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 28 February 2012

Officers or Gentlemen? The Frankish Aristocracy in the Sixth Century: Part 4

[In the last installment of this piece (which begins here), I examine the role of royal service in maintaining aristocratic power.  It is argued to have been essential and this fact allows us to see the sixth-century Merovingian kingdom as a state in a very real sense.]
Part 3 is here

Gregory describes the redeployment of royal patronage on one occasion.  Following the suppression of the ‘Revolt of the Dukes’ (587) in Austrasia, Gregory says that a number of other dukes were demoted and others promoted to replace them.  All of the Frankish magnates whom we encounter in the pages of the Histories occupy positions within the administration of the realm, as duke, count, patricius and so on.  None of these titles, it is worth reminding ourselves, was hereditary; as implied in the preceding discussion, it is very difficult to find a title (even the general rank) remaining in the same family before the late sixth century.  Obviously it is impossible to know how they obtained their position in the hierarchy; one imagines that they probably came from the loosely-defined de facto aristocracy mentioned earlier.  Nonetheless, what emerges very clearly from Gregory’s account is the absolute dependence of these powerful aristocrats upon the royal court for their status.  We simply do not find an equivalent of the late Roman senatorial nobleman who spent brief intermittent periods in the service – negotium – of the state between longer spells of otium on his estates, or of the brooding late medieval noble plotting in his castle.  When demoted from office, such figures disappear from history.  Consequently the possibility that one might lose one’s position at court drove magnates to extreme actions, such as the Revolt of the Dukes just mentioned. 

The actions of the Austrasian dukes and those of the Neustrian aristocrats after Chilperic’s assassination (584) are very revealing.  The Neustrian magnates who found themselves away from court (generally accompanying an embassy to Spain) sided, on the whole, with Gundovald, a claimant with Byzantine backing.  The Austrasian dukes planned to murder Childebert II and replace him with his infant sons.  Clearly there were two principal options.  Ideally, an aristocratic faction wanted a malleable king (especially a child) whom they could control.  This allowed them to monopolise access to the court and to control the distribution of royal patronage, while still acting with the legitimacy of royal officers.  It is significant, therefore, that the Revolt of the Dukes occurred as Childebert II entered his majority (technically from Christmas 585); the faction of the dukes involved had controlled the court for most of his minority and were now threatened with demotion and replacement.  The Neustrian aristocrats who were at court in late 584 found themselves in the happy position where they could control the court of the infant Chlothar II.  On the other hand, the palatine magnates accompanying Princess Rigunth to Spain were now likely to find their return to favour barred by other factions.  Therefore they took the option of siding with an adult king, Gundovald.  An adult king, even if unlikely to be controlled by particular faction, at least had the corresponding merit of distributing and redistributing patronage between groups, ensuring that all had access and all had their turn.  This was better than exclusion from the court by a rival group.  Note, though, that there was no question of replacing the Merovingian dynasty.  By this stage the Merovingians had clearly established the idea that they were the only throneworthy family in the realm.  There was no Frankish noble lineage to match them; certainly none that could persuade other families to accept their pre-eminence.  Thus it is significant that one of the rebellious Austrasian dukes, Rauching, took to claiming that (like Gundovald) he was a son of Chlothar I.

Those excluded in this way tended to leave the kingdom altogether, as Duke Lupus did during Childebert II’s minority.  Similarly, after the Revolt of the Dukes, Gregory tells us that many men moved to other regions for fear of (pertimescens) Childebert II.  One assumes that they were linked in some way to the subdued rebellious faction.  Their fears were justified.  Childebert II was not a man to cross.  In one of Gregory’s more chilling stories, when the king and his court were watching bear-baiting at Metz, a certain Magnovald was dragged from Childebert’s presence and butchered, his body being thrown into the street as a sign that this killing had been carried out legitimately.  This was done causis occultis.  Some suggested that Magnovald’s alleged murder of his wife was the cause but no one knew; there was not even a show-trial.

Magnovald’s property was confiscated by the fisc.  This indeed was the fate of the property of all those who fell foul of the king: Guntramn Boso, Rauching, Eberulf and many others.  Some stories imply, furthermore, that the king was by no means necessarily disinheriting their heirs in such action.  Even those who were not in disfavour might have their property taken back into the fisc when they died.  The fact that the sons of another Bodegisel were able to inherit their father’s land without forfeiting any of it appears to have been a point of some note to Gregory.  The implications of the story are elucidated by another tale wherein, after their father was killed (evidently near Poitiers) in pursuit of a petty dispute, the sons of Waddo had to go to the king (again, Childebert II) before they were allowed to inherit his lands.  Childebert seems, however, soon to have changed his mind, stripped them of these estates and given them to his rebellious cousin Chlothild.  Royal interference in matters of inheritance is further suggested by Gregory’s repeated complaint that Chilperic of Neustria tore up the wills (testamenta) of those who died leaving land to the church.  This might in part stem, as Edward James suggested, from the fact that Pactus Legis Salicae did not envisage that the inhabitants of the northern Frankish realms could make wills at all.  Indeed no sixth-century northern Gallic wills survive.  If correct, this in turn would support the interpretation of the Law as having some practical import.  Another reason for Chilperic’s ire, however, might be that – given that the king is unlikely to have taken much interest in the social strata below that of the more powerful aristocrats – royal grants were not gifts in perpetuity and thus not the recipient’s to give away.  Parallels can be sought in a roughly contemporary antiqua (old law) in the Visigothic Forum Iudicum, concerning property acquired by a saio whilst in someone’s service.  Any weapons given by the patron may be kept if the saio leaves the patron’s service, but any other property ultimately remains that of the patron.  The Visigothic Law concerning the permanence of royal donations belongs to the seventh century, when it would – equally – evidently not be out of step with contemporary Frankish practice.  

Frankish practice might descend from late Roman, with the ‘salary’ of royal officers being paid through the grant of tax revenue from particular, defined assets – probably referred to as villae.  The existence of the system is confirmed by a story in the Histories and it makes sense of the apparent ease with which villae could be granted away, taken back and re-granted, as in the story of Waddo’s villa near Poitiers.  It also allows us to understand the meaning of the term villa as it appears in seventh- and earlier eighth-century usage.  A further advantage of this reading is that it makes sense of the social relationships witnessed in the Pactus Legis Salicae.  A freeman (ingenuus) granted the right to take the tax from a designated villa or small geographical unit would remain a freeman like the inhabitants of the villa; the latter would stand in no legally inferior or dependent status to him.  This means of salarying state officers would make sense in an effectively non-monetary economy like that of sixth-century northern Gaul.  The relevant surplus could be extracted and used by the recipient without leaving the locality, any responsibility for converting payments in kind into gold (whether from villae assigned to him as his salary or from other villae from which he was tasked with tax-collection) becoming that of the salaried officer in question.  The only currency we know about in Gaul at this time (outside a small-denomination coinage effectively restricted to Marseille) is imported Byzantine solidi and Gallic imitations struck by the Merovingian kings.  Such coins were of very high value and doubtless therefore circulated (alongside their late Roman predecessors) only among the most powerful socio-political classes, something that made these coins ideal vehicles for royal and imperial propaganda.  With a more or less fixed content of 1/72 of a pound of gold, they should, however, be understood more as convenient units of bullion than as coins proper.  Even in the monetized late Roman period, the value of the solidus to the other elements of currency had been inflationary.  The only other coinage in use appears to have been residual Roman currency, which is frequently found in the pouches with which many Merovingian males were buried.  When taken alongside the presence of balances in lavishly-furnished male burials, we might conclude that local leaders, and especially (one assumes) those with an official title, had a responsibility for ‘weights and measures’.  These old coins presumably circulated as bullion and it was a local leader’s word that determined their worth as pieces of precious metal.

Taken together, the advantages of royal service were clearly considerable for northern Gallic aristocrats.  Legal privilege was only one of these.  In addition, responsibility for taxation and for levying the army from an increasingly broad stratum of ‘Franks’ constituted an important source of local patronage.  The revenue from royally-bestowed villae could be considerable, especially where numerous such ‘fiscal assets’ were accumulated by an officer like a duke or count.  Royal gifts were also prestigious as well as valuable.  Thus it is not surprising that the competition for patronage was fierce and that this outweighed any nascent magnate identity.  It is not surprising that a king like Childebert II could very rapidly call out an army of a sufficient size to spell doom immediately, even for an allegedly large force assembled by the rebellious dukes Ursio and Berthefried.  With all these points made, it is clear that the early Merovingian kingdom can and should be considered a state.  It had the ability to raise an army and use it as an independent coercive force, as well as being able to pay its officials through a, however obscurely, functioning taxation system.  It thus had the means to make its writ run in the myriad communities that were located within its bounds.  Ideologically too, it appears to have managed to monopolise conceptions of legitimate power.  Nowhere is this clearer than in the ability of the third generation of Merovingians (the sons of Chlothar I) to maintain a tight grip on authority while rarely taking the field at the head of their armies.  Unlike other realms, the Frankish kingdoms endured serious military reverses in this period (at the hands of Bretons, Goths and Lombards) without political crises ensuing.  On the other hand, while failure does not seem to have redounded to the kings’ shame, any credit for victories does seem to have allowed the Merovingians to maintain the ideology of triumphal rulership.



