Featured post

More Posts you might have missed on the other site

Here, in order from oldest to most recent are the not-exactly-numerous posts that have appeared on the other site in the past two and a half...

Tuesday, 8 December 2020

Ethnicity in Late Antiquity (5): Ethnic transformations in the West, c.600

As we have seen so many times already, major change took place in the late sixth century. The most important aspect of this was the spread of non-Roman identity. By the middle quarters of the seventh century everyone who mattered in the north of Gaul was a Frank; something similar seems to have taken place in Spain, with Gothic identity; the Lombard Laws don’t seem to feel any need to deal with Romans other than those from outside the kingdom; The early Anglo-Saxon law-codes similarly pay little attention to Romano-Britons or Welsh. Furthermore, by this time the earlier situation where ethnicity in the law was generally limited to adult males had come to an end. The types of ethnicity with which the law was concerned were now held by adults and children, men and women. It ran in families.

What’s more, when the seventh-century law-codes do talk about Romans they are now firmly in a lower legal position. This is very clear in early Anglo-Saxon law-codes’ references to the Welsh, who are now often even slaves, or at least unfree. Whereas 6th-century Salic Law had envisaged the Roman population of northern Gaul as a roughly parallel population to the Franks, even if the latter were privileged, seventh-century Ripuarian Law saw Romans within the Ripuarian territory as essentially half-free, and requiring a Ripuarian Frank to speak for them at law.  In other words, all of those among the free population who hadn’t managed to make an effective claim to a non-Roman identity had effectively missed the boat completely.

Even in areas like the south of Gaul, where non-Roman identities didn’t become dominant, we can still see a shift to Germanic personal names by the seventh century. In the church, too, whereas Roman names had predominated before around 600, after that date Germanic names became more common. It is difficult, once again, not to link all this to the responses to Justinian’s wars and his ideology of reconquest. Roman identity lost whatever was left of its cachet. It’s difficult not to link this with the shifts in ideology we’ve already encountered, and with the final triumph of the martial model of masculinity, as well as with the changes in the organisation of the state, taxation and military service that I discussed in [a previous lecture that I hope to upload in future].

One of the other developments around 600 was the introduction of the idea of the personality of the law. It is often said that after the collapse of the western Empire law became personal. In other words, if you were a Frank you were tried according to Frankish Law; if you were a Roman you were tried according to Roman law, and so on. Since about 1980 it has been shown increasingly that post-imperial western law-codes were territorial, not ethnic as had been believed: in other words they applied to everyone in the kingdom rather than just to the Franks or the Goths. This changed around 600. To my knowledge the first mention of the personality of the law – i.e. the right to be tried by the law of your people – comes in the Frankish 7th-century code, Ripuarian Law, which explicitly states that people of particular non-Ripuarian Frankish identities should be tried according to their own law. Obviously (or to me it seems obvious anyway) this can only have been an aristocratic privilege, so this would be something else associated with the emergence of a more secure social elite around 600 in the north-western areas of Europe. The one place where personality of the law didn’t seem to apply was Iberia, where seventh-century Visigothic law said that no other code was applicable within the kingdom. I suspect this is another example of the universalising claims of the Visigothic kingdom – its claim to be the legitimate heir of Rome.

Related to these developments was the creation of a range of new ethnic law-codes in the first decades of the seventh century. As well as Ripuarian Law, which seems to have been envisaged as the code for the eastern Frankish kingdom of Austrasia, the reign of the Frankish king Chlothar II also saw the drafting of law-codes for the Bavarians, Alemans and other peoples under Frankish hegemony. It is possible that the Kentish laws of Æthelberht are another example of this.

By the second quarter of the seventh century, then, not only were kingdoms made up of different ethnic groups but the elites within those groups could take their legal identity with them if they moved around.  We have seen that in the Roman Empire Roman ethnicity was bound up with the process of socialisation and of being a man; everything about your identity was crucially linked to your membership of the Roman body politic. By about 625 there seems to have been no part of the western world – or perhaps even the Eastern Empire – where this was still the case. This too must surely have meant an undermining of the old late antique state as discussed in [previous lectures that I hope to upload in future].

Ethnicity in Late Antiquity (4): Ethnicity in the post-imperial kingdoms

Functional Ethnicity

In the fifth and sixth centuries the most obvious feature about at least some types of ethnic identity was that it referred to function within a kingdom. As we’ve seen, by the start of the fifth century, the Roman army had adopted a range of barbarising identities, which I have called the martial model of masculinity. Indeed, as we saw, the army probably included more barbarians than had earlier been the case, even if probably not to the extent that used to be believed. Certainly, and importantly, more non-Romans were able to reach its highest ranks. We have also seen that in the fifth century, politics increasingly became focused upon the rivalry between different regional factions based around local armies, increasingly made up of people who claimed a barbarian identity.

