Ethnicity is a hot topic of research at the moment, and several collections of essays have recently appeared that address questions of identity in the Middle Ages. Strategies of Distinction is part of the series The Transformation of the Roman World (TRW), the result of a series of meetings held by an international group of scholars, sponsored by the European Science Foundation between 1994 and 1997. The project looked at different kinds of changes that took place in the politics, economy, society, and culture of western Europe between the fourth and ninth centuries. [1] The period 300-800 was one in which issues of identity, at both the individual and the group level, were changing and being manipulated on a large scale, and thus ethnicity is certainly a topic appropriate for inclusion in the series.

This volume contains an introduction by Walter Pohl, thirteen papers, and a conclusion by Ian Wood, the overall editor of the TRW series (a list of the papers, with descriptions of their contents, can be found at the end of this review). Pohl, in his introduction, defines the purpose of the volume: "What did names, law, language, costume, burial rites, rhetoric, culture, royal representation or ideology mean, and to whom?...the papers assembled here....all deal with the ways how ethnic distinction became a political factor in the post-Roman world." (9) That ethnic identity did have meaning to people is assumed by the authors of all the papers; the question of what it meant, and more importantly how and why, is dealt with in some papers better than in others.

One problem often found in books of collected essays is lack of coherence between the papers. One might think that in the TRW series, broad coverage was intended, and that each volume would represent the "state(s) of the question". Instead, this volume contains a fairly typical collection of conference papers, with themes and subjects dependent on the interests of the participants. In some ways, quite broad coverage of methodologies and topics is achieved, and in other ways it is not.

The book is divided into three thematic sections. The papers in the first section, "Signs and Identities" (Pohl, Daim, Heather, Jarnut) consider different ways that people set themselves apart, or were perceived to be set apart by others; in other words, they discuss the extent to which we can understand what ethnic distinctions meant to people. The papers work well together, not least because they do not all come to the same conclusions, so a variety of ways of addressing the question are presented. The papers in the second section, "Distinction and Acculturation" (Claude, Liebeschuetz, Ripoll Lopez, Sivan, Pohl-Resl, and Kazanski) are more varied; the first four deal with the Visigothic kingdom, taking different approaches, which provides a nice geographical focus, but as a result the last two papers are somewhat out of place. The third section, "Political Rhetoric and Representation" (Harrison, Hardt, Schmauder), is also quite coherent in the sense that all three authors discuss the ways that kings were representing themselves as kings, but the relation of this topic to questions of ethnicity and identity is not clear.

Geographically, the coverage in the papers is somewhat skewed. Two of the papers are about Lombard Italy (Pohl-Resl and Harrison); four about Visigothic Gaul and Spain (Claude, Liebeschuetz, Ripoll Lopez and Sivan); four about Eastern Europe (Daim, Heather, Kazanski, and Schmauder); and three have no specific geographical focus (Pohl, Jarnut, and Hardt). This distribution is both good and bad. On the one hand, the papers on the Visigothic kingdom complement each other and paint a fairly thorough picture of ethnic issues in this territory. All four studies see a transition from some kind of separation between Romans and Visigoths, whether on the basis of language, religion, attire, or marital status, to various kinds of integration that reached fruition by the seventh century. In these interpretations, ethnicity did have objective manifestations, but they were by no means uniform or constant; rather there was always in motion a shifting ethnic frontier between Romans and Goths that became easier and easier to penetrate. This comprehensiveness is less true for the four papers on Eastern Europe, most of which emphasize archaeology, and the two on the Lombards, which deal only with texts. There is nothing specifically on Francia, except in Pohl's and Hardt's papers, and even more surprisingly, there is nothing on Anglo-Saxon England, certainly a place in which questions of individual and group identity were being played out in the period in question.

Methodologically, the authors make use of a variety of different types of evidence. Five of them deal primarily with textual sources (Heather, Liebeschuetz, Sivan, Pohl-Resl, and Harrison), three primarily with archaeological material (Daim, Kazanski, and Schmauder), and four discuss both texts and archaology (Pohl, Claude, Ripoll Lopez, and Hardt). The authors who look at texts have generally taken to heart Wood's conclusion that ethnicity is contextual; the authors look at different types of texts (histories, lawcodes, charters, royal decrees, etc.) and, for the most part, attempt to determine the context in which ethnic terms are used in them. For example, Pohl-Resl shows that ethnic terms in documents and laws before 774 do not show that there were clear differences between the ways Romans and Lombards lived and acted 'under the law'. Hagith Sivan explains that the Visigothic ban against intermarriage was politically motivated, against a Frankish enemy, not a question of ethnic purity. Harrison examines the rhetorical strategies by which Lombard kings presented themselves as Lombard, Christian, and/or Roman, and concludes that these changed over time as the nature of the monarchy changed.

