Rethinking Warfare and Politics in Britain 400-600

Introduction

The traditional view of warfare in Britain in the two centuries following the end of imperial Roman government there envisages things very much in aquatic terms.  A ‘tide’ of migrating Anglo-Saxons ‘washes’ or ‘flows’ over the former provinces.  The newcomers come in ‘waves’ and ‘swamp’ Romano-British culture, ‘flooding’ it with new influences.  The spread of Anglo-Saxon population, culture and influence is seen to spread across the land, from east to west with a discernible ‘front line’: Romano-Britons to the west of it and Anglo-Saxons to the east.[1]  We can plot the movement of the newcomers from their cemeteries (inhumations and cremations with grave-goods) and some sorts of settlement (with certain types of timber hall and sunken buildings or Grubenhäuser). Supposed exceptions to this picture are described as ‘reservations’ – metaphors from the frontiers of European colonization of America and elsewhere are also common in discussions of the period.  The sides involved in the fighting are seen as two quite separate cultures with very different ways of waging war.  In the east, the Angles, Saxons and Jutes wage war on foot.  Some may ride to battle but dismount to fight.  Their weapons are principally spears and shields, though their leaders carry swords and all bear the seax (knife) that supposedly gave the Saxons their name.  To the west the Britons try to maintain a more Roman way of doing things.  They employ mounted as well as dismounted warriors, and their leaders are thought to ride in this cavalry force.  Their campaigns are based upon refortified Iron-Age hillforts, such as have been excavated at places like South Cadbury (Somerset).

This account represents an extreme version of a particular view, to which few scholars today would subscribe in its entirety.  But it is certainly no caricature.  These images and metaphors are so common that people who write about the topic fall into them seemingly unconsciously.  The assumptions behind them permeate discussions of the subject, even when an author is trying to call older ideas into question.  In recent decades many of these ideas have been disputed but without a clear consensus emerging among scholars.  One reason why a more moderate version of the image sketched in my first paragraph probably remains the majority view of the period, and why attempts to revise the picture have made little headway – in spite of the many problems inherent in the older view – is that a series of unexamined assumptions that underlies almost all investigations into post-imperial Britain.  In this essay I want to uncover those and expose them to some scrutiny.  I will do this by placing Britain more firmly in a broader European context. Much of the misunderstanding that bedevils current approaches stems from a failure to do this.

Evidence

Inevitably we need to begin with an account of the sources for this enquiry, and their many problems.  How do we know what we know about Britain between c.400 and c.600.  Indeed, do we know what we often think we know?  Discussions of the evidence for post-imperial Britain can make for rather depressing reading.  To be brief, there are almost no written sources that can be relied upon to aid our enquiry.  The events surrounding the rebellion of Constantine ‘III’ and his ill-fated attempt to usurp the western imperial throne with the support of the British army (407-11) are well enough documented but after that an almost impenetrable darkness descends upon Britain that lasts until the arrival of the missionary saint Augustine in 597.  The Life of Saint Germanus of Auxerre by Constantius of Lyon tells us how the bishop was sent to deal with allegations of Pelagian heresy in Britain in 428-9 and, while there, helped a British army to defeat an army of barbarians.  This source has its problems but gives us some insights into what British society was like in the 420s. 

Other than that, the only written work definitely composed in Britain between 400 and 600 was the De Excidio et Conquestu Britanniae (On the Destruction and Conquest of Britain) by Gildas.  This is a sermon, mostly concerned with the (in Gildas’ view) shocking state of the British church and (to a lesser extent) the misdeeds of its rulers but near its beginning it sets out a brief account of Britain’s history between the Roman invasion and Gildas’ own day.  This formed the basis for every subsequent attempt to write a history of Britain in our period.  It is here that we hear of the Britons’ unsuccessful appeal for help to Aëtius the magister militum in Gaul, about the general Ambrosius Aurelianus and his wars against the Saxons, and about the battle of Mount Badon. Unfortunately, the account is highly rhetorical, polemical and stylised, and impossible to place in any sort of neat sequence.  If we had any other written account of British history in this period, we would probably view it very differently.  There is no good reason to doubt that the appeal to Aëtius (which must belong in the late 440s or early 450s), Ambrosius’ wars or the battle of Mount Badon happened, but more than that is impossible to say.  We have no certain idea who Gildas was or where or when he was writing (which could have been at any time between c.475 and c.550 – perhaps later).

