[I have been expanding and revising for publication this conference paper from a couple of years ago. Here is the text of the (provisionally) finished piece but without (esp. n.33!) finished footnotes etc. I will revise and repost when those are done but for now I would value any thoughts or responses. There are some quite important difference of argument from the previous version.]
Until quite recently, one of the stranger stories in Gregory of Tours’ Histories had largely gone unremarked upon, in spite of its interesting possible implications, or implied possibilities. When the Nuns’ Revolt at Holy Cross, Poitiers (590) was supressed, says Gregory, the nuns’ leaders, Chlothild and Basina, made a series of accusations against Leubovera, their abbess. One of these was that the abbess kept about the nunnery a man dressed as a woman, so that he could attend to her without arousing any sort of suspicion. Chlothild pointed the man out. And so a man stepped forward, dressed as a woman. He said that he could not do ‘manly work’ and that that was why he dressed that way. In any case, he claimed, he lived a long way off and, though of course he had heard of the abbess, he had never actually met her. This testimony was enough for the bishops (including Gregory) who were sitting in judgement and the charge was dismissed.
One might say that the bishops had been surprisingly easily duped by this implausible tale and assume that Leubovera had indeed been rather less than ideal in her abbacy. The presiding bishops had a vested interest in suppressing the rebellion and so simply squashed the evidence that it had been justified. That possibility should be no means be ruled out, although the story rather loses its potential interest in the process. It is, furthermore, interesting to speculate about why we would want to dismiss the nameless Poitevin’s own account. His testimony, that something as disturbing and seemingly ‘modern’ as cross-dressing might have occurred in early medieval society, is perhaps just too troubling, so we seek reasons to discount it. Thus we silence a voice from the past but put blame the on the tribunal. Nonetheless, as stated, it might just have been that the bishops did simply ignore evidence of Leubovera’s back-sliding, by accepting an invented story.
Alternatively, however, one might side with the bishops and take the Poitevin’s testimony at face value; the rebellious nuns’ were searching desperately to justify their actions and made up a story based around the presence at the tribunal of the man in women’s clothing. The judgement was not entirely uncritical of Abbess Leubovera. It made no condemnation of her lifestyle but there is no reason to suppose it would not have done so, had the evidence been compelling. After all, female power was troubling in the early middle ages and women in positions of authority, judged by men, were always in a difficult position. The rebellious nuns were frequently of the same aristocratic senatorial origin as the presiding bishops and two of their leaders, Chlothild and Basina, were royal princesses; other reasons stated for the nuns’ actions included resentment of the abbess’ strictness. Nonetheless, Venantius Fortunatus’ poems to the abbey’s founder, Saint Radegund, do not suggest that the abbess’ dinner table was as ascetic as one might have expected. Accepting the Poitevin’s account might be somewhat naïve but he might have been telling the truth.
Gregory’s narration provides little to help us decide between these alternatives. He largely relates what the participants said, and passes no comment. There is no analytical resource in his silence. For him the whole episode illustrated a key theme of the Histories: that transient worldly life and status could not be translated into the eternal merit of the truly saintly. The only thing that might help us evaluate the story is the fact that, if Gregory’s account of the tribunal is a broadly accurate depiction of what he witnessed, the Poitevin had come up with this explanation in the first place in the expectation that it would be believed and had come to the assembly dressed as a woman. If Chlothild’s accusation were correct, it certainly seems odd that the abbess’ former attendant should have attended to tribunal in his usual disguise. This, however, is not something that the historian can second guess.
There are two areas where an interpretative decision is required. The first concerns what Chlothild and Basina’s accusation concerned, and the second is what the Poitevin meant by his explanation that he could not perform opus virile. A further problem is the unauthorised assumption, made in most readings, that the Poitevin’s explanation related directly to the nature of the accusation.