When the Merovingians took over the governance of northern Gaul in the later fifth century they were, with few exceptions, not confronted by a powerful, independently wealthy magnate class.  Over the next century, the kings managed to reduce the status of this group further, so that it became yet more dependent upon royal service for its local pre-eminence.  The kings’ actions should probably not be regarded as policy.  Many can be viewed as continuations of the later Roman emperors’ style of governance, maintaining checks and balances and the even distribution of royal favour.  Because of this they find parallels in the activities of other post-imperial kings.  It may be no more than the happy genealogical accident that the Merovingians were able to maintain father-to-adult-son succession for four generations between Childeric I and the sons of Chlothar I that enabled them to be more successful than any other post-imperial dynasty.  Nonetheless that factor reveals the inherent weakness of the system.  Much like the late Empire, political stability relied upon the tenure of the throne by an adult capable of distributing and redistributing patronage and maintain balance between different aristocratic factions.  Again like the Empire, a series of royal minorities, in this case after Sigibert I’s assassination in late 575, produced a critical situation.  This occurred, furthermore, at a point when other contingent factors could come together to exacerbate any crisis in the legitimacy of government carried out in the young king’s name.  The transformations in the nature of the Frankish aristocracy that occurred as a result of this combination of circumstances lie outside the scope of this paper.  As the sixth century entered its last quarter, contemporaries could only have perceived the Merovingian kings as exercising a tight control of legitimate power and a literally life-and-death authority over their aristocrats.  In these circumstances the idea that the latter formed an independently powerful social class would have seemed a long way from reality.


Officers or Gentlemen? The Frankish Aristocracy in the Sixth Century: Part 3


[The third installment of this draft article on the sixth-century aristocracy (which begins here) looks at the significance for the debate of the legal evidence and of terminology for aristocrats in the narrative sources.  The argument is that both forms of evidence, while they permit the reasonable supposition that some families maintained social pre-eminence within their localities, this position was not secure or recognisable by southerners as amounting to a real form of social distinction.  The failure of the laws to recognise an aristocratic stratum might have originated as a royal strategy but by the middle of the sixth century it seems to have been a reasonable approximation of reality.]

Part 2 is here

In this context we can reconsider the earliest of the Frankish law-codes, the Pactus Legis Salicae, and its failure to mention any sort of a noble caste.  The Pactus only recognises two types of freeman: the Frank, with a wergild of 200 solidi and the Roman tributarius, whose wergild was 150 solidi.  The only men raised above this general stratum are the presumably Frankish members of the trustis regis (the royal retinue: wergild 600 solidi or double that of the Frank), the Roman members of the convivia regis (the royal table: wergild 300 solidi or double that of the tributarius), the pueri regis (the “King’s Boys”) and comites (companions; here meaning counts: royal officers).  All of these exceptions are defined by a connection to the king.  As mentioned, the interpretation of this fact has been a matter of debate.  On its own, it is difficult to make the absence of a legally-recognised aristocratic stratum bear much weight in proposing that such a class did not exist in reality.  Arguments against it have been numerous.  The kings deliberately did not recognise such a group as a means of strengthening their own position.  Alternatively, the law-makers felt that they could not legislate for a noble wergild; in practice the magnates would set their own blood-price.   Another possibility is that the class of ‘Franks’, with a higher wergild than ‘Romans’ is itself the aristocracy.  Or the coverage of the Pactus is hardly comprehensive and the absence of laws specifically relating to the aristocracy might not therefore be significant.  Indeed there are legal ‘documents of practice’ which decree that ‘Salic law’ be followed in circumstances not covered at all by the Pactus.  Finally, the law itself was only a symbolic text with no necessary grounding in reality; texts actually showing that the Pactus and its provisions were followed in practice are entirely lacking.  By contrast we have later documents recording legal decisions that flatly contravene what the Pactus said should be done.

These arguments against the significance of the Pactus’ silence are of uneven weight but there is something in all of them.  The weakest – like a crude progression from the law’s silence to actual absence – turn on the idea that the law was directly applicable in practice.  In the sixth century, the extent to which this was the case is impossible to gauge.  In defence of the idea that there was some connection between code and practice, we can point to some similarities between the processes it envisages and those described in narrative sources and the fact that the law set geographical boundaries on the area within which it thought it was applicable (between the Ardennes and the Loire).  The best way forward seems to be to remember that the law was a deliberate composition, not a simple, passive transmission of age-old custom.  Even if, as seems very plausible, it was assembled at least in part from pre-existing royal decrees, the Pactus is very consistent in its vocabulary, especially when using ethnic terminology (here it differs from Burgundian law, for example).  It does not seem unreasonable to hypothesise that its authors assembled materials, and phrased their discussion of issues, in a deliberate fashion, whether this was to provide a practical law-book, or simply an object text that symbolised Clovis’ capacity as a law-making king, in Roman fashion. 

Viewed thus, several suggestions are possible.  One is that, as Chris Wickham wrote nearly twenty years ago, it is significant that the range of relationships and legal situations envisaged by the legislators did not encompass the activities of major, wealthy landlords or their intrusions into a society of small-landowners.  For it is not simply the absence of a recognised aristocratic stratum, with its own wergild, that matters.  There are no references to the relationships that suggest the existence of such a class, either.  This becomes clear through comparison with seventh-century Lex Ribvaria.  Comparison with other codes suggests that the failure to mention a magnate stratum with a higher wergild than the normal freeman is deliberate.  We should not expect this absence, as a matter of course, because of the de facto ability of aristocrats to set their own blood price.  Burgundian, Anglo-Saxon and Lombard laws all recognise such a class.  Furthermore, why ordinary aristocrats could not be legislated for because of their ability to set their own wergild, but aristocrats with royal backing could be so restricted is a problem left unanswered.  Thus the lack of recognition of an aristocracy with status independent of royal approval might indeed have been a deliberate royal statement.  This need not imply that such a group did not exist in practice but the absence from the law of other indices of its presence, which comparison with other codes suggests would normally act to restrict its power, argues that it was not, in reality, very significant.

Other sixth-century evidence permits the argument that even if, when the Pactus was issued, the law-makers’ deliberately ignored a magnate class as a statement of royal power, by the middle of the sixth century the dependence of northern Gaulish local leaders upon royal legitimation was a normal state of affairs.  This is suggested from the analysis of the region’s rural cemeteries during the middle quarters of the century, as already discussed.  That reading of the excavated data finds support in some aspects of the written sources.  The first such aspect to consider is that of terminology.  Gregory of Tours, for example, only employs the term nobilis once to describe a northern Gallic family, referring to the brothers Bodegisil and Bobo.  Otherwise, within Gaul, Gregory restricts his use of the term nobilis to the Merovingian royal family itself, and to southern Gallic senators.  His description of an embassy to Constantinople, one of the occasions when Gregory refers to Bodegisil, suggests why this might have been.  He describes the ambassadors as ‘Bodegisil son of Mummolenus, from Soissons, Evantius son of Dynamius, from Arles, and Grippo, a Frank’.  For Gregory, then, it seems that one’s civitas of origin and the name of one’s father were the principal criteria for inclusion in the social stratum that encompassed the senatorial nobility.  In this, he does not seem too different from his near contemporary Isidore of Seville, whose definition of nobilis, within the category of ‘terms for people’ (de quibusdam vocabulis hominum), was one whose ‘name and lineage are known’.

Wickham downplays the significance of Gregory’s failure to use the word nobilis for northerners, by stating that Venantius Fortunatus, unlike Gregory, does employ the term for Frankish aristocrats.  In fact, scrutiny reveals Venantius’ use of the term to be more or less the same as Gregory’s.  He describes as ‘noble’ southern Gallic senators, the Merovingians and one northern Frank, who dwells in Toulouse.  Otherwise he uses the word to qualify virtues or objects in a religious context.

Nonetheless, looking at the usage of particular terms is not without its problems, given the almost total lack of written evidence from northern Gaul itself.  The sources available are essentially from Aquitaine or – in the case of the works of Venantius Fortunatus – of a poet from Italy.  One issue is whether the failure of these writers to discuss the northern, Frankish aristocracy in the terms usually employed for nobility is significant, or at least in what ways it is significant.  If the northern Gallic magnates were not recognisable as noblemen in the traditional Roman way, does that really mean that they cannot – functionally – have formed a similar type of ruling stratum?  There are, after all, other words which are used to describe northern aristocrats.  Some – seniores, proceres, potentes, or optimates for example – do not have a bearing on the precise enquiry as their general meaning of ‘leading men’ does not necessarily carry with it an implication about the relationship between their status and royal service one way or the other.  Other phrases do, however: maior natu for example.

This is important because, if nothing else, it demonstrates that no position taken on this topic is likely to represent a hard-and-fast ‘right answer’.  Indeed none of the argument presented thus far implies that there was no Frankish aristocracy, defining the latter in a loose sense as a group of people more powerful, in de facto terms at least, than their fellows.  The analysis of the archaeological cemetery evidence discussed earlier does not by any means suggest that some families did not remain dominant within their localities right through the period from the fourth century to the seventh.  Members of such kindreds might indeed be described as maiores natu.  In some areas, in any case, what we might term a noble class – using the word noble to distinguish an aristocracy more narrowly defined by its birth and culture as well as its wealth and standing – certainly did exist.  The ‘senators’ of Trier are the clearest example. 

Another point that must be made is that royal service contained mechanisms that could foster some of the characteristics of such a nobility.  Service in the schola palatina – the royal military household, where the training of the Pueri Regis presumably took place – clearly included literary education, enabling Frankish service aristocrats to partake in Latin literary culture to some extent at least.  Bonds forged between the Pueri themselves could create a sense of group solidarity (maintained by an epistolary culture derived from that of late imperial times), as is most famously attested at the end of the period covered by this paper with the circle of Desiderius of Cahors, which was formed when in the service of Chlothar II and Dagobert I.  Royal service, like late imperial, also brought with it certain honorific titles.  Viri illustri are attested and, in the capitularies, even viri magnificentissimi optimates.  These honorifics might also be used to foster a class identity amongst an essentially service aristocracy.  Such titles, however, were not hereditary. 