Consequently it’s not surprising that across the Western kingdoms we can find a situation where the army was made up of ‘barbarians’, which is to say people claiming a Frankish, Gothic, Burgundian or Lombard identity. It is very likely in my view that Angle and Saxon identities played the same role in lowland Britain at the same time, though there’s no real way of knowing. On the other hand, the Romans, as I have mentioned in earlier lectures, staffed what remained of the civil administration and the church

In the 6th-century law-codes, what is striking is that an ethnic identity is generally only assigned to male adults. This seems to underscore that this was an identity that was generally only deployed in particular contexts, and that it might be something that was acquired rather than simply inherited. In other words you’d have to grow up and serve in the army to be a Frank like your father. The range of issues where ethnicity is discussed is actually pretty low and generally encompasses interactions or disputes between ‘barbarians’ and Romans, issues of inheritance – presumably because of the link between this and military service, or tax-paying.  Another clear feature of the 6th-century law is that the ‘barbarians’ have a higher wergild than the Romans, a wergild being the sum paid to the family of the deceased by their killer. This may well have been a continuation of a higher status for soldiers; it’s difficult to say. The Roman and Barbarian populations are broadly parallel but we can see an increasing privilege attached to being a barbarian: tax-exemption, higher legal standing, other privileges.

Ethnic Change

One of the other debates about late antique identity concerns whether one could change ethnic identity and, if so, how easily. At one extreme, an American historian called Patrick Amory – who later went off to become a hot shot record company executive – really – argued that because of the functional role just mentioned, ethnic identities were little more than badges that could be chosen and swapped without much difficulty. At the other, Peter Heather responded showing that that ethnicity wasn’t so easily changed and argued for a much greater reality to ethnic identity and groupings. The problem is that both failed to appreciate two things. One was the sort of plurality of identities in the ‘ethnic’ sphere that I discuss in my 2007 chapter, so that ethnic change wasn’t a simple one-for-one straight swap but the acquisition of another identity – as when a Roman joined a barbarised army unit. Related to do that was the element of time. Heather was absolutely right that there is little or no evidence of simple adoption and discarding of ethnicities but people nevertheless did change ethnic identity; it is just that it was a longer process than Amory suggested, taking a generation or more. One person might adopt a Gothic identity in addition to his other identities but then, over time, stress it more and more until it was the one that he used most often, and his children might in turn see themselves more unproblematically as Gothic – especially if they had inherited his military obligations, tax-exemptions and so on.

Signs of ethnic identity in the 5th and 6th century

It’s probably not surprising, given that there seem to have been no deep-seated unifying features within ethnic groups, that all sorts of material cultural and other signs were employed to try to provide that sort of unity, or to proclaim an ethnic allegiance. One that receives quite a lot of contemporary comment is costume. In [an earlier lecture, which I hope to post at some point] I mentioned that there are regional variations in fifth- and sixth-century female costume across the west and that these have been linked to ethnicity. There are problems with this. One is the point just made that ethnic identity in the laws seems not to have been something assigned to women, except in contexts of whom they were married to. If there was an ethnic component to this costume, maybe it actually related to their husbands or fathers. Another problem, though, is that closer study shows that differences in female costume were often related to age and position in the life-cycle. Some of these differences might simply be regional and while, as I argued in my book, that might be just as ethnic as identities related to membership of ‘peoples’ it would relate to a different level of ethnic identity from the one assumed. The linkage between ‘barbarian’ identities and military service might have led to the carrying of weapons as a sign of ethnic distinction.

Hair-styles were another means of showing ethnic identity. Sidonius Apollinaris describes the Frankish hair-cut, which had the back of the head shaved, and the hair put into a top-knot and thrown forward. Like this:

Paul the Deacon describes the Lombard hair-style, which was also shaved at the back but grown long and parted in the middle. Lombard men also wore long beards to mark themselves out as Lango-bardi - long-beards. Gregory of Tours mentions a Breton hair-style though he doesn’t describe it. And so on.

One of the other key signs of ethnicity was personal names: whether Germanic or Latin/Christian. We can see naming fashions change across the period. We know of people with two names – one Roman and one Germanic – and people who took new names when they went into secular service. Gregory of Tours had a great-uncle called Gundulf (a Germanic name), almost certainly not his given name; he presumably adopted it when he went to serve in the bodyguard of the Frankish king. Just as Gregory took the name Gregory when he entered the church. Romans had several names anyway, so the addition of another name – again – didn’t represent a straight swap.