Previously, such readings as this passage has received have assumed (possibly on the basis of a misleading translation given by Lewis Thorpe) that Leubovera wanted to have sex with the man, to keep him as a sort of male concubine. The female costume was a disguise. Thorpe, in keeping with this ‘sexual’ reading of the passage, understood the man’s excuse, that he wore women’s clothes because of his incapacity for opus virile as meaning that he was impotent. Gregory’s Latin, however, provides no support for such a reading. He simply reports that Chlothild accused the abbess of having a male personal servant, dressed in women’s clothes. This would be in flagrant breach of the monastic rule in itself, even without any additional charge of sexual incontinence. In this reading the term opus virile would not have any sexual connotation; the term is attested elsewhere as simply meaning ‘manly work’, agricultural labour in the fields.
Neither reading is entirely satisfactory. The latter turns a blind eye to the sexual reasons for the segregation of nuns and thus the fundamental reason why it was deemed wrong for a man to be in attendance on the abbess. It fails to consider why, even if the accusation was simply a breach of the rule’s stipulation that no man be allowed in the monastery, the abbess would want a male servant or why she would conceal his presence. As noted, both readings assume a direct correspondence between the accusation and the man’s justification of his costume, which is by no means necessary. The rebellious nuns might have accused their abbess of sexual misconduct with a man dressed as a woman, and the cross-dressing Poitevin might still have explained his choice of costume with reference to an inability to work outside in the fields. Alternatively, the accusation might simply have been that Leubovera had allowed a male servant to work in the abbey, in disguise, and the excuse might have been that he wore women’s clothes because he was impotent. The double explanation – that he had never met the abbess and that he had another reason for the choice of costume – in many ways only makes sense in the sexual reading: that he dressed as a woman because he was impotent. If he dressed as a woman because he was only capable of work inside the house, as a servant, that would hardly dispel any of the fundamental problems about the abbess having a male servant in the first place. At its core, whatever Gregory’s exact words, this was an accusation about sex.
I will proceed on the basis of the core element of Gregory’s narrative of the event, of which he was a witness: that a man from sixth-century Poitou attended the tribunal of the rebellious nuns dressed as a woman and explained this decision as a result of his incapacity for ‘male work’, whatever that may have meant. This decidedly queer instance allows us to think through some changes that had taken place in gender-construction since the late Roman period and others that were under way when Gregory was writing.
It is quite interesting that, within a house of female religious, the accusation of sexual transgression was entirely in terms of heterosexual activity. The possibility of same-sex desire among the women of the abbey is concealed by silence, which is hardly unusual. What I want to concentrate upon is the relation between clothing and gender-construction. In particular, I want to explore the origins of the social space within which the anonymous Poitevin acted. This, as I have written before, enables us to read some aspects of post-imperial funerary archaeology in more nuanced and less essentialist fashion. I will not draw definite conclusions but leave different possibilities open. Much thinking remains to be done on this subject and a theme of undecidability will run through this essay.
My analysis has two bases, neither of which is fundamentally controversial. The first concerns the nature of classical gender construction; the second involves the gendered distribution of grave-goods within post-imperial cemeteries. To take the first of these, it seems well established that Roman gender was constructed around a central masculine ideal: the notion of civic Roman masculinity. This was historicised; the Romans did not think their men had always possessed such advantages in spite of the biological reasons for it that they proposed. Roman accounts of the remote past saw their ancestors as barbarous. The moment when such barbarism was transcended came with the discovery of law and – entirely related to that – the restriction of the choice of sexual partners. In Roman eyes, therefore, the very movement to civilisation been gendered. What defined the civic Roman male was moderation, reason, the control of passions and the ability to see both sides of an issue. This enabled their employment and reception of the law, it rendered their government something other than tyranny and the performance of its ideals justified a man’s involvement in such legitimate government. This was not simply psychological. As mentioned, in classical thinking it was entirely bodily, biological. The female body’s different constitution, in terms of the humours, made women less capable of such ideal behaviour.
