For reasons unconnected with the project I came back to this recently. As I have seen it alluded to a couple of times on other medieval blogs and as I think it is a clear statement of my current views of the subject, and of methodology, I thought it might be worthwhile to post it here. It was orginally given at the International Medieval Congress in Leeds in July 2009.
At the time what struck me was how many people came up to me afterwards and said they agreed with me, before revealing that they hadn't really understood what I was saying at all... Ho hum. Netwise, Jon Jarrett has a nice summary on his Blog, but he seems (to me) to see me as taking sides with those who think that historicist dominance is either a "good thing" or the only way of doing things, in that he appears to perceive an opposition between me and Eileen Joy. Insofar as I am aware of Eileen's stance (critical theory over historicism), I think I'd be on her side - politically at least.
Interdisciplinarity is one of the most used words in our professional practice, sprinkled generously to season MA course and programme descriptions. At York some moderately serious thought has been devoted to what interdisciplinarity might mean, but, if one probes what it means in practice, it is not interdisciplinarity at all, but interdepartmentality, something rather different. In funding applications it is now more or less the law of the land to use the word interdisciplinary somewhere. No matter that one needn’t actually explain how this will be achieved in practice. I know; I have traded in this bankrupt stock for many a long year! And finally, as we all know, in modern British universities (especially my own, apparently), whenever two or more shall be gathered together, they shall form an interdisciplinary centre.
Interdisciplinarity is a cherished myth of medieval studies. In John Arnold’s handy primer, A Short Introduction to Medieval History the ‘interdisciplinarity’ of the topic is cited as a reason one should do medieval history, although ‘its practitioners fret periodically about what interdisciplinary might mean’.
I am going to do more than fret. I am going to argue that genuinely interdisciplinary exchange has had important results but has had its day – perhaps killed by its own hand; at least as far as it relates to relationships within medieval studies, the word interdisciplinary has now become meaningless and should be rigorously challenged at its every usage. Such challenges are, further, absolutely vital for the intellectual health of our studies; so-called ‘interdisciplinarity’, rather than being an enabling ‘jewel in the crown’, has in fact become a ‘Bad Thing’, which threatens to stifle approaches to the written and material remains of the so-called Middle Ages.
To be interdisciplinary one needs to have some idea of what constitutes a discipline. What, one might ask, differentiates a discipline from, say, a technique or a skill? There is I suppose a standard ‘obvious’ answer at a basic level, which would equate a discipline to the sort of thing that a modern university department is named for: English, French, History, Art History, Music, Philosophy, Archaeology and so on. But this brings a lot of problems.
Such is the demand nowadays to be considered interdisciplinary that one hears some very strange (if – in this context – nevertheless understandable) claims. Here are three, which I’ve encountered. If you study medieval literature in English and French are you being interdisciplinary? Can you claim to be interdisciplinary in History and Latin? If that is possible, then any literate historian is interdisciplinary, in history and whatever language or languages she reads, including that which she speaks! Can you be interdisciplinary in palaeography and English or history? Is palaeography a discipline? If so, you could be interdisciplinary in archaeology and pottery typology. When does a skill become a discipline? Archaeology has fought a long battle – whose outcome is still far from universally conceded – to be considered a discipline rather than an ancillary or auxiliary skill or science.
To return to the disciplines that make up Medieval Studies and thus the thrust of my argument, if disciplines do not, on close inspection, seem to have much internal cohesion, one might still maintain that there are ‘disciplinary’ lines drawn round groups of types of evidence that band them together into more meaningful, larger disciplines. One might call that an external definition, based on what is excluded and included.
This notion can be attacked from several directions. The bounds of disciplines are difficult enough to define when considering subjects as different (superficially at least) as history and archaeology, as I have written myself. When thinking about what one might call the textual disciplines – literature and history – the problems are multiplied endlessly, and it’s on these that I will concentrate. I’m also approaching the problem from the perspective of someone who knows immeasurably more about the development of the discipline of history than he does about the growth of the study of literature.
The canon of sources incorporated within a discipline is historically contingent. What constituted a historical source initially depended upon historians, for good or ill (for not everyone was agreed on this), differentiating themselves from the more literary/philosophical modes of Enlightenment historical production. Writing history wie es eigentlich gewesen ist meant only using the sources that told you wie es eigentlich gewesen ist; those serious documents that told you about high politics and religion. Many – legion – were those excluded on these grounds. At least, however, the canons of historical and (as far as I am aware) a literary ones were clear (leaving aside the large number of written sources considered insufficiently literary or historical!). Clearly, though, once one got into the twentieth century and reached the Annales School and Total History, the historical canon began to include anything and everything that was written down and many things that weren’t, as well. Even Trevelyan’s very un-annaliste Social History of England entitled its chapters ‘Chaucer’s England’, ‘Shakespeare’s England’ and so on. Similarly, movements in literary studies soon moved away from the nineteenth-century canon of ‘high literature’. I read on page 1 of Derek Attridge’s The Singularity of Literature that ‘all attempts since the Renaissance to determine the difference between “literary” and “non-literary” language have failed.’ Thus defined, the boundary between ‘history’ and ‘literature’ has long been permeable, something only underlined by recent theoretical trends.