Another sign might be language. The laws often used Germanic terms for legal concepts. It is possible that declaring, for example, Frankish ethnicity (and the legal privilege it brought) at a law court required some knowledge of these, even if one couldn’t actually speak Frankish. It might also be that certain features deployed in the burial of the dead were also markers that proclaimed ethnic identity and status, but this is very difficult to identify for sure. Most of the old linkages between burial styles and ethnicity are extremely crude and unconvincing.

In an important article from the late 1990s Walter Pohl discussed these ‘signs of distinction’ as he called them and argued that the fact that these never related in a 1:1 fashion to ethnic groups in practice is significant. I don’t think this was the right conclusion. In the 1960s an anthropologist called Moerman studied the Lue in South-East Asia. He interviewed lots of the Lue and asked them what made them Lue; what were the signs of membership. He was able to compile a long list. In reality, though, these were either hardly ever worn or used in practice or, alternatively, they weren’t exclusive to the Lue. We can make the same point. The fact that Lombards weren’t the only people to sport long beards, or that Franks didn’t always go around armed with a throwing axe really says nothing about whether or not people at the time believed that these signs had particular meanings that related to particular ethnic groups.

Non-Germanic Ethnicity

Although historiographical debate about late antique ethnicity has focused on ‘Germanic’ barbarian groups, we need to remember that these weren’t the only ethnic identities that existed or which came to the fore in this period. We can list the ethnicities of the Britons in Gaul – later called Bretons – for example, or the way a British identity became important in Britain. In Gaul, identities based upon the city-district (civitas) came to the fore and became politically very important.

In the east, too, regional identities – bound up with adherence to particular forms of Christianity – may also have become increasingly politically significant.

Ethnicity in Late Antiquity (3): Critique of the Ethnogenesis/Traditionskern model

The advantage of Wolfram’s ethnogenesis model was that again, it allowed people to see peoples as ad hoc accumulations of individuals. Membership was about signing up to a set of ideas, rather than biological descent or anything like that.

This is a model that has been very seriously criticised and indeed has been one of the most bitter debates in late antique history, fought out essentially between what has been called the Vienna school, of Wolfram, his principal student and successor, Walter Pohl, and Pohl’s own students, and on the other hand a ‘Toronto school’ led by another Walter, Walter Goffart and Goffart’s students, chiefly Alexander Callander Murray. Much of the heat was produced by the Toronto school’s careless - and in my view entirely needless - accusation that Wolfram and Pohl were continuing in their works a line of Nazi ideology but had simply wrapped it up in a new set of vocabulary. Goffart is a Jew who was forced to flee Europe as a very small child because of the Nazis; Wolfram, Pohl and the others, as Austrians, obviously feel the weight of their country’s Nazi past (and neo-Nazi present) very keenly so this was hardly going to end well. On the face of it, Callander Murray had a point when he showed that Wenskus’ model of links of people to a leader had links to the ways some Nazi historians thought about the links between a Gefolgschaft (a following or retinue) and a leader or Führer. He wasn’t wrong. The problem is that the Nazis had absolutely no cogent or coherent body of historical ideas; different Nazi thinkers thought all sorts of things – many entirely incompatible with each other – about the ‘Germanic’ past. So turning a kind of intellectual line of descent for a strand of thought into a surreptitious continuation of an ideology was essentially defamatory. It’s possible that it would be difficult to find any post-war German thinking about the Germanic past that didn’t have some link to some idea accepted by some particular Nazi or other. This was unnecessarily, offensively polemical.

The biggest problem was the evidence. The ‘Germanic’ ethnogenesis model relied for its evidence upon stories written down in the early middle ages, after the ends of the migrations, usually in the post-imperial western kingdoms, in Latin, often by people who weren’t members of the ‘Germanic’ people in question but people of provincial Roman descent. The Getica, the earliest of the Histories of the Goths, was actually written in Latin, in Constantinople, by someone called Jordanes who claims to have been a Goth, but the sources he mentions were a lost history of the Goths written by the Italian Cassiodorus, whom we’ve met before and another Gothic history by someone called Ablabius who was possibly a Gallo-Roman. The History of the Lombards was written by Paul the Deacon, who was a Lombard but who was writing in the late eighth century, in Latin. Another Gothic History was written by the Hispano-Roman Isidore of Seville; there are no extant histories of the Franks before the seventh century, by which time they have acquired all sorts of mythical additions; Bede’s Ecclesiastical History of the English People was written in the early eighth century, in Latin. None of this stuff, in other words, comes from early Germanic-speaking peoples in an unmediated form. Even sources like Jordanes – if he was using authentic Gothic tradition – admits that there were other Gothic traditions that told quite different stories. Indeed, the Gothic History told by Isidore in seventh-century Spain is nothing like the history told by Jordanes in Constantinople two generations previously