Yet, the non-possession of civic male virtues was not restricted to women. A too-emotional, irrational character – similarly explained by imbalance of the humours – separated barbarians (male or female) from the Roman civic ideal, and the barbarian’s wild ferocity meant that he (or she) could be assimilated with the animal, a third axis along which one might move away from the central masculine ideal. Equally, the child who had not yet been inducted, via paideia, into society could not be expected to demonstrate the true characteristics of the civilised male. The animal, the barbarian, the feminine and the infantile could therefore be condemned by their distance from the civic masculine ideal or praised for their closeness to it, as for example in the instances of the puer senex or the virago (in its initial sense). Barbarian men could acquire the attributes of civilisation so completely that their origins were entirely effaced and, of course, male children could be educated, grow into and acquire these characteristics. A woman, though, no matter how virile, could never entirely occupy that ideal centre, because of her sex.
There are, naturally, different ways of reading this construction. The immediately apparent interpretation, and the one which doubtless structured Roman politics, is that, in sum Roman gender was constructed around the masculine. The Roman woman was secondary, defined as an incomplete version of the civilised man, defined by a fundamental lack. However, one might simultaneously acknowledge that the figure of the woman precedes that of the man, that incompletion and imperfection are logically and necessarily prior to the ‘complete’ figure of the Roman male. Indeed the historical and other cultural aspects of Roman gender construction make it very clear that the Romans themselves acknowledged this. As mentioned, with the passage to civilisation the Roman male had moved out of a situation wherein he was of a type with women, barbarians and animals. What this means is that the Roman concept of sexual difference should not be seen as a totalising binary, with the categories defined on the basis of two opposing, objectively-measurable sets of criteria, adding up to a whole. In other words, the idea was not that ‘woman’ was the opposite of ‘man’ and that the whole human population could be simply divided into these categories on the basis of that opposition. Biology was not enough. Even men had to work to be men, and the ego-ideal of the Roman male was always out of reach. In some ways both of these ideas can be represented in Figure 1. When the Poitevin said that he dressed as woman because he could not perform manly work, this could suggest that the Roman construction of gender continued into the late sixth century: the feminine was seen simply as a falling away from the masculine.
This conclusion might, however, be questioned by the evidence which forms the second basis of my discussion: the post-imperial symbolisation of the masculine and the feminine in burial ritual. From around the end of the first quarter of the sixth century in various parts of what had been the western Roman world – particularly those where a social structure based around the villae had gone into crisis with the Empire’s political disintegration – it became common to inter artefacts with the dead. What matters for the purposes of this discussion is the well-known distinction in grave-goods assemblages between men and women. This, as has now been pretty well established in several different areas of the post-imperial West, was modified by social age and position in the life-cycle. Feminine grave-goods centred upon clothing and the adornment of the body, although some symbols of female work are occasionally also found. Masculine artefacts on the other hand focused on weaponry although, again, in some areas other types of artefact were at least as significant. In the area I know best, northern Gaul, the latter comprised items such as flints, strike-a-lights, awls, tweezers and so on, often originally placed in a pouch attached to the deceased’s belt. This constitutes the second (I hope) uncontroversial basis from which my argument proceeds.
The archaeological evidence suggests two conclusions. One is that masculinity was increasingly coming to be defined by martial characteristics. The other is more controversial or problematic and suggests that there were two opposing poles of attraction in the material construction of sixth-century gender: a masculine and a feminine. In other words, the feminine was not a simple, relative absence of the masculine but was constructed around a separate set of ideals. The patterning in the funerary data suggests that these ideals were structured around sex, by which I mean the female role in reproduction (Figure 2).
Some implications in the slender documentary corpus could support this notion. The Pactus Legis Salicae suggests that the basis of female status was the life-cycle, sex and reproduction. The differing importance of the male life-cycle in governing masculine identity, as hinted at in written sources, might also strengthen the idea that there two gendered sets of ideals. This potentially suggests something different from the reading of the Poitevin’s statement alluded to earlier: that the material construction of ‘woman’ simply represented the lack of the masculine, the failure to be a ‘man’.