One could, alternatively, take a Foucaldian approach and look at how the disciplines emerged. Foucault thought that one of the defining things about the episteme of modernity, ‘the age of history’ as he called it. By ‘history’ he didn’t just mean the discipline, but ‘historicity’. This, in his reading was part and parcel with the creation of distinct histories of distinct things, with the rise of scientific method, of empiricism, with the creation of disciplines. But, in Foucault’s view (as I understand it) it also produced a concern with the inter-relationship between things, not just their classification. In those terms one could contextualise interdisciplinarity itself within the episteme of modernity, just as, as I would argue, the collapse of interdisciplinarity might be located within the episteme of post-modernity. This would underline the artificiality and contingency of these divisions and point the way to how and why they are no longer valid.
Alternatively again, you could take a Derridian philosophical approach and question those boundaries. It seems to me that the bounds between disciplines are subject to the same critique, on the same grounds, as genres were in his essay on ‘The Law of Genre’. The bounds of disciplines, like the bounds of genres, are only acknowledged as existing because they can’t actually exist. They cannot be policed or patrolled. Their very existence implies their own transgression.
To make the problem more acute, the developments of theory over the past twenty years have in any case removed whatever might have been left of any boundary between literature and history in terms of ‘focus’. Since the so-called linguistic turn, older ‘social science’ approaches to history have been on the back foot and even many of those who claim an opposition to so-called post-modernism have actually incorporated – or at least have had to pay attention to – ‘textual’ or ‘literary’ approaches to their written data. At the same time, especially in medieval studies – and here I must confess to not being as sanguine as the session’s organisers about the acceptance of ‘theory’ – pretty much all of the dominant – and even less dominant – approaches to literature are in one sense of another, historical: so-called ‘New Historicism’, Marxism, Post-colonialism, gender studies – even in some ways and to some extent Queer Theory. In my understanding at least, it looks to me as though the hard-core ahistorical, literary or philosophical approaches to medieval literature are as much on the defensive as old-style historical empiricism or cliometrics. In theoretical terms, the two disciplines often seem hardly different at all. This was where real literary/historical interdisciplinarity had its moment, its importance and – I concede – its valuable results. The upshot, though, is the eradication of the possibility of further such interdisciplinarity. One must ask what exactly a New Historicist from a literature department and a ‘post-linguistic-turn’ historian bring to the putatively interdisciplinary table, other than technique – charter diplomatic and poetics, say? Since data-specific technique – skill-sets – alone can’t, as we’ve seen, divide one discipline from another, then one really must question whether this is interdisciplinary at all. Is it different from the collaboration of two people from history departments with different specialist skills? I have nothing against collaboration, which is fruitful and important, but we should be honest about how we describe it.
The bounds of the disciplines collapse from the inside and the outside. There can thus be no such thing as interdisciplinarity.
Sharp-eared listeners might have concluded that, with the genuinely interdisciplinary scholar defined as someone who has published work within two disciplines, accepted as such by practitioners of both, and with real interdisciplinarity denied in the interplay between disciplines based on the primary written text, what is going on in this paper is just me staking a claim to be the rare real deal. I suppose, to some extent, that is right! But only in a very limited sense; it might be the case in a narrowly methodological way. I will now, however, completely undermine my own claims to be interdisciplinary at all.
We must consider the outcome of interdisciplinarity. If I take my own work, which adopted what I have called – no less problematically than any other term I’ve derided today – ‘multi-disciplinarity’, the analytical separation of different data-sets and the merging of conclusions at a higher level than is usual, then I still have to ask myself what I produced at the end of the day – if it wasn’t just good old-fashioned social history? What emerges from the literary approach to history or the historical approach to literature if not straight cultural history? Once ‘context’ is assigned importance, history, as I see it, always wins. Perhaps this is because in spite of historians’ claims, the Rankean empiricist ideal still underlies pretty much all historical work. There have – perhaps understandably – been no real moves to produce history in line with the pre-Rankean ideal of history as ‘philosophy teaching by example’. Furthermore, the kind of explanation or causation that contextualism implies is (in my view) unsatisfying even from a historian’s point of view.
So I wonder if, whatever one might say about method, whether it is even possible for the outcome of so-called interdisciplinary dialogue, to be interdisciplinary. Is it ever anything other than cultural history of one form or another? I’d even suggest that centres of Medieval Studies don’t teach interdisciplinarity at all, but a single discipline, of medieval studies, which means medieval cultural history.
Obviously, I have nothing at all against cultural history – it’s what I do – and at the level of method and approach, what is currently considered interdisciplinarity has produced enormous quantities of valuable and groundbreaking work; I’m not denying that. I don’t doubt that in some, perhaps many, areas of the study of the past, work using the ‘inter-disciplinary’ approach is far better than it would have been had it not used that method. But when it becomes the only approach, I start to worry. The dominance of interdisciplinarity has I think become pernicious. It has produced the requirement to assert interdisciplinary status that leads to the sort of vapid claims that I mentioned earlier. Even ignoring the problems with justifying the term, the ascendancy of historicist approaches and the ensuing de facto dominance of cultural history, jeopardises any approach which isn’t interested in this outcome. It is already the case that most funding bodies stress so-called interdisciplinarity and I suspect that successful bids are overwhelmingly those using this buzz-word. This endangers the funding of ahistorical approaches to literature or of historical projects that do not require the method, impoverishing both. The ubiquity of so-called ‘interdisciplinarity’ threatens to make the academy less, not more, diverse.