The other sources available to us include Bede’s Ecclesiastical History (731), the History of the Britons sometimes attributed to Nennius (829), the Anglo-Saxon Chronicle (880s onwards) and the Welsh Annals (c.975).  Many of the features of the usual histories of the period, not least ‘Arthur’ (never called a king) and his battles (first definitely found in the History of the Britons), Vortigern and Hengist and Horsa, most of the battles recorded between Britons and Saxons and so on, are found in these sources.  Unfortunately, they are, as will have been noticed, all very much later than our period.  It is very clear that the earliest of these writers, Bede, like us, knew pretty much nothing about the period between Germanus of Auxerre and Augustine of Canterbury.  Not only are the sources late; they all also have very clear political agendas relating to the time in which they were written.  This clearly influenced the sort of history they were telling.  Furthermore, although there are traces of what looks like material from ‘oral tradition’ in some of the sources, we have no idea how early or reliable it is, or how the later authors have adapted it.  There are, furthermore, no indications that any of these sources were based upon lost earlier works.

The only other written sources that might help us are some early Welsh poems allegedly by the bards Taleisin and Aneirin (in whose epic Y Gododdin is also to be found an early reference to Arthur).  It is from these that many of our ideas about post-imperial British warfare have been derived.  Unfortunately, it is impossible to be certain of the date of these poems. They purport to belong to the period around 600 but the manuscripts are very much later.  It may be that some elements belong to the early seventh century but we cannot be sure.

It is really impossible, on the basis of these sources, to construct a narrative political history of Britain between 400 and 600. This sounds like a counsel of despair but abandoning the written sources does not leave us without hope.  In fact, it allows us to break free of the obsessions of eighth- to tenth-century Anglo-Saxon and Welsh authors and rethink the problems of British history in this period.  We can use the plentiful archaeological material that survives on its own terms, rather than attempting to make it fit the story drawn from later, unreliable written works.  We can also consider our evidence in the light of what we know from the much better written sources from the mainland of Europe in the fifth and sixth centuries.

The Anglo-Saxon Migration

One of the main areas of debate in recent decades has been about the nature and scale of the Anglo-Saxon (English) migration.  The traditional view is of a massive ‘folk migration’ bringing with it all sorts of cultural influences.  Thus the archaeological traces belonging to the fifth and sixth centuries in the lowlands of Britain have traditionally been thought of as ‘Anglo-Saxon’, and read as simple cultural markers of the presence of immigrants from northern Germany.  Such elements include the settlements and cemeteries alluded to earlier.  Metalwork and other elements of culture that can be traced back to the Anglo-Saxon homelands in the north-west of Germany begin to appear in south-eastern England from c.430, a date not far out of line with the date suggested for the arrival of the Saxons (adventus saxonum) in the later written sources.  It was suggested that the advance of the Anglo-Saxons could be plotted from the spread of these cemeteries and settlement-types, to the extent that the supposed forty years of peace that followed the Battle of Mount Badon according to one reading of Gildas (Gildas does not actually say this!) could, allegedly, be seen in the archaeological record.

In the 1980s and 1990s a reaction to this view set in, which at times came close to claiming that the whole idea of the Anglo-Saxon migration was something made up by Bede!  Whether or not ‘Anglo-Saxon archaeology’ truly marked the presence of incoming Anglo-Saxons or was really derived from the north-west of Germany was debated.  The idea was proposed that, rather than a mass ‘folk migration’, the Anglo-Saxon movement to Britain was in fact a political take-over by a smaller, élite group of newcomers.  The spread of ‘Anglo-Saxon’ material culture represented not the movement of Anglo-Saxon people but the spread of Anglo-Saxon lordship.  These minimalist views have not convinced everyone.