How might this possible change from the classical situation have come about? At the core of the situation, I suggest, lay the gradual demise of the model of civic masculinity. A major problem in the factional civil wars that ended the Western Roman Empire in the fifth century was that – especially after Valentinian III’s murder in 455 – no clear winner emerged, establishing a claim to legitimacy that was accepted by all other factions outside Italy. Given that the civic model relied heavily upon links to the emperor, the ultimate guarantor of legitimate authority, this meant that, particularly in the provinces, any claim to traditional civic masculine virtue stood on shifting grounds. Furthermore, a different model of Roman masculinity had emerged during the fourth century, following the separation of the military and civil branches of imperial service under the Tetrarchy. This ‘martial model’ more consciously incorporated elements of the animal and the barbarous, antithetical to the classical ideas of the virtuous civilised man. Constructed (at least in part) in opposition to the civic male, the martial model was less affected by removal from the legitimisation of the central government. By the seventh century, it had become dominant across the West, with only the model of Christian religious masculinity existing as an alternative.
It would, however, be mistaken to think that the civic model’s demise was simply connected to the western Empire’s fragmentation by 476. After all, although by the late 470s contemporaries were doubtless aware that the western Empire was effectively not functioning, they had no reason to suppose – and there is no evidence that any of them did suppose – that the West had definitively ‘fallen’. The ‘fall of the West’ and its consequent need for ‘reconquest’ was something only decreed outright by eastern imperial ideology from the 520s. In the sixty years between the deposition of Romulus and the beginning of Justinian’s wars, the West lay in a sort of ‘twilight zone’, notionally inside the Roman Empire but with no effective imperial government. Most western rulers had, nevertheless, claims to be legitimate imperial officers and the titles to go with them, and it is should not be assumed that none harboured ambitions to take over the currently-vacant western throne. For that reason, a ‘Roman’, civic masculine model continued to be effective, even if losing status to the ‘barbarian’ or ‘barbarised’ martial model. Indeed, in the absence of a western emperor of contested legitimacy, the status of this form of masculinity may have stabilised, linked now, as it was, however nebulously, to the indubitably legitimate Emperor in Constantinople. The Pactus Legis Salicae, illustrates this situation well when it sets out different rates of compensation for the murder of free Franks and Romans of different grades.
This situation ended with the Justinian’s wars and especially with their ultimate failure. The imperial proclamation that the western realms lay outside the Empire cut the ground from beneath any claim to legitimate Roman authority in any area which found itself beyond the newly-redefined imperial frontiers. It was Justinian’s invasions of Africa, Italy and Spain, rather than Odoacer’s deposition of Romulus, that instigated what turned out to be the final crisis of the classical Roman civic model of masculinity. By the later sixth century then, by the time of the tribunal in Poitiers, we can identify masculine and feminine ideals. These, however, related primarily not to each other, in an antagonistic binary, but were instead, I suggest, constructed in opposition to civic masculinity, a central, definitive norm that was, by the late sixth century, disappearing from actual existence. This gives us, in the sixth-century cemetery data, the appearance of two opposed poles. Both gendered ideals were based around concepts that had hitherto been subordinated to the ideals of the Roman male.
All signs are inhabited, or haunted, by their opposites. The Roman civic male was haunted by the irrational, the emotional, the feminine, the infantile, the barbarous and the animal, which is also to say haunted by his imagined pre-civilised past. Nevertheless, for the Romans the civilised man acted as a sort of quilting point for the signifying system. I have argued before that in the fifth century this ‘point-de-capiton’ became unfixed. Yet it had never represented an absolute point of origin because as just mentioned it had always contained within it – even in Roman terms – the image of its pre-civilised Roman precursor. The latter was the supplement without which the ideal could not exist. That historical dimension meant that the space for that central ideal or image was rounded out by what it was not. What is different about the sixth-century situation is that while, of course, male and female were partly defined by what they are not, the ‘supplements’ completing their meaning came not primarily from what we might think of as their structural opposites – man:woman – but from an ideal that was fading from social reality leaving only its spectre: the civic Roman male. This was the ghost that haunted the gendered identities of sixth-century men and women; male and female ego-ideals related more to that spectre than to each other.