Nonetheless, conclusions are possible.  It is clear that, to writers like Gregory and Venantius, most of the Frankish aristocrats – even the powerful ones – whom they met or who featured in their accounts of sixth-century history did not, unlike the Merovingian kings themselves, meet the criteria for nobilitas normally used by their own class.  Some people who appeared in Gregory’s other tales or who commissioned poems from Venantius seem, in a local context, to have been members, by birth, of a stratum of leading families.  It would be difficult indeed to envisage a situation, even within the fluid sixth-century context proposed here, where such families did not exist.  What is at stake is the security of such families’ power and the relative  of royal patronage and of local pre-eminence in terms of lands, wealth and so on.

To shed further light on this issue, one obvious source of information is Gregory of Tours’ Histories, with its account of the doings and, usually, violent ends of numerous Frankish aristocrats.  Gregory’s evident refusal to incorporate the Frankish magnates within his category of nobiles has been mentioned.  One reason for this might have been the fast turn-around of post-holders within the Merovingian realm that resulted from the kings’ adept control of patronage.  This would usually mean that Gregory (and Venantius) was rarely aware of the names of these aristocrats’ fathers or their family relationships.   It is interesting to speculate, too, on the self-presentation of such men, however, and to wonder whether they presented themselves simply as, for example, ‘Grippo the Frank’.  Perhaps patronymics were not habitually given precisely because aristocratic southerners would likely be none the wiser for it.  By way of modern English analogies, questions such as ‘what school did you go to?’ or (for instance) ‘would that be the Hampshire Fortescue-Smythes?’ are questions that can only be, and therefore are only, properly answered by members of the same, privileged class as the questioner.  They aim primarily at elucidating that class-relationship and only secondarily at extracting actual information.  They do not necessarily put all the power in the hands of the questioner as difference can be just as aggressively asserted by a negative response, but social space is nevertheless marked by the exchange.


Part 4 is here

Officers or Gentlemen? The Frankish Aristocracy in the Sixth Century: Part 2


[In this second chunk of the article, I look at the dynamics of royal-aristocratic relations and how the maintenance of patronage could work to undermine the local status of aristocrats.  Archaeological evidence is then adduced to show a further weakening of local aristocratic power by the middle quarters of the sixth century.]
Part 1 is here

The exception to this rule, just mentioned, was the aristocracy of the Triererland.  This serves as an instructive case study of the problems that the Frankish kings might have faced, and how they dealt with them.  The aristocracy of the middle Moselle Valley had been closely connected with the imperial Roman presence in the region, when the western Empire’s main capital was located at Trier (western Emperors residing regularly at Trier between the 280s and 388). As well as holding large estates and lavish villas, which might have survived the early fifth-century crises better than those in other regions, it had a powerful sense of its own Roman identity.  References to the senators of Trier are known – uniquely for Merovingian Gaul north of the Loire.  Indeed, the rapacious aristocracy of northern Gaul, so long (probably mistakenly) employed as a paradigm for the late Roman aristocracy, may be specifically located to this region.  Their senatorial identity is represented not only in the statements of the evidence but also in its form.  Trier has produced about a quarter of all the late antique epigraphy from Gaul, and almost all such data found north of the Loire.  In regional context, this Roman means of commemoration stands out like a sore thumb.  As well as a self-confident Roman aristocracy, Trier also possessed an episcopate with an awkward tradition of speaking out against secular rulers, actively maintained during the first half of the sixth century by Saint Nicetius.  With all this in mind it is perhaps unsurprising that the attempts by the Merovingian kings of Austrasia to establish themselves at Trier failed.  By the middle of the sixth century they had abandoned the former imperial capital and moved up-river to Metz.  Against the physical backdrop of the old imperial buildings (in one of which, the cathedral, Nicetius staged one of his confrontations with King Theudebert of Austrasia) and against the ideological background of Justinian’s attempted reconquest, accompanied as it was by the proclamation of the West’s loss to barbarians, any early Merovingian efforts (like Theudebert’s) to present themselves simply as legitimate successors to the Roman Emperors – as emperors in new guise – was surely doomed to failure.  When the audience for such attempts was composed of a noble stratum far more confident about its traditional Roman credentials, their failure was assured.  

The Merovingians therefore, like most other post-imperial rulers, betook themselves to lesser Roman centres, like Paris and Metz.  Metz was a Roman city which offered all the elements of an ‘imperial complex’ like that at Trier or Constantinople: a cathedral, and audience chamber and an amphitheatre.  These however had not been unified into such a complex by the emperors.  Metz thus presented a suitably Roman ‘canvass’ upon which the Merovingians could nevertheless inscribe their own political identity.  Here, moreover, the Merovingians could appoint their own men to the episcopate, largely removing the threat of awkward confrontations with bishops.  As a result of this attention, Metz had, by the middle of the eighth century, largely eclipsed Trier.  Used to having their rulers living amongst them, the Trier aristocracy now had to travel up the Moselle to Metz to obtain the patronage of the Merovingian kings.  Once there, of course, they found themselves in competition with the kings’ own servants and with aristocrats from across the rest of Gaul.  Their local wealth and pre-eminence availed them little in this context.

This relationship between the royal court and local aristocrats is paradigmatic for understanding early Merovingian northern Gallic social structures.  Royal patronage was even more important to the aristocrats of the other parts of the region, whose wealth and familial standing were less secure than those of the Trier ‘senators’.  Reduced to precarious authority even within their own localities, the best means of cementing this leadership and of involvement in politics on a wider stage (such as had been possible to their fourth-century ancestors) was to obtain the patronage of the Merovingian rulers.  This dynamic is important as, as already noted, it removes the need to assume a widespread programme of political assassination or purges.
    
By the opening of the second quarter of the sixth century, which this paper takes as its rough chronological starting point, other developments had further strengthened the Merovingians’ hand.  In addition to the removal of rival branches of the Frankish royalty, already mentioned, which (as is less widely appreciated) continued into the reigns of Clovis’ sons, the defeat of the Visigothic kingdom and the acquisition of Aquitaine from 507 onwards was a decisive moment.  Although the conquest of southern Gaul was a more complex and drawn-out process than is often imagined, with much territory retaken by the Goths under Theudis and thus having to be conquered again (perhaps as late as the 530s), by 535 Burgundy and Provence had also been added to the Frankish kingdom.  These conquests massively increased the Merovingians’ store of patronage.  Northern aristocrats could be rewarded with offices (such as those of count or duke) south of the Loire.  This may not have happened on a large scale but it did not have to do so to have crucial effects.  The imposition of northerners as local civil and military governors, with royally-bestowed rights to collect taxes, hear court cases and levy the army, will have made a considerable difference to a southern Gallic nobility, used to having a powerful Gothic ruler living amongst them for the best part of a century.  Now they had to compete with Frankish royal officers for the favour of a monarch who dwelt to the north, rapidly becoming Francia.  While this dynamic enabled the Merovingian realms to secure their control over Aquitaine, Burgundy and Provence, to a greater degree than their Carolingian and later successors, it similarly maintained the kings’ ability to play off the northern aristocrats against each other in the contest for royal patronage.  This competition for patronage could be played out in various formal settings.  Gatherings of the army, the chief political assembly of the kingdom, and of the Franks especially, were occasions for the kings to reward and punish, as the famous story of the Vase of Soissons makes clear.  Gregory calls this a meeting on the campus martius.  Whether this was named for a place, on the model of the campus martius in Rome, or for a date (1 March) is not clear.  Certainly an annual assembly on 1 March (the Marchfield) seems to have been in existence by the 590s, when Childebert II issued all three of his edicts on that date.  It is possible that by the 590s a fusion of the date and the location had occurred.  Some adaptation of Roman practice is probable.

Thus the kings had, by the second quarter of the sixth century, established themselves in a position of more or less unassailable authority within their realms.  There were other policies that helped to underline this position.  One was their well-known practice of avoiding marriage with the daughters of their aristocrats, preferring low-born women and foreign princesses.  This ensured that no Frank could claim membership of the family without the recognition of a reigning king.  Royal daughters, too, appear to have been kept out of the marriage market, except for alliances with neighbouring kingdoms.  Though decried by some contemporaries, and whether or not it was designed to do so, the division of the realm in fact also helped to bolster the Merovingian supremacy.  Opposition to one king only crystallised around his brother or cousin in another of the Teilreiche.  More probably based upon the pragmatic division of the Roman Empire than upon any kind of ‘Germanic’ inheritance custom, the provision of more than one court facilitated admittance to royal presence.  This further enabled the tight control over the distribution of patronage mentioned earlier.  While this was to the kings’ advantage, the ease of access and the prevention of a monopoly of the court by any particular faction were doubtless seen as aspects of good governance by many aristocrats.

The kings’ dominance and their evident reduction of local leaders to the level of a service aristocracy can be seen further in social and economic developments of the second and third quarters of the sixth century.  This can be traced archaeologically in two or three areas.  The first is the further deterioration of the late Roman fine-ware tradition.  By 540, Argonne Ware production, which had hit a critical point in the mid-fifth century, entered a final phase of decline.  Between that point and its final disappearance from the record around 600, only a limited range of undecorated forms were produced.    A similar trend might also be manifest in the fact that artefacts of Böhner’s Stufe II (c.525-c.575/600) manifest much less craft-specialisation than was visible in, for example, the polychrome garnet-inlaid objects of the late fifth-century Flonheim-Gültlingen Horizon.    Although the belt seemingly remained a key item of masculine costume, the belt-buckles of the middle quarters of the sixth century are for the most part plain, unadorned ring-and-tongue designs, sometimes with a simple incised decoration on the loop or tongue, occasionally on the ‘scutiform’ terminal of the latter.  Where plaques are attached these too are often undecorated.  Most of the decoration that we see is invested in female apparel such as brooches but here too the objects are, for the most part, simply cast.

