priscus, presumably using Latin as the language of communication. Finally, both the Gepid traitor and Musocius' Sclavene subjects, who were lured into the ambush set by Roman troops, were accustomed to Avar songs, which were presumably in a language different from both Slavic and Latin.

9 Kozlov 1974:79. Among the ethnic groups using slovene as a name were the Slovenes of Novgorod, the Slovincy (or Kashubians) of the Baltic area, the Slovaks, and the Slovenes of Slovenia. See Ivanov and Toporov 1980:14; Schramm 1995:199 (None of the ethnonyms cited is older than 9th c. a clearly anachronic justification).

344

Second, Common Slavic itself may have been used as a lingua franca within and outside the Avar Kaganate. This may explain, in the eyes of some linguists, the spread of this language throughout most of Eastern Europe, obliterating old dialects and languages. It may also explain why this language remained fairly stable and remarkably uniform through the ninth century, with only a small number of isoglosses that began to form before Old Church Slavonic was written down.^1l This is also confirmed by the fact that Old Church Slavonic, a language created on the basis of a dialect spoken in Macedonia, was later understood in both Moravia and Kievan Rus.

The same conclusion can be drawn from the episode (642) of Raduald, duke of Benevento, reported by Paul the Deacon and discussed in Chapter 3. Raduald, who had previously been duke of Friuli, was able to talk to the Slavs who had invaded Benevento, coming from Dalmatia across the sea. Since the duchy of Friuli had been constantly confronted with Slavic raids from the neighboring region, we may presume that duke Raduald learned how to speak Slavic in Friuli. His Slavic neighbors in the north apparently spoke the same language as the Dalmatian Slavs. [12 Paul the Deacon, Historia Langobardorutn iv 39 and 44.] (I.e. the duke of Friuli learned the Slavic lingua franca from the Slavic raiders, captured or else. That is as much plausible as for the Carribean leaders to learn English from the pirate Blackbeard. Likelier, the raiders of the Benevento spoke Türkic, and the lingua franca of the Friuli was the Sarmatian Türkic of the Venethi/Wends/Wendeln, and then the duke of Friuli did not have a problem communicating with those “Slavs” in the Sarmatian/Bulgarian Türkic)

11 See Lunt 1985:203; Birnbaum 1992:7 and 1993:359. Lunt (1997:36) believes that the migration of the Slavs forced “bands from different areas” of a relatively small homeland to adapt their language to further communication, each one giving up peculiarities and substituting equivalent characteristics of their neighbors or comrades. The closing of a more or less uniform development of Slavic is set at the approximate time of the “fall of the weak jers” (i.e., the disappearance of the reduced vowels ь and ъ in certain well-defined positions) and the subsequent “vocalization of the strong jers” (i.e., the development of these reduced vowels to regular full vowels in other positions). See Birnbaum 1975:4. This sound shift cannot be dated earlier than с. б00 and some linguists argue that it should be dated to the tenth, if not twelfth, century. Another terminus ad quem for the late Common Slavic is the palatalization of velars, a phenomenon which, according to some scholars, did not take place before с. б00. Finally, the metathesis of the liquids began only after с 750 and was complete before с. 900 (Birnbaum 1975:228 and 232). One of the most frequently cited arguments for a late date of Common Slavic is Charlemagne's name (Carolus) which presents in all modern Slavic languages (in which the name designates the “emperor”) similar and archaic phonetical treatment. See Ivanov 1976:44; Patrut 1976:187
345

Slavic was also used as a lingua franca in Bulgaria, particularly after the conversion to Christianity in 865. It is only the association with this political development that brought Slavic into closer contact with other languages. This explains why, despite the presumed presence of Slavic-speaking communities in the Balkans at a relatively early date, the influence of Common Slavic on the non-Slavic languages of the area — Romanian,¹³ Albanian,¹⁴ and Greek¹⁵ - is minimal and far less significant than that of Bulgarian, Serbo-Croatian, and Macedonian. The absence of a significant influence of Common Slavic in the Balkans is also evident from the small number of Balkan place names of Slavic origin, which could be dated on phonetical grounds, with any degree of certainty, before с. 800. [16. Schramm 1981:160]

As with material culture emblemic styles, the Slavic language may have been used to mark ethnic boundaries. The emblematic use of Slavic, however, was a much later phenomenon and cannot be associated with the Slavic ethnie of the sixth and seventh centuries. [17. See Eastman and Reese 1981]

Slavs did not become Slavs because they spoke Slavic, but because they were called so by others.