Why have I referred to this process as ‘classical gender in deconstruction’? I should say at the outset that I employ the term deconstruction in its correct, Derridian sense. Deconstruction (a term which had long been in desuetude until Derrida revived it as a sort of translation of Heidegger’s term ‘Destruktion’) is a word habitually misused by historians and others, and one about which Derrida himself came to be ambivalent). It lays stress upon the fact that even putatively original meanings emerge from a (logically but not necessarily chronologically) preceding structuring semiotic ‘trace’. This emergence involves a process of deferral (because a final, originary meaning can never be reached) and difference (because all signs derive their meaning from the position within signifying chains of difference), for which Derrida coined the neologism différance. The trace is the space and the moment of différance. One of the crucial points made by Derrida is that these features apply to all forms of semiotic conveyance: written, spoken or otherwise. There is thus no vantage point that exists outside the play of différance, from which to proclaim a true, originary meaning. This was what Derrida meant by his much- (often deliberately-) misunderstood phrase ‘il n’y a pas de hors-texte’ (‘there is no outside-text’). Given that gender construction within society is a matter of language, concepts, signs and signifying behaviours, deconstruction is an appropriate analytical concept.
As has been thoroughly explained by Simon Critchley, deconstruction cannot be reduced to a simple method or analytical process, in which structural binaries are identified and reversed in importance, although this is certainly an element. It is, as Derrida said, something that happens: ‘ce qui arrive’. Instead (although I am over-simplifying a complex argument) Critchley insists that it is a form of double-reading which he terms ‘clôtural’. One reading remains a faithful commentary upon the text; the other an interpretation based on its aporias, points at which a non-empirically-based choice has to be made. An ‘oscillation’ between the two readings takes place. There is no question of establishing a ‘true’ interpretation; indeterminacy is key to deconstruction.
How, then, might that relate to sixth-century gender-construction? As noted earlier, the orthodox construction or ‘reading’ of classical gender saw Roman civic masculinity as the dominant, central, gendered ideal. That had nevertheless been partly defined by its difference from the ‘negative’ feminine and barbaric. In the course of the fifth and sixth centuries, as we have also seen, identities which actively played upon their difference from that once-dominant ideal came to the fore. However, it must also be recalled that, while there were some shifts in the power-relationships between the civic and martial modes of masculinity, both forms rested squarely upon an ultimately Roman, imperial form of legitimation, at least up to the middle of the sixth century. The ‘martial model’ played on concepts that inverted the ideals of civic masculinity but the strength or cachet that it drew from such strategies relied ultimately upon the legitimacy and prestige of those traditional Roman ideals. Further, the ultimate Roman-ness of this barbarising identity found its warranty in a link to the regional kings in whose person were brought together military and civil offices, and both models of masculinity – and the legitimacy of the royal titles or offices was, however theoretically, underpinned by a link to the emperor. At the same time, although the wild, fierce, animal, barbarism of martial masculinity may have been emphasised to suggest (in a late imperial context) a weakness in civic manhood, it could never be forgotten that those characteristics were regarded by anyone with an education as lesser, uncivilised – even feminising – traits. The orthodox ‘reading’ of the material and other symbols therefore remained in place even as an alternative ‘reading’ that flagged up its blind spots was taking place.
Thus the situation that had emerged by the middle quarters of the sixth century should not and cannot be seen as straightforwardly inverting or reversing the relationship of ‘Roman’ and ‘barbarian’, of ‘civic’ and ‘martial’, or even as a process under way towards such an inversion. The socio-political identities based on the two masculine models existed in a dynamic relationship of tension, or a semiotic oscillation, in which the strengths of each were simultaneously its weak points, in which each reading of the signs inevitably led back to the other. Put another way, the power relationship was, as yet, indeterminate.