A ‘middle way’ between these viewpoints is possible.  It is clear that there was a migration of people from the North Sea coastal areas of Germany into the lowlands of Britain.  Some cultural artefacts – not least the English language – are impossible to account for satisfactorily without acknowledging that fact.  But the English migration was not like most other ‘barbarian migrations’.  It was a product of a North Sea culture that predated the migrations, with movement in all directions across and around that Sea, exchanging all kinds of cultural influences.  In the fourth century these influences were overwhelmingly Roman, spreading north from the imperial provinces into barbaricum.  Links between north-western Germania Magna and Britain (and northern Gaul) already existed, therefore, and Saxons were long accustomed to moving into the Roman Empire, sometimes as raiders and sometimes as recruits for the army.  Saxon migration into Britain could have begun long before Constantine ‘III’ launched his rebellion.  It is worth remembering that most barbarian immigrants into the Empire left no archaeological trace.  When the western Roman Empire underwent a period of crisis around 400, its effects are as archaeologically visible in the Saxon homelands as they are in the northern Gallic and British provinces.  Stress in the Saxon homelands and new opportunities in the Empire’s north-western provinces all fed into the matrix of factors governing Saxon migration.  The appearance of ‘Saxon’ archaeology in the 430s probably marks the point at which such newcomers felt able or confident enough to proclaim their cultural identity (perhaps as the ‘Saxon rebellion’ alluded to by Gildas) rather than the date at which settlers arrived.

It is important to note, however, that movement across the North Sea was a long-term phenomenon.  Unlike, say, the Gothic migration into Aquitaine, where the Goths all arrived at once, Anglo-Saxon migration took place across many decades.  The numbers of newcomers or ‘first-generation’ migrants in Britain at any one time could have been very small but the continuing arrival of other Saxons would act constantly to ‘top up’ their culture.

The moving front

One of the migrations most similar to that of the Anglo-Saxons was probably the drift of the Franks across the Rhine into the far northern provinces of the Empire’s Gallic prefecture.  When we come to think about how the lowlands of Britain became English kingdoms the Frankish comparison might help us.  The Frankish takeover of northern Gaul has often been envisaged in similar terms to those used in the study of how the English took over lowland Britain: as a front line moving steadily southwards and westwards from the Rhine frontier.  The documentary sources for this are almost (but not quite) as inadequate as those for Britain and the archaeological evidence used to plot Frankish settlement is remarkably similar (in spite of the fact that the archaeologies of the pre-migratory Frankish and Saxon homelands are quite different): furnished inhumation cemeteries (burials with grave-goods) and settlements composed of post-built halls and Grubenhäuser).  The front line of Frankish settlement at a given date has repeatedly been plotted from the southern extent of the furnished inhumation cemeteries.

However, because our written evidence for fifth- and sixth-century Gallic history is better than that for Britain we know that this image is at best only a part of the whole story, and perhaps not the most significant component.  Two key points that emerge from a study of the fifth-century Franks.  The first is that the Frankish advance from across the Rhine was not the most decisive factor in the creation of the Frankish kingdom.  The second is that straightforward ‘binary’ warfare between Franks on one side and Romans or Gallo-Romans on the other hardly featured in the process at all.  To take these points in order, it is clear that the decisive military force in the period was the ‘Roman’ army based on the Loire around Orléans.  This army was often referred to as ‘the Franks’ because it was composed heavily of Frankish recruits and their officers but for quite a long time it was commanded by Aegidius, a Roman general.  It is even alleged that, when he was in rebellion against the imperial government in Ravenna, Aegidius used the title of rex francorum – King of the Franks.  Command over this force was evidently competed for by a Frank called Childeric, also later styled their king, and his son Clovis, and Aegidius and his son, Syagrius (later styled ‘King of the Romans’ by Gregory of Tours).  Clovis won out and was able to use this army and its control of the southern part of the Paris basin to subdue rival Frankish kings or leaders in the north and, eventually, defeat the Visigoths south of the Loire and cow the Burgundians of the Rhône valley and the Alamans, Thuringians and Saxons.  By the time of his probably untimely death in 511, Clovis was one of the two major powers in western Europe (the other being Theoderic the Ostrogoth) and, like Theoderic, evidently allowed himself to be called augustus.  Studying this course of events makes it very clear that warfare and politics in the fifth-century west was anything but the simple two-sided struggle between barbarian invaders and Roman defenders that the ‘moving front’ model envisages.  Sometimes the Franks were led by Romans; the Aquitanian senatorial aristocracy was in cahoots with the Visigoths throughout the period and an Auvergnat senatorial contingent fought for them against the Franks; and the Burgundians were closely associated with the senators of the Rhône valley.  Indeed, this picture, of warfare for control between regionally-based Romano-barbarian factions rather than between Romans and barbarians, applies throughout the western Empire in the fifth century – surprisingly even in the case of Vandal Africa. 