What is at stake here is how one draws and defines the dividing lines on the map of the academy, and what is involved in intellectual exchange. I suggest, might be a more meaningful way of thinking about the fruitful confrontation of difference – a more interesting way of drawing the lines that sub-divide academia – is the confrontation of theoretical approaches. This is why I offered this paper for these sessions. I can’t think of a single word for the confrontation of theoretical approaches unfortunately! But whatever it is, it renders disciplines/interdisciplinarity irrelevant. Discussion between a die-hard Lacanian and a historicist might involve people from different departments but if it did it would really, intellectually, be no different from debate between an ‘old-style’ social science social historian and a post-linguistic turn cultural historian within a department.
I do think that this would be a more fruitful way of dividing the field of medieval studies and giving credit to really intellectually difficult – challenging – discussions. There are two advantages to this, as I would see it. First, the difficulty – the challenge – of such confrontation opens far more the mind to what Derek Attridge calls the ‘alterity’ involved in the creation of work. He is talking about literature, but I think the point might stand more generally. Second, while so-called interdisciplinarity tends, as I have argued, to lead one way, however fruitfully in its own terms, to cultural history, the confrontation of theory might be conducive to more fruitful dialogue – two¬-way, as the word implies. If theory is – as it is – a way of seeing, rather than something to be applied, mechanically, it is something to be engaged with and refined, and this can be done from two points of view with an outcome, I suggest, that, as well as producing better theory, also reflects back, in different ways, on the areas whence those points of view started, rather than producing a single, merged, synthetic, effectively uni-disciplinary outcome.
So, my conclusion is not to outlaw or denigrate interdisciplinarity, as such, but to be much clearer about the what it is, its limits, and what it does – and the limits to that – and thus to challenge this ubiquitous demand for all scholars and all work to describe themselves/itself as ‘interdisciplinary’. On the one hand, I suppose, I do want to see value returned to work that isn’t interdisciplinary, but mostly I want to suggest a way forward which valorises engagement at the level of the development of theory, not under the empty rubric of interdisciplinarity.
Because of serial spam attacks which the Blogger platform seems unable to deal with (yes - people warned me about Blogger), I have moved the...
Sunday, 17 October 2010
Wednesday, 6 October 2010
The Woad Less Travelled: the archaeology of earlier Pictish society and politics viewed from Francia
A Frankish model, or two
I will consider Pictish cemeteries in the light of my earlier studies of Merovingian Gaul’s rich funerary archaeology. These have often been concerned with the difference between sites or periods wherein the investment in funerary commemoration was focused upon the transient – grave-goods – and those in which it was concentrated upon more permanent above ground markers. I have also examined the interrelation of the different variables involved: grave-size and shape, or construction; skeletal position; the placement of the grave-goods; the organisation of the cemetery, and so on.
As I have discussed this on countless other occasions, this description can be brief. In the sixth century, large cemeteries are known, perhaps shared by several communities, organised quite neatly in rows (hence their frequent name: Reihengräberfelder; cimetières par rangées). The families using these grave-yards all seem to participate in the grave-goods custom to some degree. There is little sign of above-ground markers though some surely existed given the regularity of the lay-out. Grave-goods are distributed according to what appear to have been communal norms, structured around age and gender. Investment in funerary display was skewed towards adult males overall and towards adults over children. As this expenditure focused on those individuals, within earlier Merovingian society, whose deaths would have created the greatest tension in social relations, I have interpreted this as indicative of a society wherein local authority and power were maintained with difficulty from one generation to the next and open to serious local competition. This interpretation accords well with the picture drawn from other sources.
In the seventh century, by contrast, the number of grave-goods decreases dramatically. Gender-specific artefacts (especially feminine objects) are greatly reduced in number; there are far less clear relationships between artefact assemblages and the deceased’s age and sex. Instead there is a greater investment in above-ground monuments: sarcophagi with lids visible at surface level, gravestones, walls around burials, reused Roman masonry and sculpture used as markers, and so on. In addition, cemeteries are smaller but more numerous than before and neat organisation into rows often breaks down in favour of smaller, probably familial plots. The investment in funerary commemoration is, if anything, more weighted towards men than women than before, in terms of grave-goods, but the other elements of the tomb, such as the sarcophagi and monuments, show a much more even distribution between different age-groups and genders. The first part of this seventh-century phase is often a phase of ‘separierung’, when locally high-status families distinguish themselves from the remainder of the burial community by less subtle means: above-ground monuments, distributing lavish grave-goods with individuals of ages that would not generally merit such investment in the earlier phase, differential orientation of burials and the physical removal of the graves to a separate part of the cemetery. In other, peripheral regions of the Frankish world this is a period when large ‘aristocratic’ barrows are built; in the region of Metz, churches start being built in the town and other administratively nodal settlements at this time. Shortly afterwards, these families seem not to have buried their dead at the communal cemetery at all, by the mid-seventh century constructing churches on their rural estates. In short this phase is characterised, with reference to the preceding period, by elite ‘rule-breaking’.