One reason for this must be that the security of local leadership was now even less than it had been in the preceding 100 years.  Surplus was now spent on a cycle of rituals that involved the display and consumption of resources, the bestowal of food, weaponry and other movables or parcels of land in return for alliances and support.  Such rituals clustered around the life-cycle, particularly the processes of socialisation (evidently a long, many-staged procedure for males), betrothal and marriage. Death and burial were the other key focus for such activity and it is from the practices associate with this that we have our most plentiful evidence in the form of archaeological cemetery data.  Between c.475 and c.500, in the Flonheim-Gültlingen Horizon proper, most furnished burials were of adult males.  In the subsequent quarter of a century, children and females became more common, giving the phase some similarities with the archaeological horizon of the so-called Föderatengräber in the late fourth and earlier fifth centuries.  We appear to have some locally prominent families marking their status within the community through the use of elaborate ritual. The increase in numbers of such graves seems to reveal a growing number of such kin-groups using material culture linked with the Franks to symbolise a claim for local authority.  The gradual use of the burials of women and children as foci for display suggests that the competition for local authority was becoming more severe, as deaths of all sorts of family-members could bring status into question and require smoothing over with ritual activity.  This activity not only displayed claims for legitimate authority to an audience assembled for the funeral, via appropriate material cultural symbols and emblems; it appears to have been accompanied by feasting and the bestowal of other gifts, maintaining or strengthening ties with other kindreds within the community.  A burial appears to have been an occasion to show that a family could inter one of its members with the appropriate obsequies to the best of its ability.  Within this, there was – obviously – a competitive element.  It is important, therefore, to underline that the ‘wealth’, or otherwise, of a burial is not a passive index of the formal rank or status of the deceased or his/her family, but an active claim to local standing.

By the second quarter of the sixth century, however, this practice had become very much more widespread.  For example, at the cemetery of Lavoye (dép. Meuse), there were only a dozen graves in its first phase (c.475-c.525) but about two hundred in the phase spanning the remainder of the sixth century.  Of these, at least three quarters were buried with grave-goods.  Rather than being clustered with particular families, though, grave-goods were now very clearly distributed according to the age and gender of the deceased.  Community norms appear to govern the forms of material deemed appropriate for individuals of particular ages and gender, as well as the overall lay-out of the cemetery itself.   Clearly the extent of competition for authority had increased with the participation of a wider range of families in the rituals that served to augment as well as cement local relationships.  Burials which are marked out from their contemporaries have to be sought more subtly than in previous phases, usually marked by having more of the correct forms of artefact – and especially gender-related artefacts – for a person of that age and sex.  More obviously unusual burials are nevertheless known, and on some sites distinction is created through the breach of the cemetery’s usual norms.

Thus we can propose from a variety of archaeological sources that authority within a community was now much less secure than had earlier been the case.  In these circumstances a connection with the royal court and its patronage was one of the best means of securing such pre-eminence independently of local political allegiances. 


Part 3 is here

Officers or Gentlemen? The Frankish Aristocracy in the Sixth Century: Part 1

[This is the second part (the first installment of the first article can be found here) of what started off as a single article about the changes in the Frankish aristocracy around 600 but which has turned into a trilogy!  This piece, again broken up for ease of reading, deals with the period between c.525 and c.575 and argues that the Merovingians transformed the local elite of northern Gaul (not hugely independently powerful in the late Roman period) into a service aristocracy in this period.  This first chunk deals with the historiographical background.
As with 'The Genesis of the Frankish Aristocracy', this piece is still very much in draft form and lacks notes or any scholarly apparatus.  Any comments are welcome - indeed invited.  Again, once the piece is properly worked up and submitted for publication I will probably remove it from the blog.]


This paper examines the nature of the northern Gallic, or Frankish, aristocracy of the Merovingian kingdom during the sixth century.  This has long been a controversial matter, with some scholars arguing for the existence of a hereditary élite class, independent of Merovingian royal legitimation, and others that the magnate stratum was dependent upon royal favour for its social, political and economic pre-eminence.  The bulk of this debate has concerned the interpretation of sixth-century written sources (or in some cases their lack), which have usually been viewed in isolation or interpreted in line with preconceptions drawn from the analysis of other types of data.  This paper looks at this evidence in broader context, addressing the problem on the basis of the full range of available sources of information.

Most of the areas of debate will be touched upon below but a brief overview of the stances taken by past writers will perhaps be useful in providing some orientation for the reader.  On the one hand, some historians have argued that the Frankish aristocracy existed simply as the extended household of the Merovingian kings, in other words as a service aristocracy.  I will term this, unsurprisingly, the ‘service aristocracy’ reading.  The principal support for this position is the absence of any reference to an aristocratic caste in the earliest Frankish law-codes, the Pactus Legis Salicae of the early sixth century and the Lex Ribvaria of the early seventh.  The only individuals recognised as more privileged than their fellows are those with a connection to the king.  Backing this up have been various technical arguments about the laws’ provisions for land-owning and the apparent absence of an independently wealthy and powerful Frankish aristocracy in the Histories of Gregory of Tours.

Opponents of this view (adopting the ‘established aristocracy’ position) do not necessarily agree with each other on the details and take ‘Germanist’ (seeing the Frankish aristocracy as a direct continuation of that which existed in the pre-migratory Frankish homelands) or ‘Romanist’ (stressing a greater degree of continuity with the region’s Gallo-Roman aristocracy) approaches in varying degrees and admixtures.  The evidence for this view is, in some ways, more exiguous than that for the ‘service aristocracy’ interpretation.  First, it turns on counter-factuals: various ways of explaining away the laws’ silence about an aristocratic rank.  Then, more reasonably, it adopts alternative interpretations of the snippets of information found in Gregory’s writings.  An apparently stronger plank of support for this reading is provided by the lavishly furnished burials found in northern Gaul between c.475 and c.525, the wealth of these burials being taken as a reliable index the deceased’s social rank.  This interpretation has some variations.  The similarity in rite between the lavish burials of this archaeological horizon and that of the Frankish king Childeric I in Tournai has been used to suggest that the men interred in these graves were the followers, perhaps the leudes, of the Merovingian royal house.  An alleged southwards spread of such burials in line with the presumed Frankish advance further supported this reading.  Of course, if the leudes of Childeric and Clovis were just service aristocrats, these graves might support the alternative interpretation.  The debate would here concern how one understands the nature of early Merovingian leudes.  Traditionally, however, the wealth in these burials has been held to indicate a more secure and established social pre-eminence of the families whose dead were laid to rest in these tombs: something more than simple royal officials.

These data apart, the ‘established aristocracy’ reading relies upon four approaches.  One is the employment of unsubstantiated (indeed largely unsubstantiatable) assumptions about the pre-migration Frankish aristocracy (often grounded in traditional beliefs about ‘early Germanic’ society, held to be universally applicable across all Germanic-speaking barbarians).  Another derives from preconceptions about the local Gallo-Roman aristocracy, equally based upon importing notions of late Roman aristocracy thought to have Empire-wide relevance although founded in data that is quite restricted in its geographical and social provenance.  Whether these ideas are applicable in this way to the northern Gallic aristocracy has not always been interrogated.  The third approach has been to take evidence from across the Merovingian period.  This view is based upon two premises.  One is that early Merovingian data are so few that (as might be clear from the summary above) they are incapable of furnishing any decisive argument for either ‘service’ or ‘established’ readings.  Thus one needs to bring in other, later material.  This method is associated with a ‘regressive’ approach, holding that the picture of the aristocracy that emerges in the better-documented seventh century can be extended backwards to the late fifth-century Merovingian take-over of northern Gaul.  Quite sophisticated defences of this approach have been offered, based around the shared Roman-Christian heritage of the early Middle Ages and the fundamentally unchanging bases of social distinction (essentially, land and honour) in the era.  The fourth approach, not dissimilar from the ‘regressive’ method, identifies features which the late Roman magnates and region’s equally powerful ninth-century Carolingian Frankish nobility shared and link the two via a straightforward line of development through time.  Nonetheless, however reasonable one may think this quadruple set of assumptions to be, it cannot be denied that it contains no decisive element.  Each view can equally reasonably be countered, even if the somewhat depressing alternative offered is simply that ‘we do not know’.

A new consideration of this well-worn topic is timely because the subject, having effectively lain dormant for some years, has been recently reinvigorated by the discussion in Chris Wickham’s The Framing of the Middle Ages.  Wickham sides with the historians who have followed the ‘established aristocracy’ reading, probably the majority interpretation amongst those who have considered the issue.  Indeed, he argues that the northern Gallic aristocracy was, within the context of the early medieval West (and perhaps even farther afield) an exceptionally wealthy group of landowners.  Wickham’s reading contains more nuance than most previous versions of this argument and an engagement with the full range of material, archaeological and documentary.  In particular, he introduces an economic dimension to the discussion, based upon study of ceramic materials.  Such an analysis was not possible even twenty years ago, and therefore adds importantly to previous archaeological investigations of the topic, which have concentrated almost exclusively upon the evidence of excavated cemeteries, the distribution of grave-goods and the existence or otherwise of ‘tombes de chef’ or ‘Adelsgräber’ (alluded to above).  Wickham poses the question of how the newly established Merovingian kings, confronted by an already extant Gallo-Roman aristocracy in the region and by Frankish noble families, could have reduced all of these magnates to the status of mere service aristocrats.  Surely, he says, the extermination or forcible removal of such people on a region-wide scale is not to be envisaged – a conclusion with which it is difficult to disagree.  Wickham’s argument also relies to some extent (though again to a lesser extent than with previous writers) upon the extrapolation of the early Merovingian situation from the better documented seventh and eighth centuries.  In the late Merovingian period, let alone the Carolingian, Wickham’s conclusion about the wealth of the Frankish élite is hard to dispute – whether or not it was as exceptional as he claims is a question that this paper does not and cannot address, but the case is well-made and plausible. 