13 The greater part of Slavic loans in Romanian seem to be of literary origin (Church literature charters, and popular literature). See Nandris, 1939. Only fifteen words can be attributed to Common Slavic influence on the basis of their phonetical treatment. For a complete list and discussion, see Mihaila 1973:16; Duridanov 1991:15. All fifteen words appear in all Romania dialects, both north and south of the Danube river. See Mihaila 1971:355. One of the earliest loans is shchiau (pl. shchei) (shkiau/shkei), a word derived from the Slavic ethnic name (Latin Sclavus), which is commonly applied to Bulgarians. See Hurdubetiu 1969; Petrovics 1987. No other word of a very long list of Slavic loans in Romanian can be dated earlier than the ninth century. See Barbulescu 1929; Patrut 1968, 1971, and 1974:104-5, 121, and 241; Manczak 1988. For a statistics of Slavic loans in Romanian, see Rosetti 1954:12; Patrut 1971:301 with n. 10. Some phonological and morphological features, such as pre-ioticization or the vocative case, may indeed be the result of Slavic language contact, but there is no way of establishing a chronological framework for these phenomena. See Petrucci 1999:53, 56-7, 105-8, 118, and 130. Moreover, phonological features long considered to have been borrowed from Common Slavic proved to be segments that developed internally. See Petrucci 1995.

14 As in Romanian, the transformation of /n/ into /r/ (a linguistic phenomenon known as rotacization) ended before the largest number of Slavic loans entered Albanian. See Brancus 1989. Only three words have been identified as certainly Common Slavic loans. See Hamp 1970; Ylli 1997. As in Romanian, the word Shqa in the Geg dialect of northern Albania refers to any Slavic-speaking group of the Orthodox faith, particularly to Bulgarians. See Mihaila 1973:16; Schramm 1995:192.

15 Among all non-Slavic languages in the Balkans, Greek has the smallest number of Slavic loans. Gustav Meyer (1894) identified only 273 words of Slavic origin. See also Popovic 1959:718; Borntrager 1989. The majority of Slavic loans seems to have entered Greek between the eighth and the eleventh century. The number of Common Slavic features, however, is comparatively higher in place names. See Malingoudis 1985 and 1987.
345

THE SLAVIC ETHNIE: THE VIEW FROM THE OUTSIDE

All written sources of the sixth century and some of the seventh use exclusively (Türkic, or Hunnic, or Sarmaian, or Scythian terms) Sclavenes and/or Antes to refer to groups living north of the Lower Danube. Though the author of the Strategikon specifically mentioned that there were many (Bulgarian or Savir) “kings” (over Sclavenes and/or Antes), which suggests more than one political and, presumably, ethnic, identity, there are no other names besides Sclavenes and Antes.¹⁸ Moreover, despite the fact that the (Bulgarian or Savir vassals) Antes were since 545 the allies of the Empire, the author of the Strategikon listed them among potential enemies. By contrast, the first tribal names (Drugubites, Sagudates, Belegezites, etc.) appear almost concomitantly in Book II of the Miracles of St Demetrius and in Fredegar. In both cases, the difference between ethnies was important, because of differing political interests linked with various ethnicities. Some of the tribes described in Book II of the Miracles of St Demetrius were among those besieging Thessalonica. They were viewed as savage, brutish, and heathen. Others, like the Belegezites, were friendly and, at times, potential and important allies, who were able to supply the besieged city with food. To Fredegar, the Wends were different from the rest of the Slavs because of their successful revolt against the Avars, and, more important, because of their role in the demise of Dagobert's power (reference to Dagobert's defeat by the Shambat/Samo Wends/Vandal/Wendeln/Nomads in 631/632). The same is true for Theophanes' account of the Bulgar migration (Wends are different from the rest of the Slavs). The two Slavic groups mentioned in connection with the conquest by Asparuch's warriors of northeastern Bulgaria have specific tribal names, because they were treated differently by both Byzantium and the conquering Bulgars. The Severeis (Savirs, the former leaders of the Western Hunnic Köturgur Western Wing “Cutriguri”) were resettled on the frontier between the Bulgar Kaganate and Byzantium, while the επτά γενεαί (Seven tribes [of Sclavenes]),¹⁹ who until then had probably been clients of the Byzantine emperor (actually, subjects of Savirs/Severeis and members of the Köturgur Western Wing), were moved on the western frontier against the Avars (to serve as Anchues, border guards).