One of the main aims of the TRW project was to include archaeologists as well as text-based historians [2], and this volume laudably includes several papers based on archaeological evidence. However, here the approaches to the material are not as inspired, and could have benefited greatly by the inclusion of some papers on Anglo-Saxon England, on which quite a bit has been written (for example by John Hines, Julian Richards, and Sian Jones). The authors who use both texts and archaeology treat the latter rather simplistically as material to illustrate the texts, without particularly considering the contexts in which the material was found. The exception is Ripoll Lopez, who talks about texts but then provides a detailed analysis of one Visigothic cemetery. On the other hand, her approach to the material, along with that of Daim and Kasanski, is that 'peoples' can be identified by equating items or assemblages of material culture with them. Thus, for Ripoll Lopez, "Because of the types of material, certain burials...can be considered to be typically Visigothic, while others may have belonged to the Roman population, or at least their contents are not characteristic of Visigothic personal jewellery." (172) Kazanski equates the Cernjahov archaeological assemblage with the Ostrogoths between 375 and 450. Even Daim, who at the beginning of his paper cautions that one cannot necessarily equate individual items of material culture with ethnic identity, then concludes that, after all, some aspects of archaeological evidence really can be equated with the Avars as we know them. The question of why individuals might choose to be buried with particular objects, and what objects in daily use might say about how they thought of themselves ethnically or in any other way, are not discussed. Archaeologists today tend to take a more nuanced approach to the equation of materials with identity, as can be seen, for example, in articles in the volumes Archaeological Approaches to Cultural Identity, ed. Stephen Shennan (London: 1989), After Empire: Towards an Ethnology of Europe's Barbarians, ed. G. Ausenda (Woodbridge, Suffolk: 1995), and Sian Jones, The Archaeology of Ethnicity: Constructing Identities in the Past and Present (London, 1997), among others.

Ethnicity is such a fluid and hard-to-grasp concept that the papers overall would have been better informed by some of the many modern theoretical discussions of it. Pohl, in his introduction, provides a very nice discussion of how ethnicity is used for political purposes, briefly mentioning some modern theoretical considerations, but concludes that "...the form in which ethnicity became relevant for European history were the ethnic kingdoms: ethnic identities as a basis for power and a key to privilege, and a force of integration in the new Christian kingdoms." (5) He follows up this idea in his paper, in which he denies that telling the difference between ethnic groups through external signs such as language, weapons, dress, and hair-styles (the signs mentioned by medieval ethnographers) was particularly significant. By calling into question potential objective criteria of ethnicity or identity, Pohl attempts, by a process of elimination, to bolster a subjective interpretation, that is, "we are who we and other think we are." A potential difficulty with this model, however, is that, if taken to extremes, it does not leave any place for using "objective" criteria, such as language and clothing, as a part of a subjective definition. That is to say, did a people never perceive its identity as in some way related to its dress or language? And might not, for example, the use of similar clothing by polyethnic groups serve as a means of creating identity? The wealth of examples that Pohl adduces where clothing and other objective factors do seem to have some kind of significance for the establishment of ethnicity and identity implies an inconsistency between his theoretical model and the complex situation one encounters in the literary and archaeological sources. Moreover, since politics in this period was something carried on by men, this definition of ethnicity effectively denies that it mattered to women, or to those who were not involved in politics.

Pohl's theory that ethnicity was only important for the political purposes it served is fortunately not adopted by all the authors in the book, but only Heather offers a theoretical discussion of how ethnicity might be defined that counters this definition, and indeed Heather's analysis of two 'ethnic' groups that seem to have survived over time in spite of their not having a political raison d'etre offers a compelling counter-example. Few of the papers in this book mention ideas which have received a great deal of attention in theories of ethnicity, such as the distinction between voluntary vs. involuntary feelings of identification with a particular group (Bentley's habitus), deliberate or incidental uses of external markers such as costume to represent identity (emblemic vs. assertive), tradition vs. fashion (although this is touched on by Daim in his paper), or the difference between internal and external (emic vs. etic) definitions of group identity, although all of these concepts are important for understanding what ethnicity meant, and to whom.

In general, the writers disagree regarding the degree to which objective criteria of ethnicity can be utilized. or even exist. Some (e.g. Pohl, Heather) present a theoretical, minimalist model that determines ethnicity on subjective grounds and denies to objective characteristics a great deal of significance. But others (e.g. Jarnut, Claude, Liebeschuetz, Sivan, Pohl-Risl, Hardt), and particularly those using archaeological evidence, take a more positive approach and attempt to determine the particular kinds of circumstances in which objective criteria of ethnicity (or identity) can be applied, while at the same time acknowledging that such criteria not only must be used with caution but also can change over time.

In his conclusion, Wood very nicely sums up the articles in the volume, and puts them into context: "While 'ethnic' words may apparently reflect biological groupings in some contexts (as in the case of the Rugi...), in other cases they reflect political affiliations, and in yet other contexts they may have had class implications." (299) "The scholar, therefore, needs to interpret an ethnic label within a kaleidoscope of changing discourses, while, at the same time, he or she should also identify the particular discourse which is of immediate relevance." (302) Wood also briefly mentions religion, not discussed by most of the other authors at all, although it was perhaps an aspect of identity that had considerable meaning to people at the time.

Overall, this volume provides some very interesting analyses of ethnic concepts and representations in the Early Middle Ages, although it does not represent the definitive word on the subject. It is clear that historians, archaeologists, and social scientists still have a long way to go before any kind of consensus can be reached regarding the ways in which senses of ethnicity and identity are created and manifested.