That indeterminacy, I would suggest, freed female identity from its former position, defined purely with reference to the civic male. After all, some of the features now held (in one reading of late Roman masculinity) to valorise the martial model were, in classical thought, shared with the feminine. As the ‘natural’ claims of the civic male to domination became more contested, and as chastity and sexual renunciation became more associated with the behaviour of dedicated religious, outside the realms of family life, it may have been possible for the female role in the household and reproduction to be stressed. It is also conceivable that, as I have argued before, the political changes of the fifth century meant that the role of marriage alliances in local and regional politics created a space within which female sexuality and status connected with that could be emphasised. Notably, this emphasis was placed upon sexual reproduction, the body and ultimately the biological differences between man and woman; those aspects which, in the view of Lacanians like Copjec, could never be incorporated in a purely linguistic or symbolic discourse of gender. These issues may have – reciprocally – contributed to the stress upon leaving the household and the sphere of sex and marriage in female saints’ lives. As markers of status associated with sex, marriage and the family became very important in secular feminine status their rejection became more obvious and more significant as an act of piety.
All of these developments, I suggest, opened up the signifying space within which the Poitevin mentioned by Gregory of Tours was acting. The classical world, of course, had concepts of mixed gender or sexuality: the hermaphrodite and the figure of Dionysius come to mind. Yet, those classical notions differ somewhat from what is signified by the man at the Nuns’ tribunal, not least in their removal from the field of the mortal. Here, again, we must take Gregory’s account at face value. As Nancy Partner said, it is worth noting that the account describes the Poitevin as a man (vir) in women’s clothes (vestimenta muliebria; in veste ... muliebri), not as a woman, or a woman who turned out under closer inspection to be a man. The implication of the story might be that the man was not considered female by his contemporaries, although the modern refusal by many to accord appropriately gendered pronouns to transgendered individuals is probably relevant. The extent to which the Poitevin lived life as a woman cannot be guessed at.
It is nevertheless interesting that the Poitevin’s construction of identity revolved not simply around the absence, or lack, of masculine items but the active employment of feminine ones. It was not enough for Gregory’s Poitevin simply to be considered less of a man, and more feminine as a result, by being incapable of opus virile; he dressed as a woman to show this. He (or she) claimed that his (or her) decision to dress in women’s clothing was motivated by the fact that, as he put it, he (or she) could do nothing of manly work. As mentioned earlier, this might be seen as in line with classical ideas of gender. And yet, this was not – or so it seems – some sort of punishment; as far as we can tell, the Poitevin had chosen this course of action him/herself.
This is where the archaeological evidence can provide a different view. It is interesting to wonder what exactly this costume – these vestimenta muliebria – comprised. Grave 32 at the cemetery of Ennery (Moselle) contained a skeleton sexed as male but been buried wearing a necklace and accompanied by an item of pottery normally (on that site, at that time) only found in female graves. No traces or indications of the type, shape or cut of the clothing survived, but the absence of a buckle suggests that there was no belt. This assemblage would, on that cemetery at that date, have been appropriate for a woman of an age above forty; it did not stress the sexuality of the deceased in the way that was common with younger women still involved in the processes of reproduction, especially teenagers. What is at stake in this relationship between gendered costume and sexual reproduction? Masculine objects in Frankish burials probably refer – at least obliquely – to the ability to start a family and govern a household. The distribution of goods does not seem to relate directly to sexual potency, at least in purely ‘biological’ terms – Frankish adolescents are rarely buried with masculine items, in spite of being of an age-group whose sexual proclivities could worry Christian moralists and which was linked in the law-codes to the kidnapping of young women. One penitential text suggests, however, that before the age of about twenty (the age, interestingly, when Frankish males start to be buried with weapons) male sexual practice, same-sex or otherwise, was more a matter of experimentation, of ‘games’.