Another reason for a new look at this old subject, implicit in my discussion of Wickham’s argument, is the dramatic increase in the range and quality of archaeological data since about 1990.  In the early 1990s it was still possible to criticise Merovingian archaeology for being excessively rooted in very traditional readings of the cemetery evidence.  At that point, although excavation of rural settlements of this period was beginning to take place, exploration (and, even more so, publication) of such sites remained scarce.  This has changed dramatically, presenting us with valuable insights into dwellings, spatial organisation and rural economy, an invaluable counterpart to the display of status visible in the cemeteries.  In addition, studies of the artefacts produced on these sites have increased in number and sophistication.  Most notable here has been the attention to ceramic data, and the construction of typologies and chronologies for the pottery found on rural sites.  Again, this body of data, whilst coming into being, barely existed in 1990.  These developments have emphasised one of the most valuable aspects of archaeological evidence, the ability to pay attention to regional and chronological diversity and precision.  Using these data we can avoid treating the north of Merovingian Gaul as an undifferentiated mass in regional as well as chronological terms.  Such a variegated picture is not possible from the written sources until well into the Carolingian era. 

Nonetheless, the diversity revealed by the excavated data allows us to look at the documentary evidence in a new light.  Rather than heaping all of the written sources together to make a ‘pile’ that looks more statistically significant, we can consider each source according to its genre and provenance in time and space and assign more significance to the differences between them.  The approach adopted is based upon the use of as wide a range of forms of data as possible – documentary, epigraphic, numismatic and archaeological – each treated in context and on its own terms and with conclusions merged at a higher level.  I have previously described this approach as ‘multi-disciplinary’.  The aim is thereby to avoid teleology and other problems arising from the merging of evidence from diverse chronological and geographical contexts. 

When adopting the highly contextualised ‘multi-disciplinary’ method used here, one result can be that the written record’s scantiness can lead to interpretations being founded on very small data-sets, sometimes upon individual anecdotes.  On the other hand, when rigorously separated out by time and place, seemingly generic sources begin to manifest differences which might have been hidden by a concentration upon a corpus of material taken as a whole.  These particularities become significant when compared with similar trends manifested in other types of evidence.  The confrontation of different types of data, of differing genres and of quite divergent forms (excavated and textual) further allows us to evaluate the significance of chronological variation and to select some interpretations over others, as better able to explain a wider range of the observable evidence.  This goes some way towards circumventing the problems created by limited quantity and indecisive statements of the written data.  The emphasis on change further entails exploration of the dynamics of change, located in the relationships between élites and their neighbours and rivals at the local level, and those between these aristocrats and the kings on a wider stage.

This paper argues, somewhat against the historiographical trend, in favour of the ‘service aristocracy’ interpretation.  It nevertheless stresses the dynamics for change and the development of an élite class in this region, an aspect of the debate which has not received due attention.  As well as chronological change, geographical diversity must also be noted.  As will be explored in in depth in a separate article, the unusually wealthy Frankish landed aristocracy envisaged by Wickham (and indeed clearly visible by the ninth century at the latest) was a creation of the important decades on either side of the year 600.  Although countered here, Wickham’s rephrasing of the debate has nonetheless led to a refinement of the ‘established aristocracy’ position, to its grounding in a broader range of evidence and to increased subtlety and sophistication.  Similarly, if my own view is ultimately rejected, it is hoped that it will make a similar contribution to refinement and nuance in any future consensus opinion.

In another separate study, I address the subject of the type of aristocracy which faced the Merovingians in northern Gaul, as they established their realm.  There I demonstrate that the Romano-Gallic aristocracy of the region was by no means a securely established, independently wealthy élite.  The principal exception to this rule (to which I shall return) was found around the former imperial capital of Trier, where aristocrats had been unusually wealthy since the early Roman period.  The Frankish aristocracy might have been more deeply rooted within the structures of society but the extent to which this remained true after the settlement of northern Gaul is unclear.  An analysis of the late fifth- and early sixth-century ‘horizon’ of lavish burials (sometimes called the ‘polychrome’ or the ‘Flonheim-Gültlingen’ horizon) supports the claim that the component of the northern Gallic élite that owed its origins to the Frankish migration was associated, or soon had to associate itself, with the Merovingian royal house.  The same was true of its northern Gallo-Roman counterpart.  This study removes most of the argument that, when the kingdom was created, the Merovingian dynasty was confronted by locally well-entrenched leaders.  In this context, selective displays of political violence, such as the eradication of rival royal families (competing foci for political legitimacy), encouraged other Frankish leaders to compete for Merovingian support.  A widespread cull, or mass deportation, of the region’s existing aristocracy need not be posited to understand how the Merovingians established dominance over the leaders of local communities.


Part 2 is here

Translated Sources Handlist

For ease of access, I have created a new 'page' for this (see the bar below the blog title, but this is the url).  Thanks again to those who have sent in additions.  This does seem to have become something of a group project and I think it will thus be a useful resource.  I haven't yet incorporated all the additions but I will, and when I have, due acknowledgement will be made.  Thanks again.

Wednesday 22 February 2012

Changing Minds around 600

[I was very kindly invited to give one of the once-a-term York Medieval Lectures yesterday (in-house speakers are not often invited), which was an honour of course.  I reprint the text as given, minus introductory burble.  It is all very provisional and doubtless all much more complex.  Indeed maybe just plain wrong, but at the moment these seem to me to be interesting lines of enquiry.

The argument goes like this.  Many things were changing around 600, of which I give a sample.  The problem is how to explain this wide range of changes across many geographical areas, social, political, military and cultural aspects, etc.  Pirenne blamed the Arabs of course, but that obviously won't work.  In the past I looked at short-term, important political events, then moved on to see things more in terms of a conflict for resources between kings and aristocrats.  Now I am exploring the idea that a combination of features, contingent political historical events, the knowledge of living 'after Rome', an ideological crisis as a result of the latter, inside and outside the old Empire, and theological worries about the end of the world (dependent upon the long-standing belief that the Roman Empire and the Sixth Age were commensurate) came together to produce a very distinctive world view, one that soon passed but left far-reaching changes in its wake.  Well, see what you think.
]

For the last two and a half years, I’ve been working on a project entitled ‘The Transformations of the Year 600’, funded by the Leverhulme Trust. 
I’ve been exploring the notion of the ‘end of the late antique state’ around 600.   This may or may not be descriptively adequate but it involves discussing the end of a Roman world view, and the need to create new ways of thinking about power, gender-relations and so on and so forth after the end of the Roman Empire.  That was what the ‘changing minds’ element of the title referred to.  But it’s that area – that mental shift – which, over the last two or three months, has really taken over my thinking about the period. It also relates to a lot of other reading I’ve been doing over the past two years about the meaning and importance of history itself.  So, what I present tonight is a long way from being a coherent, fully-worked-out thesis; it is a snapshot, as of February 2012, of where my own mind is at, in thinking about the period between roughly the last third of the sixth century and the first half of the seventh.  In descriptive terms – and in many explanatory ideas too – I have little new to say.  What I have to offer is rather an attempt to bring together various disparate elements – written and excavated data, social and political history, both of those and religious and intellectual history – and to try to confront some important issues, such as how change happens, especially what, with Foucault, we might call epistemic change.

Now, there are four caveats that I need you to bear in mind throughout what I am about to say, as I don’t have the time to reiterate them with every point I make. 
1: The first concerns the diversity of experiences and strands of the story.  If you are trying to give a whistle-stop overview of a period of between, say, eighty and 100 years in less than an hour, you have to generalise.  The period I discuss saw many different histories, whether looking at geographical regions, social groups or historical themes.  They don’t all run at the same speed and with the same chronological shifts.  Some may indeed run counter to what one might see as a main trend.  Classical and medieval thinkers liked to use music as a metaphor for unity in diversity. If you like, this paper represents me whistling something which might be recognisable, overall, as the melody of a particular piece of music but which corresponds in detail to none of the parts played by individual instruments or voices.
2: The second, related, caveat is that I am not discussing a dramatic, revolutionary change, a seismic shift from one thing to another.  I’m not suggesting that on 1 January 601 a bell tolled and everyone changed their ways of doing everything.  There were of course continuities for the ‘old’ ways and precursors, models and forerunners of the new.  What I am taking about is a change of emphasis or of pace.  It’s significant – indeed vitally important – but we need to be clear about how I understand such a change.
3: I don’t for example (and this is my third, again related, caveat), see it all as all happening suddenly, at once or all at the same time.  I might argue that the decades either side of 600 were a ‘Schwerpunkt’ but the period of change spans two or three generations. 
4: Finally, and perhaps most importantly, this is not the period of change.  I stole my project’s title, shamelessly, from that of my namesake, Guy Bois’ controversial book The Transformations of the Year 1000.  Unlike Bois, however, I am not postulating that this X marked the spot where the ancient world became the medieval world.  The period c.600 is often described as the start of the medieval world.  Indeed one short-lived journal of the 1990s, Medieval History, set its opening date, very precisely, as 594.  What happened in 594?  As I’m sure you know, Gregory of Tours died.  He bore a huge weight on his shoulders; he took to his bed, breathed his last, and the classical world ended.   But, the start of the medieval world?  What does that mean?  I don’t know; it’s one reason I don’t think of myself as a medievalist.  If it means that the world c.650 is more recognisable to people who study the twelfth or thirteenth centuries than the world c.550 then I’d have to agree, but then the world c.750 would be even more recognisable and the world c.450 even less so.  It might well be that many of what one might, were one so inclined, think of as defining features of the medieval world were significantly shifted around 600 but the danger with that sort of thinking is that, like Bois’, it assumes two periods of greater or lesser structural stability on either side of a point of transformation or transition.  The world as it was in western Europe in c.550 was a pretty recent creation and the world that emerged after my period of change did not remain unaltered for long.  A lot happened around 700. 
You must bear all four of these general points in mind, throughout what I am going to say, as qualifying an oversimplifying overview.