The second model is furnished by the study of the Merovingian inscriptions of Trier. After exhaustive study, Mark Handley argued that the investment in epigraphic commemoration in Trier matched that in grave-goods, and there are indeed some important similarities. It seems to me, however, that the investment in the commemoration of men and women is much more even than with the grave-goods custom, and relative expenditure of resources on the marking of children’s graves is certainly greater. Overall, the Trier inscriptions resemble the seventh-century phase of the rural cemeteries in many ways.
The transition from the sixth-century situation to that of the seventh century, and the phase of ‘separierung’ in particular, can be explained with reference to the emergence of a more established aristocracy, which accords with changes in other classes of data (written sources, urban and rural settlement archaeology and so on) at the same time, around 600. This period saw the end of anything that might be termed a state in Francia, though the kingdom remained fairly coherent. The patterns visible in the Trier inscriptions, which span the whole period, can be viewed as the manifestation of a particularly self-conscious (and highly unusual) local aristocracy in Trier itself, the existence of which is visible in other evidence too. Thus, there are important correlations between particular types of funerary display and certain forms of socio-political formation, which match changes in other forms of data which can be understood in the same way. Finally, the transition from one form of commemoration to another is therefore a good index of social change.
This very briefly sets out some background to my consideration of Pictish society. It sets a context for the questions that, from a Merovingian perspective, I would ask of the surviving data. It is most definitely not the aim to use the Merovingian pattern to calibrate the Pictish, to ‘read off’ the Pictish evidence against a set of Frankish correlates. By contrast, I will use the Pictish case to question my reading of the Frankish evidence. I want to see if employing insights from a better-documented area might lead us in some slightly different directions to provide a model that might be refined, corrected or rejected by future analyses.
Long cist inhumations: How Christian? How Roman?
I begin with two interpretations of the ‘classic’ Pictish long cist cemetery. The first is that these cemeteries and the rite of inhumation relate somehow to Christianisation. More subtle studies distinguish the appearance of the inhumation rite from the appearance of long cist cemeteries and, analytically, this seems to me to be quite right. Nevertheless long cist inhumations have generally, as far as I have been able to detect, been interpreted in terms of conversion or Christianisation, in spite of doubts cast upon this as long ago as 1980. The religious explanation is wanting in many ways, however, being based upon several fundamentally mistaken premises. The first is that there is something essentially Christian about the absence of grave-goods. Given the general absence of grave-goods throughout the prehistoric, Roman and early medieval eras in Scotland, excepting those furnished inhumations associated with ‘Anglian’ and later (more plausibly) Scandinavian settlement, their absence in the post-imperial era cannot be deemed especially significant. The general assumption of a Christian equivalence finds no prima facie written or archaeological support. The analogous inhumation rite generally replaced cremation within the western Roman Empire in the third century, before the significant adoption of Christianity. The late antique Church took no interest in grave-goods, except where the objects concerned were liturgical or church property, and there the issue was the alienation or diminution of ecclesiastical wealth, not involvement in un-Christian ritual. Numerous archaeological and documentary examples of Merovingian furnished inhumation under churches are well known. Thirty-three to thirty-five years after Bailey Young’s classic study of the issue, this point ought not to need repeating. Less well known, perhaps, is the fact that the Church expressed remarkably little opposition to cremation before Charlemagne’s Saxon campaigns at the end of the eighth century. The Christian associations of the rite of inhumation in a stone cist are weak – after all, such inhumations are known in scattered examples from earlier phases of Scottish archaeology. The idea that long cist cemeteries represent a phase of Christianisation is hardly more solidly based. The better assumption (albeit lacking clear empirical foundation, at this date) seems to be that Christians would wish to demonstrate their community in burial, although this is hardly the only possible reason for the appearance of communal cemeteries. The other assumption, not uncommon in British archaeology, is that the Church espoused an ‘obfuscating’ ideology of equality in death; this is simply untrue. Nevertheless, the appearance of the communal site seems to be the most important element of novelty involved in this development.
What alternatives to the questionable Christian interpretation might there be? One, more recently proposed, would see the appearance of this rite as the product of Roman influence – after the early fourth century this would not be incompatible with Christianisation. In favour of the interpretation one could point to the appearance of unfurnished inhumation in the far north of Germania Magna at the end of the fourth century, almost certainly resulting from Roman influence in the region. On the other hand, the evidence which goes alongside the northern German evidence to make that interpretation likely – the imports of metalwork, pottery and so on – are apparently absent from Scotland north of the Forth, where imperial imports seem to be an increasingly élite phenomenon. Furthermore, the rite itself, as noted, has fairly long antecedents in Scotland and why the introduction of cemeteries should be a product of Roman influence in the region is unclear.