Now, we simply don’t know what the sources were that these writers used, or how historically reliable they were, or how genuine they were as bits of Gothic or Lombard or Saxon legend. We don’t know what the writers did to these sources when they incorporated them into their Histories. Indeed, we can’t even be sure, most of the time, that they hadn’t made the whole thing up out of their own heads. Goffart has effectively argued that. I don’t think we need to go that far all the time but the bottom line is that usually – overwhelmingly usually – there’s no way of proving that they didn’t.

Another problem is the interrelationship of the sources. There are numerous similarities between Paul the Deacon’s History of the Lombards and Jordanes’ Getica. Is that because the Lombards and the Goths shared a similar ‘Germanic’ origin legend’ or because Paul had copied Jordanes (or Cassiodorus)? It’s difficult, too, to avoid the idea that these writers were in a way competing to give their people a history that was just as good as – or indeed better than – the others, whether Romans, Goths, Lombards or whatever. The Gothic history started in Scandinavia and by the ninth century it seems to have become de rigueur to have your people’s history start there.

These stories, furthermore, don’t just include elements that look like possible elements of ‘Germanic’ origin legends; they are sometimes filled with other material. Jordanes’ Getica includes things from classical myth and legend. The Goths turn up at the siege of Troy for instance, and intermarry with the Amazons, and so on. He weaves the Goths into stories from earlier Roman history. Indeed, pretty much wherever we can compare Jordanes’ account with more contemporary evidence, he’s wrong. On this basis it becomes difficult to know on what basis one can sift out the other, uncorroborated stuff as fragments of authentic Gothic history. There are two points that also come out of that. One is that both Jordanes’ and Paul the Deacon’s histories of their people were not stand-alone works; both also wrote a Historia Romana, a history of the Romans against which it was supposed to be set. In that sense we might have something comparable to Eusebius of Caesarea’s interest in weaving Christian history into the history and mythology of Rome and Greece. Finally, these origin legends of histories of people are not something that only include the Germanic-speaking barbarians. There was a fad in late antiquity for what one might call ethnic histories. Works existed in which the different peoples of Italy were traced back to mythological ancestors often via tortuous migrations (often from Troy).

Indeed, the Romans themselves had an origin myth that conforms to many of Wolfram’s elements of a Germanic origin myth. There are alliterative twin brothers, Romulus and Remus; there’s a hero, Aeneas, who crosses a sea from his homeland, kills a mighty enemy and takes his lands and so on. It’s very easy to see the writers of these sources composing works that simply gave their people a pedigree, a place, and an acceptable history in the antique Mediterranean world.

Another, major problem is that no source contains all of the elements of Wolfram’s ideal ethnic history. Wolfram’s reconstruction is way too schematic. He takes some elements from one source, adds others from another and yet others from a third. Things are interpretated according to the place where they ought to be or the role they ought to be playing in his schematic Germanic history. And sometimes he even corrects the sources – like Bede – for allegedly getting their legends wrong! According to Wolfram, Hengest and Horsa should come at the start, as Vanir brothers, not the Aesir god Woden! Silly old Bede.

This probably sounds very silly, but the reason Wolfram did this was because of his philological approach to the early medieval source material. It was for a long time believed that you could talk of ‘Germanic’ peoples as a sort of unity, sharing a unified ethos. Thus what one had in one Germanic origin myth could be put into a melting pot with all the others. This would allow you to fill in the gaps in some sources from parts of others and so on, as mentioned above. The same approach was taken to the law-codes of the post-imperial West, which were also regarded as Germanic. So you stirred all the laws in together and on that basis reconstructed the pure, original proto-Germanic custom that they all descended from. Here you ended up with the original, pure proto-Germanic origin legend. Ultimately, this had all originated in the politics of nineteenth-century Germany. In the course of the Napoleonic Wars a belief in a shared German nation had become very popular. Historical projects set out to underpin this. One, founded in Hanover and still going, but now in Munich, was the Monumenta Germaniae Historica: the Historic Monuments of Germany. Because the Middle Ages had been created when the ‘Germanic’ peoples conquered the Roman Empire, the Middle Ages were a German creation and its entire literary output could be claimed as Germanic!