CONTENTS

Walter Pohl, "Introduction: Strategies of Distinction," pp. 1-15.

Pohl's introduction establishes the historical and cultural context of the volume and introduces the questions to be considered. Pohl suggests that during the Principate, "most inhabitants of the Mediterranean world, and many in the northern world beyond it, strove to be Romans", but that "from the late fourth century onwards, ethnicity began to return..." (1). He sees the success of barbarian peoples such as Vandals, Visigoths, Burgundians, Franks, and Ostrogoths as deriving from the ability of their leaders to emulate Roman military and administrative practices. As a consequence, form followed function: "maybe it was an even greater achievement that those who had come to rule Roman provinces managed to become a people" (2-3).

Ethnic discourse provided a means by which a group could establish its own identity: "Names, narratives, and laws affirmed the separate existence of an ethnic group with an exclusive claim to power over certain parts of the ancient res publica" (2). Pohl doubts that there was a great amount of blood connection among the people in these groups, suggesting that the number of actual "Ostrogoths" in 493 amounted "quite likely to considerably less" than 1%: "Their polyethnic basis was... transformed into a singular ethnic identity, expressed in the name of the kingdom" (3). He suggests that Roman use of ethnic names probably "corresponded to widespread feelings of identity among barbarians" that "were mostly rooted in small, face-to-face groups... who came together regularly for war and political business," not in "the large ethnic communities that late Romans called Franks, or Goths", which "are highly abstract, culturally constructed ways of categorising people who might differ a lot... and might not be so different at all from people who do not fall into that category" (4). It was this latter kind of ethnicity that "became relevant for European history, with "ethnic identities as a basis for power and a key to privilege, and a force of integration" (5). Against this background, the subsequent contributions all deal, in their own ways, with how elements of ethnicity and identity were manifested among the various barbarian peoples who settled within the boundaries of the old Roman Empire.

*Signs and Identities*

Walter Pohl, "Telling the Difference: Signs of Ethnic Identity," pp. 17-69.

Pohl, begins by asking, "How can peoples be distinguished?" (18). Pohl suggests that Tacitus' "flexible handling of a set of criteria for ethnic identity" has led modern scholars to accept a model in which "language and culture repeatedly appear as decisive criteria" (19). He then argues not only that this method is based on the impressions of the observer, and not necessarily on self-perception, but also that ancient accounts of ethnic differences are not as simplistic as they seem: "Rather than simply reflecting a world of consistent ethnic diversity, they are traces of a complex communication about communities on the periphery of the ancient, and later christian, world" (19).

Pohl questions the idea that one can "define ethnicity ... by objective features like language, culture and customs, territory or political organization," and proposes rather to focus on "the subjective factor", suggesting that "strategies of distinction have to convince both insiders and outsiders that it is significant to be different" (20-22). In support of this thesis, Pohl uses a deconstructionist methodology to discredit potential objective criteria for establishing identity and to demonstrate that "language, culture, political organization, etc., do not correlate completely" (20). The use of language and attire as determinants of ethnicity are countered by citing examples of cases where they were not such. For example, the considerations that some individuals were multi-lingual, or that different peoples sometimes spoke the same language, are cited as evidence that language could not be an objective indication of ethnicity. Pohl does acknowledges that "it was among the barbarians where the diversity of languages made itself felt" (24), but he ultimately concludes that "we have no evidence that ... language was used to ... define an ethnic group" (27). (and in this regard, there is no reason to suppose that Sidonius' friends Syagrius and Burgundio are the same person, as on p.26).

In a like manner, Pohl invalidates other potential objective criteria of ethnicity. Regarding weapons, he concludes (39), "early medieval elites were too international in their outlook and too flexible in their tactics to express ethnic 'corporate identity' in their military equipment..." Attire is rejected by analogy with the inappropriateness of modern-day "national costumes"; "Similarities in the way larger groups dress" are dismissed as "scholarly constructs that should not be confounded with real social groups" (40-1). Pohl also questions archaeological evidence, suggesting that "typologies established by modern archaeologists do not necessarily reflect late antique perceptions", and that they "often ... do not correspond to ethnic units ... but to broader populations" (41).

Nevertheless, Pohl does concede that "sometimes, archaeological cultures seem to correspond to ethnic categories" (42), as in the case of the Avars. But this correspondence is challenged on the basis of "the polyethnic structure of the Avar realm," because Avar items found in a grave "might as well have belonged to a person who spoke Slavic and considered himself a Bulgar" (42). Pohl posits that clear-cut use of "costume as an ethnic sign...", such as the Lombard use of white stockings, "has to be discussed, as far as possible, in the context of symbolic codes and its social function" (42). Other evidence for Lombard dress as a sign of ethnicity is disregarded because "it had changed without any visible consequence for the identity of the Lombards" (44).