Long cist cemeteries: a slightly different approach
Neither Christianisation nor imperial influence seem, in broader perspective, to be particularly compelling explanations for the appearance of Pictish long cist cemeteries. We should probably seek an account in developments within Pictish society. In this regard it helps to look at the burial’s surviving features, which might provide clues as to any statements being made during an inhumation about the deceased or his or her family. The fairly frequent analysis of the age and sex of the deceased gives us something against which to try and read off other variables: skeletal position, the arrangement of the limbs and so on. The construction of the burial chamber itself potentially contains differences that were meaningful to contemporaries. We could look at the organisation of the cemetery and whether it appears to be organised according to communal principles or grouped into apparent family plots. Alas, some variables undoubtedly noted during excavation – such as the positioning of the skeleton – are not systematically published.
The most important feature of change seems indeed to be the shift towards communal burial – or at least more communal than hitherto. The cemeteries’ frequent organisation into rows suggests (as with the famous Merovingian Reihengräberfelder) that the burial’s location and arrangement was to some extent governed by factors other than a desire to place members of the same family in close proximity. This might in turn mean that in their rituals, funerals ‘spoke’ to each other in quite direct and distinctive ways.
If the date of the cemeteries’ appearance, around the fifth century, is correct, then it took place against a background of exiguous settlement archaeology in the region, described by Fraser Hunter as a late third- to fifth-century ‘gap’. Combined with the evidence of greater dispersion of settlements in the period immediately preceding the ‘gap’ it is possible that the new cemeteries were used by several different settlements, providing a shared ritual focus. A similar phenomenon can be suggested for sixth-century northern Gaulish cemeteries, which also appear against a backdrop of scattered, ephemeral settlement evidence. Developing this, one might hypothesise that burial became a more public, communal occasion than had hitherto been the case and that a wider audience participated in the funeral process, now apparently including some sort of procession from the household to the cemetery. These features might make the funeral the occasion for statements about the deceased, his or her family, inheritance, and so on.
Nevertheless, from the archaeology as it stands, it does not look as though the funeral was the focus for particularly intensive competitive display. It did not involve the expenditure of resources upon grave-goods, although it is worth stating that even in areas where furnished inhumation was normal grave-goods rarely represent huge amounts of wealth. Be that as it may, the deceased’s social role or identity was evidently not symbolised through elaborate costume. Or, at least, not by costume adorned with fasteners and accessories or jewellery. It was certainly not demonstrated by the deposition of appropriate symbols. Such means of public marking of the deceased’s identity can be viewed as a symptom of instability and competition for local authority. The seeming absence of such displays and the investment of effort in tomb-construction instead may therefore be understood as its opposite. It is entirely possible that the funerary process was the occasion for other activities suggestive of competition and instability, such as feasting or the exchange of gifts, which might leave no trace at all in the archaeological record. That said, some evidence suggestive of these activities shows up in the Frankish evidence. We might wonder, therefore, whether the Pictish evidence overall suggests a society wherein social organisation and stratification were stable.
Such a conclusion could be no more than provisional and might be fundamentally misjudged; indeed, for reasons that I will shortly discuss, I think it would be quite mistaken. For one thing, the conclusions drawn from grave-goods rely, to no small extent, on observable change from one situation to another, and back again. The sheer, relentless uniformity of burial rite in northern Britain – where even cremations come in cists – over long periods of time, and the almost universal absence of grave-goods, renders very difficult the simple transfer of the hypothesis where grave-goods equate with insecurity and grave-monumentalisation with stability. Indeed one might valuably take the Pictish case to question the conclusions drawn from the Frankish. Again, though, the issue of differential change through time, with quite dramatic change in burial rites in Francia, renders the application of the Pictish evidence to the Merovingian situation as potentially inappropriate as its inverse.
Straightforward application of these particular conclusions about social structure – drawn from Frankish cemetery archaeology – is thus not possible. We must return to the variables observed within Pictish cemeteries, in particular to the choice of individuals for burial in the cemetery, was highlighted earlier. These patterns permit some suggestions about fifth- or sixth-century Pictish society. It should be said at the outset that these do not match either of the Frankish models although they reveal some important similarities with the sixth-century pattern.
When one considers why younger adult women
might be made the object of public funeral ritual,
the sixth-century Frankish model becomes useful.
In northern Gaul, adolescent and especially
younger adult women were the recipients of
significant grave-goods assemblages. Adolescents
are seemingly not heavily represented in Pictish
burials but the difference might be explained by
considering that between the burials of teen-age Frankish women and those of women above about twenty. The grave assemblages of the younger age-group are very largely concerned with a costume, which I have suggested related closely to their ‘marriageable’ status, and perhaps especially to their status as betrothed or married women. The grave-goods of women older than twenty were much less exclusively related to the adornment of body and costume but were generally more numerous, suggesting a funerary status connected to a role as the mother of young children. If we remove the aspect of costume, on the basis that Pictish costume is not visible in the burial data, certainly that jewellery was not interred, a concentration on women between the ages of twenty and forty remains and perhaps a similar explanation can be proposed. If one considers the life-cycle, not only of particular individuals but of the generations above and below them, synchronically, one might see that a woman who died between the ages of about twenty and about forty was likely to have died before her children had reached maturity, especially if she died in the first half of this age bracket. To accept the analogy, one must accept similar ages at marriage and bilateral inheritance among the Picts as among the Franks, both – of course – things we know next to nothing about. One can insist on that ignorance and reject any attempt to postulate a social structure or, more reasonably, one can insist on the difference between Frankish and Pictish society, better couched as that between a specifically post-imperial provincial society and one that had always lain beyond Roman law’s effective reach, than in terms of supposed differences between ‘Celtic’ and ‘Germanic’ societies, both of which are chimerical. But one only needs to accept bilateral inheritance, not all the details of the Frankish system. Equally, my model can incorporate a higher female and a lower male age at marriage without much difficulty.