In spite of originating so clearly in a particular historical moment with a specific national political agenda, the influence of this way of thinking is still frequently to be found in writing about the barbarian migrations and the early middle ages. But if we can see past it, and people have been unpicking this for a generation now, we can get a much more interesting insight into the ways in which people wrote histories and origin legends to try to give their people a place in the world of Rome and its successors.

Ethnicity in Late Antiquity (2): Ethnogenesis

No discussion of ethnicity and ethnic identity in late antiquity can avoid discussion of the ethnogenesis debate. Ethnogenesis, though it might sound as though it’s prog rock played on world music instruments, is in fact about how peoples come into being. In particular it means the debate around a particular theory of ethnogenesis put forward by an Austrian historian called Herwig Wolfram in the 1970s.

Wolfram based his work in turn on that of a German historian called Reinhard Wenskus who published a hugely important, often cited but rarely actually read, book called Stammesbildung und Verfassung. One reason that it isn’t translated, apart from it being long and technical, is that many of the terms translate with difficulty into English, and that includes all three concepts in the title: Stamm, Bildung and Verfassung!. Stammesbildung means, roughly, the formation of tribes, though ‘Bildung’ means more than simply putting something together as, like ‘formation’ in French, it can also mean education, and Stamm has a different valence from ‘tribe’ in English – it doesn’t necessarily carry the same sense of something pre-modern, or a simpler social formation. Verfassung is trickier. It means something like constitution but like the word constitution in English, it can also mean the state of your health, and thence it can mean the state itself. Anyway, Wenskus’ book was in some ways a sort of typology of the ways in which Stamme, which I’ll translate as peoples, were formed. One of the important aspects of Wenskus was that he eschewed the notion that there was anything inherent or natural about peoples.

In earlier work in German in the 19th and 20th century there had been a great deal of work on the Volk, the people. The Volk had picked up unpleasant connotations because of its use by the Nazis. Ein Volk; Ein Reich; Ein Führer: One people; one empire; one leader. So there was a strand that saw peoples as very much biological, genetic entities, in a way that could lend itself to the sort of hierarchy of peoples of which the Nazis were fond. Not, it must be said, that the Nazis had a monopoly on this type of thinking. This was fundamental to the idea of the nation state – that they were coterminous with a people, with a history and racial unifying features, national characteristics and so on, defined – like Romanness – in opposition to set of idealised others: a constitutive outside, in technical language – meaning that a category is defined by all the things that are outside it as well as (and often more than) by all of the things it contains.

Wenskus instead took the line that peoples were formed by people crystallising in some way or other around a particular type of leader and/or a particular type of set of core beliefs about what they were as a people (here we come back to some of the things I said about identity in the last lecture). In German this could be referred to as the Traditionskern – the kernel of tradition (kernel as in core, rather than colonel). This was the body of stories, myths and so on, customs, laws perhaps, religious beliefs that gave a unity to a people. These things are very important even today. Consider the current furore about the National Trust presenting the history of Britain in ways that challenge the national historical myth; or the ways in which inhabitants of the southern USA object so violently to any questioning of the mythology of the Confederacy built up in the early 20th century; why people in loads of countries across the world really object to history that shakes up the national myth. People take it personally because they see it as an attack on their very identity. For me that is the big challenge of history: how to make people realise that history doesn’t and can’t work like that. Anyway, the key thing was the peoples were active, social and political creations, and not biological/genetic realities.

The popularisation of this idea more widely in the world of medieval studies, however, came with the Austrian historian Herwig Wolfram who worked mainly on the Goths and whose works were translated, into English and French and other languages. Wolfram employed the term ethnogenesis, which hadn’t appeared in Wenskus’ work. Wolfram took it from the work of a Soviet ethnologist called Julian Bromlej. Wolfram developed the idea of ethnogenesis in a particular way, based around Germanic philology (that is the study of the development of languages, words and their meanings, the original forms and meanings of words and texts, and so on). Now, we’ll come back to the problems inherent in this but for now it’s important to set out what what we might call the ‘strong thesis’ of ethnogenesis was.