Pohl does acknowledge that "there is no doubt that dress could be regarded as ... from a Roman point of view, barbarian" (47), but the observation of Isidore of Seville that people were recognizable by their differences is rejected as "obviously hardly supported by many practical observations" (47). Pohl suggests that such characteristics all could be perceived as being generically foreign or barbarian, but that they cannot be rigorously identified with any particular kind of named ethnic group, such as Franks or Visigoths. He also recognizes that "Late Antiquity and the early Middle Ages saw some fundamental changes in the way people dressed" (49), but sees distinctions in dress as reflective more of social than ethnic distinctions: "Innovation in dress often expresses more fundamental changes in society..." (49). Other stylistic elements that have been seen as indicators of ethnicity, including belts and hairstyles, are likewise dismissed: at most, they only "characterized specific groups of warriors" (55). Only during the Carolingian period, Pohl argues, did there develop "the concern ... with dress codes and their specific ethnic context" (45). He concludes, "barbarian rulers had to be ambiguous, and flexible handling of ethnic identities had enabled them to survive in the first place" (68).

Falko Daim, "Archaeology, Ethnicity and the Structures of Identification: The Example of the Avars, Carantanians and Moravians in the Eighth Century," pp. 71-93.

Daim provides an archaeological perspective that one can contrast with the historical one provided by Pohl. It commences with a lengthy discussion of terminology and a review of recent historiography, acknowledging that "a sensitive human being cannot approach the topic of ethnicity impartially" (71), and suggesting, "In our conceptions, ethnicity is a category of extraordinary significance. In the early Middle Ages, however, its role was rarely one of great importance..." (72). Daim prudently suggests that there never is a perfect fit between one's personal feelings about one's ethnicity (which cannot be reconstructed in any event, and which can vary depending on circumstances) and the objective manifestations (as in dress, language, etc.) of these feelings. As a result, it is unwise to attempt to define too closely the significance of objective factors. Characterizing the views of some historians, he suggests, "Instead of searching for 'objective' criteria for ethnicity, the historian now takes a minimalist view and strives to discover 'under welchen Umstanden die Ethnizitat Uberhaupt eine relevante Kategorie darstellt" (74). But for him this raises another question: "If we wish to comprehend society as an interlocking system of groups, which of them could then be labelled 'ethnic'?" (75). He ultimately accepts the definition of "ethnicity as a phenomenon of social psychology" (76), and for the description of archaeological material, he chooses a less contentious term, "identity."

Turning to the archaeological remains, Daim introduces a "model of the structures of identification," in which "we consider each cultural phenomenon with regard to its social .. function... with the aim of reconstructing ... a web of conscious or half-conscious identifications" (80). Considering the Avars, Daim analyzes nearly 50,000 eighth-century graves, which "show a multitude of strong connections between one another, so that it is acceptable to speak of a single culture" (89). He also proposes that the Moravian introduction of a hooked spur reflects "the creation of a Moravian ethnicity" (92). Daim demonstrates, therefore, that material culture can provide an objective basis for gaining insights into a people's ideas about its own identity, although he realizes that by doing so contradictions are raised with the subjective model of the social scientists: "If ethnicity is really a phenomenon of 'social psychology' ... this would imply that we are over-stressing our material by a long way" (93).

Peter Heather, "Disappearing and Reappearing Tribes," pp. 95-111.

Heather discusses the way in which "certain social units -- one might wish to call them "tribes," "ethnic groups", or something else" disappear and then reappear in the sources. In these cases, a group "survived a period of political domination by another group, seemingly without having lost all sense of solidarity, cohesion, and identity" (95). As two case studies he identifies the Heruls, who vanished from history in the late 260s, only to reappear with the Huns in the mid fifth century, and the Goths, some of whom likewise were amalgamated into the Hun horde. Both peoples reestablished their independent existence after the collapse of the Huns in 455. As another example of how a people could maintain its identity under such conditions, Heather points to the Rugians, who, having been absorbed by the Ostrogoths, "had absolutely no intercourse with women other than their own..." (99: Procop., Bell.goth. 7.2.1-3).

Heather first suggests that "identity fundamentally resides in the perception of the individual... rather than in concrete measurable things" (104). Identity therefore becomes "malleable", "manipulable", and an "evanescent situational construct" rather than an "inherited unchanging fact" (104). But he also concedes that identity is not always "something voluntarily or consciously assumed...", and that sometimes "inherited senses of identity do indeed dictate behaviour" (105). As a result, he sees "no reason to disbelieve" sources that attest to an "strong inherited sense of identity" (105). It is this feeling of identity that seems to survive during periods of political subjugation, which can disrupt political continuity, and lead to the fragmentation of a people into different sub-groups, which can adopt "different strategies for survival" (109). Nor are all groups equally keen to maintain their distinctness: the Slavs, for example, allowed prisoners to "join their groups as fully equal members" (110) after one year. On the other hand, Romans captured by Avars maintained their identity over two generations. Heather concludes by stressing that "historical narrative ... is one of the most fundamental signs of ethnic identity" (111).

Jorg Jarnut, "Nomen et Gens: Political and Linguistic Aspects of Personal Names between the Third and the Eighth Century-Presenting an Interdisciplinary Project from a Historical Perspective," pp. 113-116.