18 The same is true for the contemporary source used by Theophylact Simocatta for his account of Priscus' and Peter's campaigns north of the Danube river in the 590s (Only names Sclavenes and Antes).
19 Whether or not this was the real name of this group or simply a phrase employed by Theophanes for some sort of tribal confederation, he made it clear that they were a category separate from the Severeis (Savirs).

Danube Bulgaria Kaganate after Asparukh
S. Runciman “A history of the First Bulgarian Empire”
Avaria frontier is north of the Danube Bulgaria

346

In all those cases, ethnicity was a function of power in a very concrete and simple way. Ethnies were not classified in terms of language or culture, but in terms of their military and political potential.

Names were important, therefore, because they gave meaning to categories of political classification. If this is true, however, then “Antes” were also a similar example, since from 545 to 602, they played a completely different role for imperial policies on the Danube frontier than the Sclavenes. The Antes were constantly allies of the Romans, while Sclavenes always appeared on the side of their enemies. A different Antian ethnicity may thus have existed irrespective of the common, “utterly barbarous,” language, which, according to Procopius, both ethnies used. [20. Procopius, Wars vii 14.26] Emperor Maurice's campaigns of the late 500s against all potential and true enemies (Avars and Sclavenes) may have blurred this difference or at least made it negligible. In the eyes of the author of the Strategikon, the Sclavenes and the Antes not only had the same customs, weapons, and tactics, but both were treated as potential enemies.

347

In the light of these remarks, the very nature of a Sclavene ethnicity needs serious reconsideration. Procopius and later authors may have used this ethnic name as an umbrella-term for various groups living north of the Danube frontier, which were neither “Antes,” nor “Huns” or “Avars”. Jordanes did the same, though unlike others, he chose an ancient name, the Venethi (in fact implying Wends/Vandal/Wendeln/Nomads, or the “Huns”, the “Avars” did not exist at the time of Jordanes writing, A. D. 551), probably because he believed that the contemporary configuration of gentes beyond the limits of the Empire was a consequence, if not a reincarnation, of that described by ancient authors such as Tacitus or Ptolemy. To him, in other words, the barbarians of the sixth century, unless touched by the course of Gothic history, were frozen in time and space, basically the same and in the same places as viewed by the ancient authors. That no Slavic ethnicity existed in the eyes of any sixth- or seventh-century Byzantine author, which could be compared to the modern concept of ethnicity, is shown by Pseudo-Caesarius' usage of the term “Sclavenes”. To him, the opposite of “Sclavenes"is 'Ριπιανοί, which was not an ethnie, but a name for the inhabitants of the Roman province of Dacia Ripensis.²¹ The contrast is that between a group living north and another living south of the Danube frontier, to which Pseudo-Caesarius referred by the biblical name Physon. His focus was on the specific location, within one and the same climate, of groups supposedly different in customs and religious life. The same is true for the author of the Strategikon. If Sclavenes were discussed in a different chapter than Avars, it is because, in his eyes, they had radically different social and political systems and, as a consequence, different forms of warfare. Roman generals, therefore, ought to learn how to fight them differently. Nevertheless, when it comes to real raids, the evidence discussed in Chapter 3 reveals that many authors were not at ease pinning down who exactly was ravaging Thrace in the 580s and who, at the same time, was in Greece.

This, I must emphasize, is in sharp contrast to other authors' concepts of Slavic ethnicity.²² That to our sixth- and seventh-century authors, ethnicity was an instrument to differentiate between enemies and allies is also shown by Theophylact Simocatta's episode of the Gepid captured by Priscus' army in 593. To the author of the Feldzugsjournal used by Theophylact as a source for Priscus' campaign, this “Gepid” was different from “Sclavenes,” even if he had chosen to live among them and was a friend, if not a subject, of “king” Musocius. His “Gepid” ethnicity became apparent and important only when it became necessary to make a difference between him, a former Christian, and the other, “Sclavene” prisoners, who refused to reveal the location of their chief's village. Unlike them, the “Gepid” deserter would become a key factor for the successful conclusion of Priscus' campaign.