If then we proceed – cautiously – with a limited Frankish analogy, then the concentration on the public burial of women between the ages of twenty and forty does not indicate the high status of women – as such – as much as the importance of women as the linchpins in alliances between kindreds. In this model, a woman’s death between these ages would mean the dissolution of a marriage alliance, probably before any children produced by the union had come of age; a subtle but important distinction. This would produce stress in relations between the families, especially concerning property and inheritance. The ensuing burial of such women in the shared cemetery might involve both families in the ritual, and efforts to smooth over that tension, resolve the issues and maintain the alliance. If I am right about that, then the earliest phases of the long cist cemeteries might suggest, rather than complete stability, a situation wherein politics operated at a very local, community level, dominated, as one might expect, by systems of marriage and other alliances between families. This could suggest more or less equality between kin-groups and a certain rivalry and competition between them.
The other observable variables – and change – within the Pictish burial record concerns above-ground monuments, in which we can include square and round barrows, cairns and probably the Class I symbol stones. Although, as far as I have been able to ascertain, none has yet been found entirely in situ in association with a burial, the most plausible interpretations of these stones see them as commemorative. As already implied, there is a crucial difference between the burial marked by an above-ground monument and the ‘flat grave’ in Francia and the same surely holds for Pictland too. Whatever the ritual and purpose of the standard long-cist burial in a shared or focal cemetery, the fact remains that there was little to record the burial and its significance, except perhaps some stone settings and earth up-cast once the grave was filled in and the mourners went home. As Howard Williams has correctly pointed out, in critique of my work, this does not mean that the transient ritual was without effecting enduring and specific memories about the deceased – quite the opposite. But nevertheless such memories were confined by the lifetimes of the witnesses. The above ground monument might have given information about the deceased that was pretty limited and 1-dimensional by comparison with the actual burial ritual itself, even where straightforward textual epitaphs were employed, but this information was nevertheless available, to those who understood it, potentially until eternity. This seems to me to mark an important social shift, concerning the degree to which families’ positions in local society were established and seen as able to be projected into the future with this sort of permanent mark on the landscape.
How might one account for this pattern? If one were to combine the evident concentration upon adults who (following the provisional interpretation sketched earlier) died before their children would be old enough to have established themselves, with the sorts of stress that such a death might cause in social relations and the marking out of some sort of difference or a claim to a permanent position in local society and the landscape, I might tentatively suggest the following. These burials are those from families with a newly emerging position within society which enables a more confident statement of that status. However, this position is not entirely secure and depends to some extent upon alliance, through marriage and so on, with other families of the same standing. Such alliances are brought into question by the death of wives, as discussed above, but also by those of fathers who die before their sons have come of age or younger adult males, perhaps similarly important as the linchpins of marriage alliances. They therefore need remaking through funerary ritual. This would in many ways be a similar mechanism to that discussed earlier for the bulk of the population of the cemeteries, but perhaps operating at a particular, slightly higher social level. As mentioned, though, this can only be a provisional attempt at an explanation but the pattern as it stands does seem interesting, and to require some such attempt.
What is also worth pointing out at this stage is that the appearance of these burials, and perhaps (if I am right) the emergence of a more elevated stratum within local society, is more or less contemporary with a series of other developments. The best known, perhaps, would be the most common earliest phases of the early historic occupation of the famous hill-top fortifications in Pictland and the neighbouring areas. It would also seem to be roughly at this time that the settlement pattern generally underwent some change, with various types of dwelling loosely datable to the second half of the first millennium. It does not seem to me unreasonable to focus an explanation of all this on the emergence of a more established local élite.
Up to this point, I have proposed the existence of a society wherein local social differences were not marked, succeeded by one within which a more established élite emerges, with the point of change lying sometime in the later sixth century. So far, as I am sure you are all thinking, ‘nul points’ for originality. This fits pretty well with the received wisdom, although I do not like the vocabulary of ‘kin-based’ and ‘ranked’ societies that tends to be used for the two social formations, and I hope that I have reached the conclusion via a slightly different route (or at least via one I have not seen in the literature available to me) and perhaps suggested some different points of detail. Where I hope to have some more novel ideas comes at the stage of moving from this conclusion to some thoughts about the relationship between all this and high politics, especially the development of the Pictish kingdom.