Jarnut's contribution weighs in at just over three pages (113-116). Jarnut discusses a project designed to "gain safer criteria than hitherto of what indication names give of the origin of the bearer" (113). The project proposes to classify "ethnic specifics... so that the bearers of individual names can in future be associated with single gentes with a higher degree of certainty" (115). Name-giving processes also will be investigated. The pragmatic, objective, and statistical approach of this project is predicated on the presumption that objective criteria of ethnicity do exist, and, when treated with appropriate caution, can provide insights into ethnic identity. Jarnut's statement that "both positive and negative generalisations are problematic" (114) carries with it a rejection of the presumption that objective criteria cannot be usefully applied to issues of ethnicity.

*Distinction and Acculturation*

Dietrich Claude, "Remarks about Relations between Visigoths and Hispano-Romans in the Seventh Century," pp. 117-130.

Claude begins by asking "where contemporaries saw characteristic differences between ethnic groups" (117). Possible distinguishing factors mentioned by Isidore of Seville include dress and language, and "have contemporary value" (118), although on the basis of archaeological evidence, Claude also concludes that "towards the end of the sixth century at the latest, dress ceased to be a distinguishing feature between Visigoths and Hispano-Romans" (120). Although by this time the Gothic language had dropped out of use, Roman "inhabitants saw the Gothic ethnicity as an important distinctive feature when it came to name-giving" (122); like Jarnut, Claude accepts that ethnicity can be established on the basis of nomenclature (e.g. Glismoda: pp.123, 125). Claude repeats the rather out-of-fashion assumption that the Code of Euric and the Roman Law of the Visigoths represented "personal law" (123: on this issue, see Liebeschuetz below), and consequently the territoriality of Visigothic law probably can be moved back well before "the middle of the seventh century" (123). He sees a steady movement toward "a rapprochement between Goths and Hispano-Romans" (124). He suggests that ultimately, "the only difference between both peoples .. lay in an aspect of social behaviour... a Germanic tradition to use 'leading names' (Leitnamen) within each family and to hand down parts of names" (125). But, in apparent contradiction to this conclusion, he also sees "the aristocracy as bearers of a particular Gothic consciousness..." (127).

An apparent problem with Claude's view of rapprochement and integration arises with the seventeenth canon of the Sixth Council of Toledo in 638, which states that when a king died, "nullus ... extraneae gentis homo, nisi genere Gothus ... provehatur ad apicem regni" (127), a statement that seems to indicate that at least in this regard ethnic descent continued to be of great significance. But, observing that "this appears to be doubtful in the light of our preceding discussion" (127), Claude attempts to refute this apparent prohibition on Romans becoming king by defining "extraneae gentes" as "ethnic groups that lived outside the Visigothic kingdom." But the sources on which this argument is based both specify not "extraneae" but "externae gentes" (Lex Visigothorum 9.2.8: AD 673; Julian of Toledo, Historia Wambae regis 28: AD 672), such as "the Franks or the Aquitanians." In legal contexts, terminology meant something. The usage of extraneae here may, rather, be akin to that involving the extraneae mulieres with whom clerics were forbidden to fraternize, in the sense of those with whom one did not have a close, and in particular a familial, connection. It is difficult, therefore, to explain the requirement that the Gothic kings had to be Goths without having it mean something close to home.

In other regards, too, Claude denies the Gothic kings their ethnicity by suggesting that they had "placed themselves in a Roman tradition at an early date," by, e.g., using the title "Flavius" (129), and he concludes by seeing the term "Gothus" as having social rather than ethnic significance: "it was used for any member of the kingdom of aristocratic origin" (129). But this line of reasoning seems to have an inconsistency: it permits continuing objective indicators of ethnicity to the Romans (as in the title "Flavius"), but denies them to the Goths. In this model, Goths who adopted elements of Roman-ness perforce lost their Gothic-ness. And this in the face of other late evidence (buried in a note in another section of the study) for continuing distinctions between Romans and Goths, as in Lex Visigothorum 9.2.9, "seu sit Gothus sive Romanus," which is dismissed on the grounds that "legal language is always conservative", and because it is simply "an example for jurisdictional perfectionism than for survival of a conscience of ethnic diversity" (130 n.87).

Claude argues that "the ethnic identity of the Visigoths was under a permanent process of erosion... which led to an almost complete assimilation into the Hispano(tm)Roman environment. The barriers that costume, confession, law and language had once formed, were gone... What remained were ... family traditions" (130). Only at the very end does Claude acknowledge that the process of assimilation was a two-way street: "the Hispano-Romans ... took on important ideas of political order from the Goths." Ultimately, "this process resulted in the formation of a new Hispanic 'nation' in the late seventh century" (130)

Wolf Liebeschuetz, "Citizen Status and Law in the Roman Empire and the Visigothic Kingdom," pp. 131-152.

The following contribution, by Wolf Liebeschuetz, "Citizen Status and Law in the Roman Empire and the Visigothic Kingdom" (131-152), considers the often neglected question of the nature of Roman citizenship during the late Empire. Liebeschuetz first surveys the original difference between citizens and non-citizens, which could be ill-defined, in practice if not in law. This distinction disappeared with the Constitutio antoniniana in AD 212, to be replaced by that between honestiores and humiliores, and the appearance of the status nomina Flavius and Aurelius, respectively (135). The term peregrinus came to refer to "an individual living in a city or province other than that of his registered origo", and politeia to "groups of non-naturalized barbarians who had acquired the right to live within the Empire" (135). The distinction between slave and free also gained added significance, and the honestiores became fragmented into different groups with their own special statuses. There were a "multitude and refinement of status distinctions", in the face of which "A sense of Roman identity survived but it had been depoliticised... The population of the late Empire lacked a strong sense of ... shared interest" (136-7).