It was an almost rule-like theory about 'Germanic' legends about the formation of a people, which were then 'borne' as the 'kernel of tradition' by the aristocratic core of a new people, which would then attract new recruits along the way. This, was where he developed Wenskus’ work

In this reconstruction, the Germanic origin legend went like this. Once upon a time, the people in question left their ancient homeland. To do this they had to cross some sort of physical frontier, usually a river or a sea. They did this under the leadership of brothers, often twins, referred to by Wolfram using (ironically) the term from classical Greek legend dioscuri. Once they arrived on the other bank or shore, the people fought what one might think of as their defining enemy, from whom they took the land. Various other changes were held to take place at this point in the story. For one thing, the people underwent a religious change. Wolfram said that this stage marked the shift from worshipping the Vanir gods, with their twin founders related to a goddess, to worshipping the newer Aesir gods, with Woden as their chief deity. Wolfram also thought there were changes in the political constitution of Germanic peoples. The old Germanic languages had several words for king. In old English the word is cyning, which is related to kin, and suggests a role within a greater kindred, perhaps. The words Wolfram was more interested in, however, were the Gothic Thiudans and Reiks. Thiudans, a bit like cyning, is a term that links the ruler to the people; it is cognate with modern Deutsch, which derives from a word meaning ‘the people’. Reiks on the other hand is a word with a wider Indo-European meaning – cognate with Latin rex, Irish , Sanskrit raja and so on – and has a clearer monarchical sense. Wolfram wanted to link these terms to a passage in Tacitus, which I have mentioned before. In the Germania Tacitus says that the Germani reges ex nobilitate, duces ex virtute sumunt (‘the Germani choose their kings according to their nobility and their war-leaders according to their prowess’) to argue that there were two types of king: a ‘sacral’ king who belonged to a royal family and who had long-lasting powers but largely religious functions, and a war-leader-king chosen only at times of crisis but who had more wide-ranging but short-lived authority.

I am going to criticise this pretty seriously, but it’s important for me to say first that Wolfram is a tremendous scholar and actually a very flexible thinker. He has changed his mind about various aspects of his theory and responds generously to other people’s work. But I will leave that critique for the next [blog-post].

Ethnicity in Late Antiquity (1): Rethinking Ethnic Identity

In this [blog-post] I want to revisit the chapter about ethnicity from my 2007 book Barbarian Migrations and the Roman West to set out a number of areas in which I think my views have changed since that publication.

The first issue to raise is quite a big contradiction in that chapter, which you might have noticed. Much of the theory that I discussed in the discussion of how ideas of ethnic identity have changed is concerned with groups of people – and generally when we’re looking at historical sources what we tend to see is discussion of groups. But then the model that I propose is really not about groups at all but about specific people interacting with other specific people. This isn’t entirely surprising given the theory that influenced me most at the time, as I’ll come back to in a moment, but it does make for a bit of a fault line in the discussion. 

That’s quite a serious flaw but probably the important issue, and it’s hilariously ironic (hilarious to me anyway) given the title of the chapter, is that nowhere in that chapter do I actually define identity. Obviously there’s a lot about what ethnicity is and how it might work, but it never really gets down to discuss what an identity is, and how that might work. To be fairer to myself, this is actually pretty general in history. Here’s a little challenge. Go away and find one of the million and one books or articles with the word identity in the title and see if they ever give any sort of idea of how they understand the concept of identity. I pretty much guarantee it won’t. Identity has just become a sort of trendy place-holder for more or less coherent thoughts on what might be called identification, labelling or categorisation; at best on how people thought they fitted into history or the world.

But we need a rethink of all of this, and I have been trying to put one forward for the past few years. In my very earliest work I was very influenced by Pierre Bourdieu and Anthony Giddens, whom we’ve encountered already in this course. From them I took the idea that social change and social structure came about from the cumulative interaction of people of different categories, constantly building up a memory bank of correct and inappropriate social interactions, which could never entirely be maintained in the same state; the structure, that was made up of people’s knowledge of that perpetually growing memory bank, could never entirely reproduce itself. So social change was always on-going. Stasis and absolute continuity would be far more interesting – and unusual – than change. 

My thinking is still ultimately based on that but what I have added to it since then are insights from some of the philosophy that I have already alluded to in this course. The most important change – I think it’s an advance; you might not – concerns exploding – opening out – what those social categories are. When I have talked about interactions between old and young, man and woman, aristocrat and commoner, Frank and Roman, what do those groups mean? I think it’s too common just to think of these things as self-evident entities. Here we ought to go back a couple of weeks to [an earlier lecture which I hope to post on the blog at some point] and the very brief reference made there to interpellation and subject-position: the way that a social interaction calls you into a particular persona, and gives what you say a specific kind of weight, authority and so on. We can extremely briefly also refer back to Jacques Lacan in that context. One of my favourite quotations from Lacan is one that I allude to, I just noticed, in [an earlier lecture which I hope to post on the blog at some point] as ‘the quote at the start’ but which in my reorganisation of the materials for this course, somehow got edited out from where it had been. The quotation is ‘A fool who thinks he is a king is no crazier than a king who thinks he is a king’. What Lacan meant is that there’s nothing intrinsically king-like about someone who is a king; what makes them a king is how people see them, how they see themselves, and how they behave; and especially the idea that someone with a particular persona, whether king, young man, Roman, village elder, professor, student, whatever, conducts themselves not simply according to the way they see other people and their relationships of power with them but also a projection of what they think those other people expect or want from them. People are always constrained by what they think the other expects of them. 