This is underlined in the archaeological example, if we assume that the physical anthropological analysis was correct (and there is no decisive reason not to do so). The deceased’s family displayed this feature of identity in the grave, in public ritual. One might wonder whether this suggests something more positive about the reception of identity. Sixth-century furnished burials were carried out in public, in view of the community; indeed this point is essential for understanding the deposition of grave-goods. Although the precise motivations of the people responsible for determining the nature of the burial (and indeed who those people were) lie beyond our purview, it seems reasonable to conclude that the deceased’s family appear to have acknowledged the dead person’s choice of life style. However, as elsewhere at this date, the community and its norms seem to have played a significant role in the sixth-century northern Gallic burial rite. There might therefore have been an expectation that the deceased person was buried in feminine costume, regardless of their biological sex, whether or not the family wished to do this. Nonetheless, the dead person was allowed burial in the communal grave-yard. Although the grave was oriented north-west:south-east, this orientation was shared with the neighbouring interments, one of which (grave 11) was an unusually well-furnished burial of an adolescent. Its location seems peripheral but we do not know whether, or where, the edges of the cemetery were reached. A NW-SE orientation was used to display distinction at Ennery but not necessarily in a negative way. Such graves included the clearly prestigious and equally peripheral burials 70 and 71 (which broke other community rules to emphasise masculinity). However one looks at this burial, it suggests that views of a biological male who lived life as a woman were not simply negative.
This social identity was not, therefore, seen simply in terms of a distance from the masculine ideal, but also – perhaps – as something more positive: a movement towards valued feminine ideals. If so, and this can only be a suggestion, this underlines the increased ‘bipolarity’ of sixth-century secular concepts of gender. In this reconstruction, the ‘oscillation’ discussed above, between civic and martial models of masculinity, the tension within which masculine ideals remained contested and their relationship undetermined, permitted the expression of antitheses without necessarily implying a negative judgement. If my reconstruction is correct, it underlines that, as in many other areas, this was a quite unusual period of western European history.
How one reads opposite instances, where biologically female skeletons are associated with masculine grave-goods, like weaponry, is something I cannot address. I suspect, however, that this is a rather different circumstance, perhaps more akin to the classical situation and thus, I suspect, not to be read crudely as ‘transgression’ – perhaps quite the opposite.
The subsequent history of gender-construction in Francia perhaps supports the tentative reading I have offered. Indeed the tribunal at Poitiers took place at about the time that the unusual situation I have suggested was coming to an end. The Justinianic Wars redrew a frontier between the Roman Empire and a new barbaricum in the former Western provinces. As stated earlier, within the western kingdoms any link with the emperor or imperial office lost its value in legitimising status and authority. New forms of legitimation, and new foundations for ideology, were required. In this context, what had been an ‘oscillation’ between the orthodox reading of gender, with a dominant civic masculinity, and an alternative view, valorising the ‘supplements’ that rounded out the orthodox interpretation, slowed and changed into a more straightforward reversal and inversion of the earlier imperial hierarchy.
Seventh-century cemeteries reveal the change. One difference from sixth-century sites is the reduction of feminine gendered grave-goods and a greater stress on the masculine. Although the numbers and variety of the weapons deposited with the dead are reduced, the percentage of masculine burials with weapons increases. This seems to represent a final triumph of the martial model of masculinity, which had by now entirely supplanted the civic. This situation may be more akin to the Roman, in that it saw a central masculine ideal, with the feminine judged by proximity to or distance from this. What was different from the situation in c.300 was the nature of that central masculine ideal, which was now based upon the martial model and a different, more warlike, set of virtues. How male sexual impotence was judged, and whether it was marked at all, within that system is not something the evidence of which I am aware allows us to discuss. I suspect, however, that in a society more defined by masculine heads of lineages, that it was rather different: not symbolised in clothing and certainly not valorised by a family. The case of Gregory’s Poitevin suggests a glimpse of a historical moment when, however briefly, something different was possible. The importance of that is that it cautions us not only against seeing modern categories as ‘natural’ but against positing an essential, unchanging ‘medieval’ set of categories with which to compare them.