There also was a legal differentiation between "Romans and barbarians" (137), although this issue seems to have been of little practical significance: "For the upper classes, what distinguished Romans from barbarians was literary culture" (136), and barbarians "never... demanded to receive the rights of Roman citizens" (138). Barbarians were, however, fitted "into one or another of the hierarchy of status groups...," and if citizenship were a pre-requisite, it was granted "as a mere formality" (138). Continuing distinctions between Romans and Goths are seen in the prohibition in the Gothic kingdom of marriages between Romans and barbarians, which remained in force until overturned by Leovigild (568-586). The explanation for such a marriage bar "is surely that maintenance of ethnic cohesion, one might almost say of national identity, was a matter of the greatest importance for them..." (140); it was needed "to preserve the ethnic consciousness and cohesion of a group threatened with disappearance through assimilation" (141). And as for "what constituted a Goth," an issue not discussed in the law codes, Liebeschuetz proposes that "the kings reserved for themselves the right to decide whether an individual was a Goth or not" (141). Turning to the issue of governance, Liebeschuetz proposes that it was unlikely that Goths and Romans had separate laws, for "the conflict of laws would have raised difficult questions" (142). The Code of Euric "was valid also for Romans... a king who was the head of state of both peoples is not likely to have spoken with two voices about the same legal issue" (142-144, cf.147). Liebeschuetz concludes that "by the seventh century there will have been very little to separate Goths from Romans" (148): burial customs that had been distinct at the beginning of the sixth century were abandoned in the seventh (149).

Yet, the Goths still preserved a "sense of nationhood, or ... their ethnic identity... Nobody but a Goth could be king of Spain" (149). Liebeschuetz also makes the point that even if "actual differences...disappeared, symbolic differences such as personal names, and the titles of certain public offices (saio, triumphadus, gardingus)...became more important" (150). He proposes that Isidore of Seville "does not suggest that in the Gothic kingdom...the difference between ruling people and ruled was being eliminated" (150), an opinion rather different from that of Claude, who concludes that "Isidore understood Goths and Romans as equal ethnica in the regnum Gothorum (130). At the same time, nomenclature ceased to have an ethnic association, and, for Romans, Germanic names became more fashionable (150-1). In this regard, Liebeschuetz expresses surprise "that a culturally more advanced majority population should take on the 'national' -- or 'ethnic' -- identity of a ruling minority..." (151), but ultimately concludes that, given the new political realities, there was "little reason why a Roman should not wish to identify with the gens et patria Gothorum" (152). Consequently, Liebeschuetz's conclusion is consistent with that of Claude: by a process of assimilation of aspects of each other's cultures, a single society with an amalgamated corporate identity developed in Visigothic Spain.

Gisela Ripoll Lopez, "The Arrival of the Visigoths in Hispania: Population Problems and the Process of Acculturation," pp. 153-187.

Ripoll Lopez begins by citing "a clear historiographical tradition ... of the existence of two large groups [viz. Goths and Romans]...apparently confronted by a question of origins and culture" (153). Ripoll Lopez proposes to reinterpret literary evidence and analyze new archaeological material in order "to qualify the traditional view...as regards differentiation and confrontation" to show "a mutual process of acculturation" (154). She provides a useful description of the great ethnic diversity in late antique Spain, including communities of eastern origin (Greeks, Jews, and Syrians), and Africans, and a detailed survey of size and location of elements of Roman (7-12,000,000) and Visigothic (130,000) populations. Burials show "the gradual abandonment of the typically Visigothic form of dress..." (164), an observation that necessarily contradicts the suggestions made earlier in the volume that such identifications cannot be made with any certainty.

Ripoll Lopez then turns to Leovigild's repeal of the ban on intermarriage (Lex Vis. 24.3.1.1, "sancimus ut tam Gotus Romanam quam etiam Gotam Romanus si coniugem habere valuerit .. facultas eis nubendi subiaceat."), which "has been used as an instrument to justify supposed national unity" that would incorporate both Goths and Romans (164). Among other things, this ordinance indicates that at the time there continued to be a distinction between Romans and Goths. Ripoll Lopez suggests, "This change encouraged the mixing of the population and accelerated the acculturation process...the newcomers became fully integrated within the Roman social, economic and cultural structure"; ultimately, "differences...from the point of view of material culture were to be almost imperceptible" (165). Consistent with the observations of Heather about reassertions of identity above, she also suggests that subsequent to integration "names of Germanic origin re-emerged, certain signs of identity thus being recovered" (166).

The remainder of the study looks at objective criteria of ethnicity as manifested in grave goods. Burial patterns indicate that "the foundational, Visigothic nucleus was governed by a kinship system and only gradually gave way to the monogamous family structure...." This phase was followed in the sixth century by an "occupation of burials that can no longer be defined as strictly speaking Visigothic, but which instead incorporate elements of a Roman nature" (178). Here, again, one sees the merging of what had been two different ethnic groups, initially distinguished by different customs and material culture, into a more integrated aggregate.