So, what that in turn means, returning to the specific issue of ethnic identity, is that when someone takes part in a social interaction from the subject position of, say, a Roman or a Frank, or a Gaul, or a Briton, that ethnicity is not some simple or straightforward thing that can be referred to. It is composed – firstly – of the person’s idea of what that category means, as an ideal. Here the most obvious illustration would be the Roman because we’ve discussed it often already and because we have the best evidence for it. Someone participating in a social interaction from the position of a Roman, judges their behaviour against the ideal behaviour of a Roman: moderate, reasoned, self-controlled, and so on. Internally, he (and we’ll assume it is a he, because of the unrelenting masculinocentric focus of Roman thought about good Roman-ness) he will consider himself acting in the view of other Romans, judged by them as Roman but also as against the things that were outside that category: woman, barbarian, and so on. And then that person will have a set of ideas about what they expect from the person they’re interacting with, based around ideals, presuppositions, prejudices perhaps, as well as societal norms. Now, the thing is, you can never be that ideal; it’s always something you’re moving towards; you can never be a category in yourself because a category is a sign, with signifiers and signifieds which, as we’ve seen, is constituted as much by what it isn’t as what it is, and you can never get back to a point where a sign and a thing are completely self-identical. That’s Derrida, if you remember, from [an earlier lecture which I hope to post on the blog at some point]. Also, though, because the ideals shift all the time, as I said earlier, as a result of the cumulation of interactions, that ideal itself is always on the move. And obviously the other person is also acting according to the same ideals and constraints and reflections. So, what you have is not, as you might see in my work before about ten years ago, an interaction between two coherent things, described as identities, two coherent categories, but rather between two shifting, fluid, incoherent bodies of ideas, signs and so on. 

And this leads me to my second major rethink. Everything I have just said, obviously points up the idea that one social actor’s view of their social persona, the ideal of their identity, what it represents, how it ought to behave, and how other people should conduct themselves in regard to them, need not coincide very closely with the ideas of the other party to the interaction. What I have argued in recent work is that every social interaction between people whose subject position is based upon particular identities is a wager: a bet if you prefer. A bet on the fact that the other person will accept your subject position, first of all (will, e.g., they even accept that you are a Roman), but also that they will share – more or less – your ideas of what that means for your behaviour and their behaviour towards you. 

What do those developments mean overall? The most important thing is that they make ethnicity and identity even more fluid than I thought they were, but they also make social structure – the body of correct and unacceptable ways of behaving – even more difficult to pin down and reproduce. That in turn makes it even more important to think about how identities – as those bodies of ideas and ideals – are marked out and how space and other cues are used to prompt the ‘correct’ behaviour, or flag up whether interactions are formal or informal. That brings us back to the uses of material culture discussed in [an earlier lecture - I hope to post this at some point]. Notice, too, though, that the material culture used to signify identities itself changes across our period. Our historical categories are always moving categories even if, at any one time, people might think they have a good idea of what they mean and – indeed – that that is how they’ve always been. 

The final major area in which I need to rethink what I published in 2007 concerns the relationship between race and ethnicity, but I will return to that in more detail in [a future blog-post].

Ethnicity in Late Antiquity: Introduction to a series of short blog-posts

[This term I had to reorganise my second-year 'Histories and Contexts' course 'The World of Late Antiquity, c.300-c.650' to allow it to be taught on-line. For me, this meant reorganising what had been my lectures into 'lecture packages' for students to work through. Each lecture I divided into up to 6 'elements', either a 10-minute video lecture or a background essay for the students to read. I soon discovered that I found it really difficult to talk to a 10-minute time limit coherently, without a script. As a result, I ended up with probably a book's worth of lecture scripts and background essays. Perhaps I will write these up into somesort of introductory overview volume (?Introducing Late antiquity: A Western Perspective?). In the meantime I thought it might be of use or interest to upload these to the blog as 'short reads'. I have replaced some of the terminology - so, for example, 'in the next video-lecture' has become 'in the next [blog-post]' with the square brackets denoting the replacement. 