 Gregory of Tours, Libri Historiarum 10.15. MGH SRM 1.1, ed. B. Krusch and W. Levison (Hanover, 1951), pp.501-6.
 Dinshaw; Sedgwick.
 This was pointed out to me by Dr Erin Dailey, whose very good University of Leeds PhD thesis on Gregory’s women will I hope be published at some point.
 Charters; Wickham, Framing the EMAs.
 Halsall, Sex, gender and transgression
 Biology: Pliny; Vitruvius; Vegetius
 General bibliographic reference.
 Refs to barbarians.
 If I understand her correctly, Joan Copjec argues that (for Lacanians) sexual difference resists complete representation in language because it exists, effectively, within the sphere of (in Lacanian terms) the Real. That would indeed give sexual difference an important qualitative distinction from differences between identities based on class or ethnicity. This could be assimilated with the Roman world-view. J. Copjec, Read my Desire: Lacan Against the Historicists (Boston MA, 1994; repr. London, 2015). See also S. Žižek, Tarrying with the Negative: Kant, Hegel and the critique of Ideology (Durham NC, 1993).
 Refs to general description.
 Sett & Soc
 Halsall, ‘Female status’
 Halsall, ‘Merovingian masculinities’; ‘Growing up in Merovingian Gaul’.
 Refs. Halsall, Barbarian Migrations.
 Halsall, Barbarian Migrations, pp.; Halsall, ‘Gender and the End of Empire.’
 An important role in bringing about these changes was doubtless played by Christian debates on sex and whether the chastely married might be valorised over the virgin. Similarly, the role of asceticism and of competitive display of extreme renunciation would move behavioural ideals in some areas away from traditional civic norms of moderation. Regrettably I am not able to pursue these issues here.
 Titles of Burgundian, Gothic and Frankish kings.
 Theoderic and Clovis styled as augustus. The notion that ‘barbarians’ were prevented by their birth from becoming emperor is a modern fiction.
 PLS refs. Halsall, ‘Transformations of Romanness’,
 Justinian treaty with Theudebert, re Provence.
 The idea of the quilting point or point-de-capiton comes from the psychoanalytic philosophy of Jacques Lacan. Here I am assimilating it with Derridian thinking by interpreting it more pragmatically, within the Symbolic order. I do not see it as representing a metaphysical, absolute point of origin, present unto itself. Lacanians might regard this move as illegitimate (cp. Copjec, Read my desire).
 Derridian bibliography.
 On which see J. Derrida, Heidegger: La question de l’Être et l’Histoire. Cours de l’ENS-Ulm 1964-1965, ed. T. Dutoit (Paris, 2013), pp.23-44.
 E.g. A. Rio, . Deconstruction does not simply mean a close analysis of the various parts of a text or topic, or the breaking down of the latter into its component parts.
 Above all, S. Critchley, Ethics of Deconstruction: Derrida and Levinas (3rd edition; Edinburgh, 2014).
 Again, I must draw attention to, without being able to explore, the effect of Christian ideals in this process. Some areas of the civic masculine ideal – most notably its stress on moderation – were appropriated by the Church. This may have led to the association of such ideals with the celibate, outside family life. At the same time, the practice of extreme asceticism by holy men and women served, in some views, to weaken the claims of ‘mere’ restraint and chastity.
 Copjec, above.
 Partner, ‘No sex,no gender.’
 Halsall, ‘Sex, gender and transgression’
 Ref. An interesting foreshadow of Freud’s idea that such ‘perversions’ were things one left behind after youth. Elizabeth Freeman refers to this as a teleological view of heteronormativity.
 A chance discovery in the 1980s suggests that a seventh-century phase of the site remained unexcavated. Ref. Simmer.
 Halsall, Settlement and Social Organisation, pp.70-71.
 I discuss this issue at greater length in
 Similar in ASE.
 Refs to later Frankish gender. Garver etc.