Interpretation: Social Development
An outsider can detect an understandable reaction in Pictish studies to old-fashioned notions of the ‘mysterious’ Picts, speaking their own pre-indo-European language, using their unique matrilineal succession system and carving mystic symbols in some sort of misty Scythian/Eskimo elf-land. This reaction seems, however, to have gone almost to an opposite extreme, postulating a ‘precocious’ statehood among the early historic inhabitants of Scotland north of the Forth. A ninth-century Pictish state (let alone a seventh- or eighth- century one) would indeed be quite something, seen in a broad European context, but suffice it to say that I have not yet seen any evidence adduced that would convincingly sustain a notion. This is not to deny the existence of an effective and organised kingdom, fully involved in the western European currents of political thought. That the Pictish realm was about as sophisticated a structure as most of its western contemporaries is wholly convincing. But an organised and effective kingdom is not necessarily a state. Even the eighth- and ninth-century Carolingian Empire cannot, for all its size, wealth and military might, convincingly be labelled a state. Until such time as we have evidence of the Pictish rulers’ ability to deploy an independent coercive force, and given the evidential constraints mentioned above I cannot imagine that such evidence is ever likely to be forthcoming, any reference to ‘the Pictish state’ will be premature.
Nevertheless, that something significant was going on in northern British society, whether Pictish, Scottish or British, in the last part of the sixth century seems fairly clear and it is not controversial to propose that this change is associated with the emergence of a social group that was more secure in its position and on occasion at least capable of drawing on manpower resources to construct defended élite settlements.
My objection to the more linear plotting of kingdom-growth is essentially based around two sides of the same, problematic idea: that the development of a local élite marches hand in hand with a growth in centralised power. I think this is a mistake but not one limited to studies of the northern British kingdoms; it detrimentally affects work on the origins of early Anglo-Saxon kingdoms too, for example, and I think it is the most significant methodological problem with Chris Wickham’s The Framing of the Early Middle Ages. One side of the problem is the fact that the absence of an established élite at a local level does not imply the absence of an effective, even a powerful, centralised government. If we take the comparatively well-documented sixth-century Merovingian kingdom as an example, we see this particularly clearly. The northern Gallic cemetery evidence shows that society was based essentially around age and gender, as I mentioned earlier. There is scant evidence of a local élite, independent of royal service, in the archaeological or the written evidence: I think Chris Wickham is wrong to propose that the Merovingian northern Gallic was unusually powerful and wealthy throughout the early Middle Ages. From about 600 onwards he has a point but in the fifth and sixth century the situation is very different. Anyway, what is abundantly clear is that the lack of a rigid social hierarchy in local communities by no means infers a lack of powerful royal authority. It is not problematic to call the sixth-century Merovingian kingdom a ‘state’; it taxed and it had a coercive military force, even if this did not take the form of a standing army. The key lies in the point just made about the absence of an established local aristocracy other than that based upon royal service. The best way of assuring dominance in the local community, breaking out of the ever-changing network of alliances and gift exchanges was through the acquisition of office in the royal administration. Such offices were extremely desirable as a result and it is clear that the kings rotated their patronage between different groups, distributing and redistributing their favours to ensure that no one faction acquired a dominant position. This meant that the kings held all the aces and that their authority was drawn down into local society along these arteries of patronage.
Many of the precise ways in which the Frankish realm worked descended directly from the late Roman Empire; indeed the mechanisms through which the early Merovingian state functioned mirrored fairly closely, albeit on a smaller scale, those through which the late Empire had operated. These precise things, for fairly obvious reasons, could not be postulated for their sixth-century Pictish contemporaries beyond the old frontier. That said, there were plenty of ways in which an analogous state of affairs could have worked, producing an extensive kingdom at least.
One of the most important problems in studying the early medieval Pictish (or Scottish, or British) kingdoms is the simple, brute, inescapable fact that we have absolutely no reliable information about political history before the period around 600. I have argued before that it is very questionable to try and project northern British political geography, as it appears in the written sources of the seventh century and later, back into the late Roman era, or even, I suggest, into the fifth and sixth centuries. One reason why is precisely the fact that so much seems to have been changing in the period when these areas begin to emerge into a dim light of political ‘histoire événementielle’. These changes probably determine the date of the earliest written data just as they govern the appearance of particular forms of archaeological evidence. It is – it seems to me – eminently possible that the kingdoms, as we see them in the fragmentary Irish and Anglo-Saxon records of the seventh century (very few of which are contemporary), might be quite recent creations. That is perhaps not a very novel thing to say, but what I think might very well be the case that, if they are new, then, rather than emerging from some sort of more or less inchoate tribal, ‘kin-based’ society, they might result from the fragmentation of earlier, larger-scale polities, perhaps the sort of extensive ‘Pictish’ confederation that the Romans thought existed north of Hadrian’s Wall. (And it’s worth reiterating that in late Roman political geography ‘Picts’ start at the Wall, not at the Forth.)
At least it should be conceded that the evidence as we have it accords this suggestion plausibility equal to that of its alternatives. Some points might lend it added credibility. Certainly, the period in which these regions, like those further south, begin to be visible in documentary sources, seems to have been one of political turmoil, extending far into the seventh century. Extensive overlordships are fought for, acquired and lost. A Scottish king is seriously defeated whilst evidently on his way to attack Banburgh in 603. A successor might even have reached and besieged the Bernician capital a decade or two later. A Northumbrian king’s son becomes over-king of the Picts and later one of his relatives exercises a fairly effective hegemony of Britain south of the Great Glen, until his son is killed in battle, when the pendulum evidently swings in the opposite direction. The instances can be multiplied. What seems clear is that the framework within which northern British – like southern British or Anglo-Saxon – politics operated was one of extensive overlordship. This might be a hang-over from a preceding ‘late antique’, confederate state of affairs. The rivalry for dominance within the realms between regional, familially- or dynastically-based aristocratic groups might be the new feature, linked with the developments that we can trace in the excavated record.