Hagith Sivan, "The Appropriation of Roman Law in Barbarian Hands: 'Roman-Barbarian' Marriage in Visigothic Gaul and Spain," pp. 189-203.

Sivan expands in detail upon the Visigothic ban on intermarriage discussed by Liebeschuetz and Ripoll Lopez. Sivan suggests that in such legislation "one expects to find traces of Germanic customs," but concludes that the Goths "appear to have faithfully followed Roman patterns" (190). She suggests that past interpretations assume that the ban resulted either from a Gothic attempt to maintain their "national purity" (as Liebeschuetz above) or from "religious antagonism" between the Arian Goths and Nicene Romans, and "imply that the Goths had a specific self-image which required complete separation from 'the other'" (190).

Her discussion focuses on the identity of the "barbarians" whose intermarriage with Romans was forbidden in an interpretatio of Codex Theodosianus 3.14.1 found in the Visigothic Breviarium Alarici (3.14) of AD 506. Past analyses have assumed that these barbarians had to be Visigoths, but Sivan wonders, "How likely ..were the Goths to exclude themselves from marrying locals in their own Gallic dominions?" (194). Such a question of course presupposes that the ban was not an effort to maintain the Goths' ethnic identity. Sivan then considers Procopius' account of intermarriage between Franks and Arborychi (i.e. Armoricans), and concludes that the barbarians who were prohibited intermarriage were not Goths, but Franks, and that the ban "is best perceived as a temporary measure taken under pressure" of an impending Frankish invasion (199).

Sivan then turns to the repeal of the ban, perhaps issued in AD 580 by Leovigild, in which it became permitted for Goths to marry Romans. But here Sivan's Frankish hypothesis runs into potential difficulty, for if the law being annulled is that of the Breviarium, as Sivan argues it is, then it is clear that the barbarians of the Breviarium were thought to have been Visigoths, not Franks. Sivan deals with this obstacle by supposing that "Leovigild was unaware of the circumstances that had prompted" the earlier ban (201), for in 580 "the ban made sense only as a prohibition on marriage between Goths and Romans" (202). She also argues that a de facto ban on intermarriage would have continued as long as the Visigoths remained Arian and any ecclesiastical prohibitions on marrying heretics (such as that found in a Spanish canon appended to the canons of the Gallic Council of Agde of 506) remained in effect. By the time of the repeal, moreover, "the law was understood as an instrument of class distinctiveness and not of voluntary religious separation" (202).

Ultimately, Sivan returns to the orthodox interpretation of the ban on intermarriage as a means of maintaining "national purity," hypothesizing that "the Gothic rulers ... may have regarded the ban as a useful tool to maintain the superior position of the Gothic minority in Spain," and as "an expression of ethnic diversity" (203). It was repealed only "Leovigild contemplated a united Spanish kingdom, with both Goths and Romans loyal to their Gothic monarch" (202).

Brigitte Pohl-Resl, "Legal Practice and Ethnic Identity in Lombard Italy," pp. 205-219.

Pohl-Resl looks at Lombard law codes and charters to show "how everyday legal practice influenced and changed the laws," and wonders about "the role of ethnic identity .. in this context" (205). She proposes to look for ethnic terms, and to consider the role that ethnic identity had in legal practice as evidenced by the distinction between Romans and Lombards. She suggests that ethnic terms could have either geographical or chronological significance. They also could be ambiguous: "Romani," for example, could refer either to Roman subjects of the Lombard king, or to Byzantines. She also suggests that, unlike some other barbarians, the Lombards maintained two separate legal systems, although "people were free to choose between these two law-codes". In matters of inheritance, however, a person was expected to "stick closely to his or her law" (209). In some cases, ethnic identifications were explicitly stated, but nomenclature (at least in the examples cited) gives little clue to what this identity was.

Over the course of the eighth century there was "more and more adoption of Roman legal traditions" (211). Some Roman legal practices survived in the church, although others, such as the keeping of civic archives, disappeared. By the late eighth century, as a consequence of Frankish rule," ethnicity began to matter even in monasteries," and in the ninth century, "legal practice and ethnic identity were closely connected" (219). After citing numerous examples in which ethnic identity seemed truly to matter, Pohl-Resl concludes that they "should not be taken to reflect any strong sense of ethnic identity" (ibid.).

Michel Kazanski, "Le Royaume de Vinitharius: Le Recit de Jordanes et let Donnees Archeologiques," pp. 221-240.

Kazanski returns to the Goths, and considers a group of Ostrogoths, ruled by a certain Vinitharius, who were under the domination of the Huns in the late fourth and early fifth century. Kazanski proposes to examine the question of what territories were occupied by these Goths. The historical record places them between the Don and the Dniestr. Archaeologically, much of this area was occupied in the fourth century by the Cernjahov culture, which is identified as representing the Goths, but which often has been thought to have disappeared in the late fourth century, coincident with the arrival of the Huns. Kazanski, however, suggests that "tout cela se revele faux" (223), and that theCernjahov culture can be shown to have survived in full form through the first half of the fifth century. The arrival of the Huns, in fact, left few clear manifestations on these sites: "La culture materielle de la civilisation de Cernjahov ne montre pas non plus de changements notables" (224). What does appear, however, are items of foreign origin, such as ceramics identified with the Alans to the east and the Przeworsk culture to the north. This contribution provides a striking archaeological example of the phenomenon of submergence and recovery discussed in Heather's study above.