21 For a different, but unconvincing, interpretation of 'Ριπιανοί, see Dragojlovic 1972—1973.
22 See Pohl 1988:107—8. In relation to the Slavic raids of the 580s, and especially to the evidence of the Chronicle of Moncnwasia, Walter Pohl believed that the reason for which many authors muddled Avars and Slavs was the fluidity of the early medieval concept of ethnicity. Those who viewed themselves politically closer to the Avars, chose to leave at the end of the raid, together with the Kagan. Those who presumably remained and settled in Greece, became Slavs. The analysis presented in Chapter 3 shows, however, that this interpretation does not stand against the existing evidence.
348

Viewed from this perspective, ethnicities were just labels attached to various actors in historically determined situations. Like all labels, they were sometimes misleading. The author of the Strategikon warns against those still claiming to be “Romans” (Ρωμαίοι), but who “have given in to the times”,  forgot “their own people”, and preferred “to gain the good will of the enemy”, by luring Roman armies into ambushes set by the Sclavenes. To the experienced soldier who wrote the Strategikon, any ethnicity, including a Roman one, should be treated with extreme suspicion, if not backed by a politically correct affiliation. [23. Strategikon xi 4.3 1]

Byzantine authors seem to have used “Sclavenes” and “Antes” to make sense of the process of group identification which was taking place under their own eyes just north of the Danube frontier. They were, of course, interested more in the military and political consequences of this process than in the analysis of Slavic ethnicity. Chiefs and chief names were more important than customs or culture. When customs and culture came to the fore, as in the case of the Strategikon, it was because its author believed that they were linked to the kind of warfare preferred by Sclavenes and Antes. A similar concept may have guided Procopius in writing his Slavic excursus. It is because of their military skills that the Sclavenes and the Antes caught the attention of the Roman authors. As early as 537, Sclavene mercenaries were fighting in Italy on the Roman side. The first Sclavene raid recorded by Procopius predates by only five or six years the publication of the first seven books of the Wars. In his work, Procopius viewed the Sclavenes and the Antes as “new” and their presence in the Lower Danube region as recent.

Although he constantly referred to Sclavenes in relation to Huns or other nomads, there is no indication that he believed them to have recently come from some other place. That he considered them to be “new” can only mean that they had not, until then, represented a political force worth being treated like the Lombards, the Gepids, the Cutrigurs (mentioning Cutrigurs in this context defeats the argument of the author: the “Cutrigurs” were “Huns”), and other “allies” surrounding the Empire. It is because he thought the Sclavenes and the Antes were not politically important (or, at least, not as important as Lombards, Gepids, or Cutrigurs) that Procopius failed to record any chief names. To be one of Justinian's ευσπουδοι, one needed first to have a “king.” The irony behind the episode of the “phoney Chilbudius,” with its plot setting imitating that of a neo-Attic comedy, is that the Antes, who eventually became Justinian's ευσπουδοι, did not have a true leader, for they had “lived from old under a democracy”.
349

The making of the Slavs was less a matter of ethnogenesis and more one of invention, imagining and labeling by Byzantine authors. Some form of group identity, however, which we may arguably call ethnicity, was growing out of the historical circumstances following the fortification of the Danube limes. This was therefore an identity formed in the shadow of Justinian's forts, not in the Pripet marshes. There are good reasons to believe that this identity was much more complex than the doublet “Sclavenes-Antes” imposed by the Byzantine historiography. Book II of the Miracles of St Demetrius and Fredegars chronicle give us a measure of this complexity. That no “Slavs” called themselves by this name not only indicates that no group took on the label imposed by outsiders, but also suggests that this label was more a pedantic construction than the result of systematic interaction across ethnic boundaries.

The first clear statement that “we are Slavs” comes from the twelfth-century Russian Primary Chronicle. [24. Cross and Sherbowitz-Wetzor 1953:63] With this chronicle, however, the making of the Slavs ends and another story begins: that of their “national” use for claims to ancestry.