2: Description of Period

The Column of Phocas, Rome
Nothing says 'the end of the classical world'
like the Column of Phocas
So, you might be asking yourself, what is so interesting about the period around 600?  Well, I’ll tell you…  This will of necessity be a rather breathless and selective round-up.  I will return to the elements that concern me; others can perhaps be dealt with in discussion.  One topic that most interests me is time and how people thought about it, so this listing has something of a focus on that area.  As ever Gaul/Francia is the area about which I know most and it will retain its centrality within my discussion.  I’m not assuming that what happens in Francia must necessarily happen everywhere else; in some cases clearly it doesn’t; in others there are distinct variations on the theme.  It nevertheless gives a reasonable approximation of the general directions in which things were moving across the West.  One thing that has long been said, and it still seems to be true, is that it was this period that saw a decisive shift of focus towards the north of France and the North Sea cultural zone province from the Mediterranean.  This might be of more general cultural significance than I hitherto supposed.
The material that first drew me to the importance of the changes around 600 was the funerary archaeology of north-eastern Gaul.  I have talked about this JustSo – Very often, not least here, that I will – you’ll be pleased to hear, be very brief but it is important and must be mentioned.   The principal issues are these.  First of all, one of the major shifts in the form and design of artefacts took place sometime between c.575 and c.600.  This involved, not least, a radical shift in the balance of investment in material and decoration from feminine to masculine artefacts.  At the same time, the focus of funerary ritual moved from the comparatively transient (if quite multi-dimensional) display of the grave-goods ritual, to a greater degree of investment in more permanent – if more one-dimensional – commemoration in above ground monuments, and perhaps in other, written or liturgical  forms of memoria.  Monuments include the construction of barrows on the fringes of the Frankish realms, churches (in northern Gaul moving gradually out from the cities, via the smaller administrative nodes, to the countryside), grave-stones, sarcophagus-lids visible at surface level, walls around burials, and so on.  Commensurate with this is a significant growth in the number of cemeteries, usually smaller than their sixth-century precursors. 
Now, it is possible to see similar shifts in other regions of the West, both inside and outside the former frontiers of the Roman Empire.  The rash of lavish barrow burials in Anglo-Saxon England, most famously Sutton Hoo, of course, most recently Prittlewell, is well-known.  Not just the investment in the display is noteworthy, but also the process of separation – both of which find their parallels in Francia, particularly on its fringes.  Many of the changes involved in the move towards what used to be called ‘Final Phase Cemeteries’ – or ‘Conversion Period’ cemeteries (I don’t think the description is very much more helpful) – are recognisably similar, though rather different in detail.  The increased concern with above-ground commemoration extends to the reuse of prehistoric barrows and other monuments – something which has some parallels in Francia, especially in the east, though elsewhere it might be that the standing remains of Roman villae were more important.  North of the old Roman frontier in parts of the Pictish kingdom it is the case that above-ground monuments were becoming more popular at this time, too, with the use of symbol stones, cairns and various types of barrow. 
The usual political interpretation of the Anglo-Saxon burials is as a sign of the emergence of kingship.  The Pictish developments are usually placed in a sequence allegedly charting the growth of the Pictish ‘state’.  I will present some reasons to rethink this interpretation later on.  The standard interpretation of the Anglo-Saxon trend towards the reuse of barrows and other ‘ancient monuments’ is that seventh-century people were attempting to associate themselves with the ancestors, in other words with the past.  I find this reading entirely unconvincing, for reasons I have set out at length, especially with regard to John Moreland’s reading of the Wigber Low barrow in Derbyshire.  There seems to me to be no evidence to support it and, in broader context, especially looked at across Europe, it is far more likely that this represents an over-writing of the past and a concern with the future.
Whilst were on the subject of the future, here are two clauses of Frankish law concerning what you have to do if you can’t pay a legal fine.  The first is clause 57, De Chrenecruda, of the early sixth-century Pactus Legis Salicae – a favourite of any student who has to read the text.  The guilty man goes before the mallus – the court – and declares publicly, with twelve oath-helpers, that he has no property left, above or below the ground.  Whereupon he must take handfuls of earth from the four corners of his house and throw them over his shoulder onto those of his closest relatives, on his father’s and mother’s sides, who must help him pay.  He does this to all those who are related to him within three generations.  Something over a hundred years later, the equivalent clause, 12.2, of Lex Ribvaria, probably issued in the 620s or 630s, though this is uncertain, simply reads that if a man cannot pay the fine, his sons can pay over three generations.  What we have here, whilst the concern with three generations remains constant, is a shift from a public, ritual display bringing in living relatives, towards a spreading of the responsibility into the future.
This is not the only concern with the future in Ripuarian Law, which contains, to my knowledge, the first reference to the procedure whereby, if a written document is not drawn up, young boys are brought to a legal transaction and boxed around the ears to make them remember it.  The reference to documents is important.  Ripuarian Law, and in this it is entirely different from Salic Law, repeatedly says that customary law, partible inheritance and similar matters can be modified if a document exists.  And indeed it is from around 600 that documents start to survive in Francia.  Various chance explanations for this have been proposed but none is very satisfactory.  In the tenth century, the oldest charters Flodoard of Rheims could find in the cathedral archive were from the period between 580 and 600, apart from a couple of bishops’ wills, notably that of Saint Remigius, presumably kept as a relic.  The oldest original charters belong to Chlothar II’s reign (584-629), there’s a possible interpolated charter from the reign of Chlothar’s cousin Theudebert II (596-612) and the oldest genuine copies come from the earlier seventh century.  The numbers of authentic charters really take off from about the 640s and into the second half of the century, though.  The most forged Merovingian king is not, as you might expect, Clovis, founder of the dynasty; Theo Kölzer could only find six forged charters of Clovis.  It is Dagobert I (621-639), for whom Kölzer tracked down no fewer than forty.  Forgers, it seems, picked a reign from which it was plausible that a charter might have survived.  Clearly, no one had ever seen a charter of Clovis, evidently making attempts to claim him as a donor almost as suspect to early medieval readers as they are to modern ones. 
Forged Merovingian Charters, by Reign, Childeric I
to Dagobert I

It is interesting too that for the century or more between Childeric I and Childebert II’s accession in 575 Kölzer found seventy-four deperdita – charters referred to by, or inferred from, other sources but now lost.  He catalogued no fewer than 133 for the sixty-four years between 575 and 639; mostly from Dagobert’s reign.  Now, that evident three- or fourfold increase in the relative frequency of documents must be relativized by the growing survival of written evidence from around 600: a vitally important change in itself.  Doing so reminds us that the sixth-century Merovingian kingdom was highly literate, generating all sorts of documents – almost none of which survives.  Indeed, there are almost as few surviving documents from Gaul north of the Loire before 600 as there are for the British Isles.  Apart from warning us against assuming that a documentary blank indicates illiteracy or illiterate government, this shows that the increased document survival after 600 is not an issue of the introduction of writing as much as the decision to retain documents (an issue not even addressed in John Moreland’s book).  Why that might be the case is an issue to which I will return.  It sheds a different light on the production of northern formularies from the seventh century, for one thing.
Associated with the survival of these documents is the development of aristocratic estates.  Northern Gallic Merovingian aristocrats might have had more secure control, more eminent ownership of, lands and a greater ability to dispose of them as they pleased – far more wills survive from the seventh century than the sixth; Salic Law did not even recognise the possibility of testamentary disposition and a story in Gregory of Tours suggests that some kings did what they could to back this up.  This might also have been because sixth-century royal grants were grants of government and of revenue, rather than of ownership of the land.  Higher-status settlement sites become archaeologically visible in the seventh century.  Associated with this is the growth of immunities, preventing royal officers from entering specified estates.  That itself suggests two crucial developments.  It illuminates the well-known decline of taxation around 600 but the prohibition on royal officers entering estates to collect fines for non-performance of military service suggests, with other evidence, an important change in the raising of armies.  In my book on the subject, I argued that sixth-century armies still functioned effectively as royal coercive forces, raised from administrative units, commanded by royally-appointed officers.  Seventh-century forces appear to be ever more focused upon aristocratic retinues.  There are other military changes at this time that I don’t have time to discuss.  The rites to extract local surplus and military service seem by the middle of the seventh century to have devolved to local aristocrats.  This may have been the case in Spain as well as in Francia, although the Spanish case seems to me to be more complex and to require more thought. 
The key shift that unifies most of the preceding points is a growth in aristocratic power.  Thus one can detect changes in the description of aristocrats.  Magnate dynasties begin to be securely traceable in northern Gaul.  One reason for this is their promotion of rural monasticism, in Gaul often associated with the Irish monk Columbanus but which, as Ian Wood has shown, rapidly became Frankish to all intents and purposes.  The foundation of these monasteries was a focus for the donation of lands, a means of securing aristocratic control over lands and so on. 
The growth in aristocratic power is certainly to be associated with the economic revival of the North Sea zone from around 600.  The trading sites or emporia become more visible than hitherto, and northern Gallic towns experience an upturn after two centuries of stagnation.  An increase in craft-specialisation is detectable in the metalwork and ceramic evidence, and smaller-denomination coinage reappears.  This is doubtless related, too, to the well-known shift in trading patterns at about this time, from the so-called Mediterranean system that predominated in the fifth and sixth centuries to the ‘Continental System’ of the seventh.
Finally, there are major cultural shifts too.  The use of the Bible, and especially of the Old Testament, in political writings increases at about this time.  About twenty years ago William Daly wrote an important piece about the difference between Gregory of Tours’ Clovis and the Clovis who appears in contemporary sources, showing that the Clovis of his own day was a much more Roman and Christian figure than Gregory’s.  Daly didn’t make the point, though, that Gregory’s Clovis is very much a Clovis for the late sixth century: an Old Testament king who walks righteously in the sight of the Lord and smites His enemies.  Robert Markus brilliantly discussed the importance of typological thinking to Gregory the Great, and how, when compared with earlier thinkers (most notably Augustine) Gregory is able much more easily to pass through the sign – the word – to the signified beyond.  This sort of development, I think, is more widespread than perhaps Markus thought. 
Jacques Fontaine argued that the seventh century saw an acceleration in the importance of rhythmic verse but especially of a type of poetry that was composed in autonomous strophes making more use of parallel phrasing and synonyms.  This can be set alongside Gregory the Great’s thought and Gregory of Tours’ historical writing.  Another shift that has been remarked upon is in the greater use of miracles in religious writing.  It has been argued that Caesarius of Arles makes almost no use of the miraculous in his Sermons, but that his vita, composed in the 540s contains many miracles and by the time one arrives at the Gregories, especially the Bishop of Tours, the miraculous can be argued, as by Ray Van Dam, to dominate explanation, to the exclusion of theological debate.  Finally the period I am concerned with is, more or less, that of the floruit of Salin’s Style II, which shows a number of key differences from its late fifth- and early sixth-century predecessor, funnily enough known as Style I.
The world's most famous Style II Object
It’s possible to sketch all of these developments too starkly, as I have doubtless done, and there are, as stated, many regional variations, but the selection given at least sets out the background to what I want to talk about in the main section of the lecture.  One thing I would like to stress, to any Anglo-Saxonists in the audience, is just how well the Anglo-Saxon picture fits alongside that which I have just sketched, which I hope will make you reconsider the way in which explanation is still widely hung upon conversion to Christianity – at least ponder whether conversion was a cause or a symptom of change.