*Political Rhetoric and Representation*

Dick Harrison, "Political Rhetoric and Political Ideology in Lombard Italy," pp. 241-254.

Harrison returns to the Lombards, and proposes to investigate "rhetorical and ideological aspects of the Italo-Lombard monarchy (568-774)" (241), looking for Christian and Roman influences in evidence that includes legal material, ceremony, coinage, royal insignia, and buildings. A discussion of legislation and charters picks up on the topic of Pohl-Resl. Harrison suggests that the prologue to the Edict of Rothari in 643, which cited eleven generations of Rothari's ancestors, "is in fact one of the clearest manifestations we have of non-Roman political self-consciousness in the so-called successor states to the western Roman Empire", and that "Rothari and his councillors consciously inserted this list instead of...making an attempt to place Rothari within a system that was explicitly Roman and imperial" (242). Like Pohl-Resl, Harrison sees a growing Roman influence: Liutprand, for example, "introduces a Christian ideology" (243), and Ratchis' purpose "is very Christian" (244); the latter even attempted to put Rothari "within his own catholic world... not mentioning the fact that Rothari had been of the Arian faith" (244). With regard to ethnic terminology, all the kings referred to themselves as kings of the Lombards, but "the problem is what the word Langobardus really meant," for "there was a shift from the sixth to the eighth century from an old tribal meaning to one with a more socially and economically charged significance" (245).

Matthias Hardt, "Royal Treasures and Representation in the Early Middle Ages," pp. 255-280.

Hardt proposes to use the evidence of both literature and material culture to look at the role played by treasures in the early Middle Ages. He notes that past scholarship has seen "the treasure of barbarian kings as important basis [sic] for their rule", and that images of hoards and treasures in heroic poetry might "reflect actual circumstances of the migration period" (256). Citing several military campaigns of Franks and Byzantines, he suggests that "royal treasures played a particular role in the wars.... in a sense they constituted war aims..." (260). Possession of treasures also were important in succession settlements, and in the creation of marriage alliances. Hardt enumerates the different sorts of things that could be considered as "treasure," including, along with gold, silver, jewelry (including crowns), and precious stones, such things as clothing, tableware, weapons, books, and documents.

As for the sources of treasure, Hardt cites plunder and "extensive looting campaigns", "tribute that was transferred between gentes", gift exchange (including the de rigueur gifts that ambassadors were expected to present), fines, bribery (and presumably extortion), the confiscation of property, grave robbing, treasure trove, and mining. It is difficult to assess, however, how much income accrued from taxes and tolls. Hardt then turns to the means of storing treasures, as in treasure chambers and fortified towns. As for expenditures, treasures could be useful "to ward off disaster or keep potentially plundering neighbours at bay" (277), for distribution to one's followers, for bequests to churches and monasteries, and for household and ceremonial expenses. Hardt concludes that "the royal treasury ... was the instrument by which royal rank was determined..."; it "represented the pinnacle of an economic system to which broad sections of the barbarian conquerors .... were still committed" (279-280). Implicit throughout, although not explicitly stated, is the assumption that the possession of treasure was one of the things that maintained the identity of a barbarian king.

Michael Schmauder, "Imperial Representation or Barbaric Imitation? The Imperial Brooches (Kaiserfibeln)," pp. 281-296.

Schmauder considers four brooches known as "Kaiserfibeln" for their similarity to fibulae worn by Roman emperors. Their most significant element is two or more hanging strands of pearls, precious stones, and golden pendants. After a detailed discussion of the brooches, Schmauder concludes that three of them are of Roman workmanship and "represent the bestowal upon barbarian rulers of insignia which were certainly very similar to the imperial costume, if not in all elements and design" (293). The fourth, however, from Pietroasa, was "certainly manufactured in a barbarian workshop," and Schmauder suggests that the "presumptuousness" of this requires explanation (293). He concludes that, unlike such rulers as Theoderic the Ostrogoth, barbarians living outside the boundaries of the old Roman Empire were not "restricted by numerous legal and customary restraints" and were free "to imitate late Roman forms of representation of the imperial court" without hindrance (295-296).

Ian Wood, "Conclusion," pp. 297-303.

NOTES:

[1] This compilation is one of several resulting from the work of a sub-group of the project that dealt with the topic of "Imperium, gentes et regna." A previous volume emanating from the same sub-group was W. Pohl ed., Kingdoms of the Empire. The Integration of Barbarians in Late Antiquity (Leiden: E.J. Brill, 1997), which includes (ix-x) a foreword that also is appropriate to the volume being reviewed here.

[2] See Ian Wood, "Report: The European Science Foundation's Programme on the Transformation of the Roman World and Emergence of Early Medieval Europe," in Early Medieval Europe 6 (1997): 217-227.