The third possibility, again equally plausible, is that the kingdoms visible in the seventh century existed earlier, and with the same extent but were somewhat different in their internal composition and political dynamics. The latter would be based upon the political power of local aristocratic élites and this leads me to the other side of the problems with the ‘linear’ model. The emergence of an élite more securely rooted in local socio-economic superiority makes things difficult for any central ruler: be it king, emperor or whatever. Unless a ruler has an independent coercive force, paid for by taxation (taxation levied with the threat of intervention by the coercive force), it becomes very difficult indeed for the writ of the monarch or the central government to run in localities that lie outside the ruler’s direct control.
The evidential problems of Pictish society make this a difficult issue to explore, let alone resolve. It might, for all we know, be the case that the Pictish kings did indeed have such a coercive force. In my past brief forays into Pictish history, I have tended to assume that the kings had some ability to call upon manpower, manifested in the construction of the hill-forts. There might be something in this. Perhaps sites like Dundurn are simply the seats of royal officials. Equally, though, it might be that these sites are the seats of local magnates. The first option gives us a kingdom ruling through officers ensconced in fortifications dominating local society; the second gives us aristocrats equally ensconced in dominating fortifications but potentially acting between the king and the remainder of his subjects. I can think of no obvious way of resolving the issue, but I offer the following as reasons for preferring the ‘local aristocrat’ reading to the ‘royal officer’ interpretation. One would be that it would coincide with the other evidence suggestive of the emergence of such an élite at about the same time as the earliest fortification. Another would be – perhaps – a difference from the earlier situation (as I understand it) which seems to have but few, larger fortifications (like Burghead). The change might imply some sort of fragmentation and it would lead to a third reason, which would be to ask why such lower-level ‘administrative’ fortifications only appeared at this time – which is not to say that such a question could not be answered. I think that, for reasons set out earlier, seventh-century history supports the idea of more fragmented power or weaker royal authority, but would gladly concede that it is hardly the only way of reading the evidence.
Perhaps a better reason why I prefer a weakening of power in the face of increasing local authority would be the fact that the above-ground monuments seem to be a fairly short-lived phenomenon. Over time, as the forts become more elaborate, the barrows seem to die out and the Class 2 Stones appear, evidently later in the seventh century. Alongside the foundation of churches all this looks – from a Frankish perspective – like classic signs of the entrenching of the power of local aristocracy. The later abandonment of the hill-forts at the same time as the establishment of other types of royal site, like Forteviot, also makes sense as a sign of royal authority asserting itself over such regional centres. The evidently ritual focus of site complexes like Forteviot is to be expected as the establishment of ideologies investing kings with religious sanction and loyalty to them as the ‘meet and right’ thing to do is a classic later first millennium – Carolingian, if you like – means of confronting the absence of coercive force.
All this nevertheless leaves the million-dollar question unanswered: why? And only have suggestions in response. One obvious starting point would be some form of economic development at a local level – again tied to the development of an aristocracy. It is of course at roughly this time that continental pottery finds its way to the north-east shores of the Irish Sea, a fact which should, I imagine, be linked to socio-economic development in the importing area, enabling exchange. Such imports do not make their way to the east coast in numbers but one might, given the archaeology, propose that similar things were taking place. One might still wonder why this might take place, not simply why economic development might have occurred at this time, but also, and I think more interestingly, why – if it did take place and if the increased surplus was being garnered by local élites for their own purposes perhaps in opposition to old structures of kingship – such élites were adopting this strategy. It is perhaps too easy to see – as I have tended to do in the past – a somehow natural or innate desire of local élites (or potential élites) to shrug off the overlordship of central or otherwise distant kings or overlords. What one might wonder is whether the changes that were taking place across Europe at this time, around the late sixth century, and which I currently would link to the end of the old Roman ways of seeing the world, had similar effects beyond the old frontier. Given the saturation of barbaricum with Roman ideas of power, at least east of the Rhine and south of the North African limes, it does not seem unreasonable to me to propose that such changes would have consequences in northern Britain too. Perhaps Pictish kings too had to find new ideological supports for their power, and for expecting people to conform with their desires, and perhaps this produced a crisis of legitimacy at moments of crisis as it did at this time in other areas. There were, again as in other areas, new ideological forms becoming available to local aristocrats to bolster their authority – in (as indeed in Gaul and northern Italy) Irish monasticism.
What I conclude – and I assure you that this was by no means what I thought I would conclude when I started thinking about this – is that economic, social and political developments in Pictland were very much in line with those taking place elsewhere around the end of the sixth century, that this part of Europe found itself very much in the same currents of change that were flowing through the rest of the West. What these developments imply, though, is not the straightforward growth of royal or central power, a movement up a neat linear trajectory towards the kingdom of the Scots. The road less travelled, that I am proposing, is much more uneven and twisting, and its destination more unpredictable but it is therefore, in my view, more interesting.