3: Explanations

All of which brings us neatly to the issue of explanation.  The most famous explanation is, I suppose, Henri Pirenne’s from the early twentieth century, blaming seventh-century changes on the Arab conquests.  This might still be popular as an explanation in some quarters, for obvious reasons.  But the Arab conquests come too late to explain the western Mediterranean changes and again you have to ponder the extent to which they were themselves cause or symptom of the changes taking place at this time.  Pirenne, famously, thought that the rupture of the Mediterranean from the north-west caused Europe’s centre of gravity to shift towards, well, his native Belgium, really.  That economic dislocation, though, can be traced as early as the fourth century and the Mediterranean economic crisis comes too early to have had much to do with the Arabs – although it might have had crucial knock-on effects in the British Isles and it does have a role in explaining some of the variations in trajectory between northern and southern Europe.
Twenty years ago, I pinned the Gallic changes on the royal minorities that occurred between 575 and 613, decisively altering the balance of power between the royal court and the aristocracy, towards the latter.  I still think that that’s very important in explaining the end of the Late Antique state in Gaul.  But to what extent does, or can, it explain developments elsewhere?  It probably did have effects in southern England but there’s only so far that one can push such an explanation.  It’s true that there were political upheavals in Spain around 600, too, that could have had similar effects, but one must also note that (according to Shetelig in the 1930s!) there were changes at about this time on Norwegian cemeteries that look very like those that took place in Gaul.  Frankish and Gothic political instability surely didn’t have effects that rippled out that far!
One also has to ask why dynastic crises in Spain and Gaul should have had the same sorts of effects (if they did!).  So I moved on to consider the end of the late antique state.  Many western European polities in the sixth century fit a reasonable, uncontroversial definition of a state, whereas they don’t (as I see it) in the seventh.  My explanations for that focussed on a rivalry for control of comparatively finite material resources, both at a local community-level and at a higher political level, between kings and aristocrats.  That nevertheless questions why crisis should automatically lead to a hoarding of power by the aristocrats.  It raises the problem, as Chris Wickham put it in 1984, using evidently Gramscian terminology, of ‘how the state lost consent’.  The materialist explanation only goes so far here.
So we need to enter the world of ideas.  Markus discussed the period 400-600 as one of ‘desecularisation’.  There was an ‘ascetic invasion’, a shift towards more monastic ways of thinking about the world, which left little room for the secular.  Markus was a very fine historian and obviously there is a great deal in his argument.  Yet, it too falls short.  Like all of us, Markus had his blind spots.  One was a curious lack of interest in the contemporaries of Gregory the Great, who, for Markus, simply stands at the end of a line that starts with Augustine and passes through Cassian and Cassiodorus.  Markus says little about the other Gregory – of Tours – or Isidore of Seville, let alone other types of text.  It’s difficult, moreover, to avoid the impression that Markus saw Gregory as illustrative of a something of a sorry decline when held up to Augustine.  Again, as far as I can judge, there’s something in that but I think that we need to interrogate this more clearly.  Markus saw the epistemological shift around 600 in terms of decline.  His reading is rationalist, as from a sort of analytical philosophical viewpoint but while instinctively I agree with his judgement I think there must be a more positive or creative way of reading the shift.
Markus’ idea of an epistemological crisis is one in which the old ideas cannot provide answers to new questions, forcing a radical shift in ways of thought.  But, says Markus, like a true classicist, the old ways could solve all of these problems.  How could classical civilisation not?  The idea that that sort of thinking might be found wanting is clearly unimaginable!  This is what I want to explore in what remains of this paper.
In 1988, after a paper by Jocelyn Hillgarth about eschatology and political thinking around 600, Karl Ferdinand Werner said (in my translation),
“I just want to say that I am wary of whether an ‘atmosphere’, or certain intellectual ideas can be a driving force.  Does a particular thought dominate men and their behaviour?  I’d like to express some scepticism about taking that too far.” 
It is a fair question.  How far was the rarefied air of a Pope Gregory breathed by kings, or the men hauling Sutton Hoo Man’s ship up a hill in Suffolk, let along those carving Pictish symbols in areas that the Roman Empire had never governed?  This is what I want to explore, using a potpourri of ideas from a range of thinkers, used in my own Humpty-Dumpty-ish sort of way.  They will cluster around the reciprocal relationship between the subject – the social or historical actor – and the symbolic order.

4: Attempt to try and bring all this together

The Immediately Post-Justinianic West

In Barbarian Migrations I dealt with the issue of narrative and the tyranny of a particular narrative of fifth century history – a theme I develop in my new book, Worlds of Arthur.  I tried to write a history in the ironic mode, a history, in other words, determined by unintended consequences of actions aimed at doing something else entirely.  My view of history is as something entirely chaotic.  The period from 550 to 650 presents a very different, if related, set of issues.  For one thing there is no overarching narrative of any sort, once the Roman Empire dissolves.  That is one reason why the importance of the decades around 600 has not really been realised, unlike in the East, where an imperial master-narrative still pertains. 