This first cluster of such 'short reads' is about ethnicity. Remember they originate as second-year undergraduate lecture elements and so lack any kind of aparatus. I may come back and add further notes and comments in the future. Most of these were written under pressure of time, too, so there might well be silly errors in amongst. Nonetheless I hope you might find something of interest or use in them. 

Below is the background introduction to the original lecture package which was posted on our VLE (Virtual Learning Environment).]


One of the most studied and frequently most hotly-debated areas of late antique history is ethnicity. The crisis of the 5th century (the so-called fall of the western Roman Empire) involved regional factions focused on groups with non-Roman identities and eventually settled down into 'barbarian' kingdoms. Traditionally, this was thought to represent the conquest of the Roman Empire by the barbarians, but what had become of all the Romans? Were they all killed, enslaved or driven out? Were the Franks, Goths, Lombards and so on all descendants of barbarians who had migrated into the Empire? It became clear to 20th-century historians that these 'peoples' must have been a more complex social phenomenon. So what did it mean to invoke an ethnic identity?

[The first element of this package was to read the discussion of ethnicity in my 2007 book Barbarian Migrations and the Roman West]

[Part 1]: Re-thinking ethnic Identity. In this [post] I want to revisit the issues from that chapter and set out the ways in which I think I would revise my views now, 13 years on.

[Part 2]: Ethnogenesis. One of the key areas of debate has been around the idea of ethnogenesis: the creation of peoples. This [post] talks about the work of two historians - Reinhard Wenskus and Herwig Wolfram - on how peoples formed around an elite group that bore the traditions (customs, beliefs, myths) that unified the people. Herwig Wolfram developed an idea of the origin myth that 'Germanic' peoples used to structure their past and give them a historical identity in the present.

[Part 3]: Critique of the Traditionskern/Ethnogenesis model. This ethnogenesis model has been heatedly - even viciously - debated between one group of scholars known as the Vienna School clustered around Wolfram, his students, most notably Walter Pohl, and Pohl's students) and the Toronto school, centred on another Walter, Walter Goffart, and his students. In this [post] I will set out some of my own criticisms of what I call the 'strong' ethnogenesis' thesis before suggesting that we might well accept a 'weak' thesis.

[Part 4]: Ethnicity in the post-imperial kingdoms. In the fifth century, probably developing from the state of affairs in the 4th-century Roman Empire (as discussed last week), in the developing western kingdoms, ethnicity was largely fucntional.  Those claiming a 'barbarian' identity (Gothic or Frankish, etc) were the army; those with a Roman or provincial Roman identity were tax-payers and staffed the civil bureaucracy, the Church and so on. Another heated area of debate has been how easily (if at all) people could change their ethnicity in the 5th and 6th centuries. I will summarise this and set out my own view.

There were numerous ways in which ethnic identity could be proclaimed in the 5th and 6th centuries.  How did these actually relate to ethnic groups.  Walter Pohl has suggested that the fact that these never related in a 1:1 fashion to ethnic groups in practice is significant.  I will suggest that he didn't get this right, using a discussion by Moerman of the South-East Asian Lue people as a focus.

[Part 5]: Ethnic transformations in the West, c.600. As ever, the post-Justinianic world saw radical changes.  The most important was a sharp decline in the value of 'Roman' identity.  With the unpicking of the binary between 'Roman' and 'non-Roman' within the western kingdoms, we see a different way of seeing ethnicity in the West.  Much of this is focused upon the law.  I argue that the concept of the 'personality of the law' emerges at this time - that is that some (élite - in my view at least) groups could claim the right to be tried according to the law of their people.  Much of this might be associated with the reign of the Frankish king Chlothar II (584-629).

At this point we can end by considering whether there was any link at all between the processes of socialisation and membership of a political grouping.

A Note on the Germanic

People used to talk about the Germanic-speaking barbarians as though they formed a unified group. Other than the fact that they spoke languages that were related to each other, there is no evidence of a pan-Germanic identity or ethos in antiquity. The idea that there was is is something that was essentially invented at the Renaissance and especially in the 19th century. To talk about Goths, Vandals, Saxons or Franks interchangeably as 'Germanics' is grossly misleading - it is worse still to refer to them as 'Germans', which is still sometimes done. Part of the problem is that English no longer has a word for ancient Germani, different from its word for modern Germans, unlike modern German (in which  ancient Germani are die Germanen as opposed to die Deutsche) or French (which has les Germains - ancient Germani - and les Allemands - modern Germans). So when I use the word 'Germanic' in these lectures, unless I am talking about the languages they spoke, you need to hear those inverted commas!