The Point de Capiton

It seems to me to be crucial to understanding this period to acknowledge that, after Justinian, his ideological output and especially after his wars and their failure to reconquer territories that he now proclaimed had been lost to barbarians, it was impossible to go on believing that Roman Empire still existed in the West, just as before.  By the end of the period that concerns me it was impossible to believe that in the East either, and many of the responses were analogous.  This has important implications.  For one thing and this is a weakness in Markus’ thinking, it cut away the key anchors of the old signifying system.  One is born into a world of language, of signification, which is why Lacan and Derrida were correct to reverse Saussure’s formula and insist on the priority of the signifier over the signified.  But the removal of, in Lacanian terms, the points de capiton of the classical signifying chain meant a fluidity of meanings.  If one were to combine that with Derrida’s not dissimilar view of language, then it seems to me that this would precisely render the old ways of thinking unsatisfactory in a new context, in the way that Markus failed to acknowledge. 
Help, however, was at hand.  One of the crucial changes of the fifth century – perhaps the crucial change of that century, more so than the old narrative of barbarian invasion, which makes it ironic that I didn’t really spot it in my book – is a shift in the way of seeing the world.  In the classical Roman vision, the centre, the core, of political legitimacy was defined by the ideals of the civic Roman male.  Closeness to that core, or distance from it – along analogous axes of barbarism, femininity or animalism – defined relative political legitimacy.  That nearness or distance was of course ultimately determined by those who controlled the imperial court.  For various reasons, that centre fractured in the fifth century and the world came to be structured in terms of religious orthodoxy, and distance from that, towards various heresies for example, was what defined illegitimacy.  You can track that shift at all sorts of levels of fifth-century politics and across a wide spectrum of cultural forms.  One advantage was that the centre, here, was less fixed but it is clear that this was still played out within an essentially Roman framework.  Many Christian values at this time mapped directly onto those of the civic Roman.  Compare the preambles of Valentinian III’s Novels with the laws of his great-grandfather and namesake Valentinian I and see how things have changed.  Even into the sixth century in the West, religious orthodoxy was still a matter linked with the Emperor.  In the East of course it continued to be but in the West what the Emperor thought largely became a matter of indifference after the Three Chapters Schism.  That increased concern with Christianity provides both the cause and the cure of the post-Roman anxieties of the period around 600.
The problem about being Christian in a post-Justinianic world was this.  At least for two and a half centuries, since Eusebius had synchronized biblical history with classical, the sixth age of the world had been seen as commensurate with the Roman Empire.  Christ, after all, had been born during the reign of the first Emperor.  Gregory of Tours added that Saint Martin, his hero, had been born during the reign of the first Christian Emperor.  So, what now?  Now that the Empire had so clearly ceased to exist?  This, I propose, is the key to understanding all sorts of aspects of this period.  Many of those alive in the decades around 600 – which they knew was 600 years after Christ – thought they were living after the end of history, after linear time.  Hence the new and apparently qualitatively different apocalypticism that one finds so strongly so strongly marked in Gregory the Great’s writings.  Less widely appreciated is the fact that similar concerns appear in the works of his direct contemporary and namesake, the Bishop of Tours.  Gregory seems to perceive the approaching end, almost malgré lui in Giselle de Nie’s reading.  In de Nie’s view, Gregory began with the idea of reassuring his readers that the end of the world was not nigh but by the time he finished the Histories the evidence of his eyes seemed to tell a different story.  Signs and wonders were on the increase and with them false prophets.  There is a lot in this.  The only problem is that the Preface to Book V of the Histories contains an unmistakable apocalyptic allusion.  Die Nie thus has to argue that V.Praef. was composed late and inserted into an earlier composition.  I’m not convinced by this and have argued that it is actually – at the opposite extreme – the first part of the Histories that Gregory composed.
So, what do you do in a world after time?  In a world devoid of its centuries-old fixed points?  The solution seems to have been to turn to another source of fixed points and underpinnings, the Bible.  Given the synchronism of Christianity and the Empire and the unpicking of the quilting points of the Roman signifying system, then it is not surprising that the Old Testament should have become more of a constant touchstone than before.
This new set of reference points works in a number of directions.  For one thing it allows western kings to find new forms of legitimation.  This is the moment when kings become new Davids and Solomons rather than new Constantines or Trajans.  Religious underpinnings of kingship come to the fore and in a new way.  Earlier, I contrasted the legal pronouncements of Valentinian I and Valentinian III.  If we skip forward another 100 years to the Edict of King Guntramn of Burgundy, issued before the Council of Chalon in 585 we see something qualitatively different again.  This, it seems to me, is the appearance of kingship as divine ministry.  Kingship is a God-given position you don’t find that under the Empire) and the king is responsible for the spiritual well-being of his people.  What I find interesting is that this royal document predates Gregory the Great’s Pastoral Rule by eight years or so.  This suggests to me that Gregory’s thinking, where there’s really nothing to allow a distinction between spiritual and secular pastors, may be rather more typical of his age than has been appreciated.  It suggests to me that the idea of the pastor as answerable for the souls of his flock – for which there were plentiful Old Testament reference points -  was keyed into the apocalypticism of the period.  A judgement was coming. 
Nonetheless, that was useful for royal ideology in what I suspect might have been something of a hegemonic crisis.  In the Empire there had been a what one can label, in Gramscian terms, a civil society in which normative codes served to maintain and legitimise the status quo.  With the unpicking of the classical points de capiton in the period up to c.550, all this will have been questioned.  Why should one do what one’s king said?  A religious – what has been called a monastic or ascetic – replacement must have helped to resolve that crisis.
It might also explain some other religious synchronicity.  In 589, Reccared of the Goths called the Third Council of Toledo, where he renounced the Arian creed that his people had followed for more than two centuries.  Within a decade of that, Æthelberht of Kent allowed Augustine’s mission into his lands and shortly afterwards converted to Christianity.  I have long thought that these events are linked in a current of political and religious thought.  The late and post-imperial trend of adopting a religion that differentiated the military élite from the civil, provincial population was coming to an end.  Perhaps this was because the two groups were by now indistinguishable.  By the time that Ripuarian Law was issued, for example, Romani, rather than being a parallel if legally disadvantaged group of the free population, were semi-free, dependant at law upon free Ripuarians – Franks.  Perhaps, too, in the new world, that way of thinking, locked as it was into an imperial framework, was no longer viable. 
In the intellectual climate of the late sixth century, with the idea of royal ministry gaining ground, perhaps universaility was more important.  This might be equally important inside and outside the former Empire and among former pagans as well as former heretics if – as I suspect was the case, the consciousness of living in a post-Roman world caused ideological crises there too.  This point is of course more readily grasped once one abandons the ‘Invasions’ narrative and sees barbaricum and the Empire not as two antagonistically-opposed worlds but as two inextricably linked parts of the same world. 
The other element of this conjuncture of material, social and mental circumstances was indeed those political crises.  I mentioned at the start the question of why these should have produced what I see as the collapse of the state at this time.  A crisis of the old hegemony would create conditions in which a new order of (in Gramscian terms) civil society could be created but in which also the rules of politics were significantly renegotiated.
A regional aristocracy that acquired a security of local pre-eminence and of the control of surplus could have done what it had done under the Christian Empire, got involved in urban building, sponsored and sent its children to civil schools, had its children educated in the old way, forged a class identity, a control of culture and taste and so on.  But the reference points for that were lacking.  The fixed points according to which the old culture was triangulated had gone.  So new aristocratic cultural forms – like royal – were needed.  Therefore it is at this time that aristocratic (and royal) involvement in monasteries really takes off across the West, from Rome itself as far as Britain.  This expression fits the intellectual currents I have described, not least the concern with ministry and the care of souls, and I suspect a concern with the Day of Judgement.  None of this denies the material advantages bestowed upon the élite by these new forms.  It serves to explain or at least thicken our description of the change in forms.
The coexistence of apocalyptic world view and continuing material politics can be well enough understood via the Lacanian formula much cited by Žižek: Je sais bien mais quand-même: I know very well [that the world is ending] but nevertheless [I’ll go on trying to do the best I can for my family via-à-vis everyone else’s].
We can thicken the description further by considering the nature of the changes in thought.  How does one conceive of a world after time?  The increase in the use of typology is key.  The epistemic change can be nicely illustrated by comparing two large-scale, long-term historical projects: Orosius’ Seven Books of History Against the Pagans, written c.417, and Gregory of Tours’ Ten Books of Histories, written 576-592.  The projects of both are similar to the extent that Gregory quotes Orosius in the key Preface to Book V.  Orosius, though, was more up-beat in his message than Gregory, something significant in itself.  However, although Orosius’ message is very much ‘it was ever thus’, his history unfolds in characteristically classical linear fashion.  Gregory’s Histories, by contrast, concentrate on his own time and are entirely discontinuous.  They work typologically.  The Histories begin not with Creation but with the Cain killing Abel, the type of the event that led Gregory to begin his work: Chilperic’s murder of his brother, Sigebert, in late 575.  After the end of linear time, consequences come about vertically, from God to earth according to the nature of human behaviour.  Gregory writes history as though it were hagiography.  Self-contained anecdotes have their own outcomes manifesting divine presence and they are piled high, typologically, repetitively. 
Pictish symbols: metaphors?
I have already mentioned the prevalence of the miraculous in explanation.  It has been shown, most recently in Charlotte Kingston’s PhD thesis, that Gregory the Great’s discussions of the miraculous manifest the same intellectual traits as his theological writings.  Gregory of Tours seems to reject theological debate in favour of miraculous proofs.  It might be the case, though, that the two Gregories can be brought together, alongside other cultural shifts around an increasingly metaphorical way of thinking.  The Isidorean Style of poetry works, I read, in analogous fashion to Gregory’s history/hagiography; it is episodic and works through synonym, paraphrase with no linear connection between verses.  The model, evidently, is the Psalms. The effect seems to me to be to unsettle and, thereby enable a shift to meditation on meaning, on the issues in question.  In that sense I think that Gregory the Great’s ‘freewheeling’ theology and Gregory of Tours’ preference for the ‘sign’ over the text (so to speak) might be more alike than has been appreciated.  It’s possible, idly, to wonder whether Style II on the one hand and the Pictish symbols on the other might themselves be different manifestations of a metaphoric current of thought.
Je sais bien mais quand-même…  We can return to the attractions of all this in a new political world, a new civic society.  Disjuncture, discontinuity, typology and metaphor might be very attractive to new or redefined élites.  The aristocracies on either side of the Channel, maybe north of Hadrian’s Wall and in other areas, even in Italy, were frequently new groups.  This is where typology and metaphor are so valuable.  They imply the ‘always, already’.  Groups with no lineal descent – even if trying to create such – find their justification typologically. 
So we can end by returning to the above-ground monuments.  If there is a link on occasion with the past, I think we ought to conceive of this in other than simply ancestral terms.  I think it is more important to see it as a discontinuous over-writing but above all the concern is with a monument that will last into the future.  The irony is that that concern with the future, which I see in the burial monuments, the retention of documents and their increased use, is that on the eve of the Apocalypse it was essentially a very short-term view!
Apocalypse soon?  Agilbert's Sarcophagus, Jouarre (c.670)
Does anyone know of an earlier depiction of the Day of
Judgement?
That leads me to my concluding thought.  The conjuncture of political, material and intellectual circumstances around 600 created a very specific world-view.  Within half a century the moment passed.  The world didn’t end.  Of course Apocalypticism persisted, but Bede, born not long after the end of my period, returned to the Augustinian idea that the Day of Judgement was not coming soon.  It is interesting that according to Mark Handley’s studies, after III Toledo, the use of the Spanish Era largely dropped out of use for a few decades.  To me it’s as though they thought they had better stop counting.  By the second quarter of the seventh century they had started again. 
Apocalyptic worries, c.700.  An inscription
from Mellobaudes Memoria, Les Dunes,
Poitiers: ALFA ET INITIUM ET FINIS.
QUONIAM QUIDQUID QUOTIDIAE
PEIUS ET PEIUS QUONDIAM FINIS
APPROPRINQUIT
In Barbarian Migrations I narrated the end of the Roman Empire in ironic terms. 
The most ironic thing of all, however, is that during the preceding century it is almost impossible to identify a single figure who had actually tried to cause its demise.  All the decisive acts in bringing down the Empire were carried out by people attempting to create a better position for themselves within the sorts of imperial structures that had existed in the fourth century.
Similarly, if the mental, cultural and political world c.650 was more recognisably medieval, if many of the tramlines that defined ‘medieval’ thought, culture and politics were set in place around 600, then this was largely because many people thought the world would end at that sort of time.  That would mean that – if you like to think in these terms – the creation of the medieval world, like the end of the Roman Empire came about by accident.  It was a monument to history’s inevitable irony.