So, where does this leave us?  First of all, it tells us that if British politics between 400 and 600 did take the form of a ‘moving front’ war between Britons and Anglo-Saxons then it was unique: the only part of the western Empire where this sort of binary opposition occurred.  But our basis for that unique picture is entirely unreliable.  The late histories mentioned earlier all had reasons to portray the origins of their kingdoms or people in a simple story of invasion and conquest by one people at the expense of another.  Frankish origin stories saw their history in these terms too by the time our earliest insular history (Bede’s) was written and we have seen how wide of the mark this idea was.

Another issue of vital importance that emerges more clearly from the Gallic/Frankish evidence is that the rite of burial with grave-goods is not an imported barbarian custom, but begins inside the Roman Empire.  It is found in regions that underwent social competition, especially where a villa-based social system collapsed as in both northern Gaul and lowland Britain.  These graves, therefore, cannot be identified as the burials of Anglo-Saxons (or Franks) rather than of Romano-Britons (or Gallo-Romans).  This has the happy result of enabling us to see the weaponry in these graves not as exclusive to ‘Saxons’ but as very likely wielded by ‘Britons’ too.  The other feature that has emerged from recent scholarly work is just how fluid identities like ‘Saxon’, ‘Roman’, Frank’ or ‘Briton’ could be.  Evidence from the better-documented European mainland gives us numerous examples of people adopting new identities.  Families could shift their dominant ethnic identity over a generation or two.  We know, because our evidence is so much better, that very ‘Germanic’-looking people in seventh-century northern Gaul were actually the sons and grandsons of Roman aristocrats.

This also means that we need not see the known hillforts of the period as exclusively ‘British’.  Indeed, it is worth noting that the two pieces of diagnostically fifth-century metalwork found on the Cadbury excavations were of types usually (probably misleadingly) supposed to be Anglo-Saxon.

Implications for warfare

Clearly this discussion has huge implications for how we see warfare in fifth- and sixth-century Britain and permits an entirely different vision.  Frustratingly, it can never be filled in with precise details of people, places and events or be more than a hypothesis, but it does allow us to postulate a political and military history that is grounded in reliable, contemporary evidence and which better fits what we know of the fifth century more generally.

First of all, the political struggles for power in the lowlands of Britain were probably waged by different factions wherein Romano-British aristocrats were allied with particular groups of ‘barbarian’ soldiers.  The command of these factions was probably contested too, between different families and perhaps between rival groups of British or Saxon origin.  Although the group was probably identified by the ethnicity of the majority of (or the dominant group in) the army, we ought not to assume that all of the warriors were of that ethnic identity or origin.

Second, it may be that the key areas in this struggle, where English political dominance was first established, were on or near the coast.  The western band of the villa area, on the border with the highland zone (running from the Dorset coast to the East Riding of Yorkshire) was the most prosperous zone of late Roman Britain.  Note that the furnished burials suggestive of societal stress are rare in this region perhaps suggesting a greater degree of political stability.  This band was also the area where the most powerful Anglo-Saxon kingdoms emerged (Wessex, Mercia and Deira) and this is unlikely to be coincidence.  Whoever controlled these areas could lord it over regions nearer the coast, including areas settled by incoming Anglo-Saxons.  This would resemble the pattern we saw in northern Gaul, and indeed the domination of the south-eastern coastal regions (East Anglia, Essex, Kent and Sussex) by the powerful ‘inland’ realms was the norm through most of the documented periods of Anglo-Saxon history.

Another feature that emerges from the study of the European mainland is just how fragile many kingdoms were.  Many evaporated after a single serious military defeat, leaving little trace.  The kingdoms of the Thuringians, the Vandals and the Burgundians were all extinguished in the 530s and would be entirely unknown to us if the documentary record was as exiguous as it is in Britain.  Syagrius, ‘king of the Romans’, who competed with Clovis for the control of the southern Paris basin and the Loire army, is known to history thanks only to his appearance in a lost saint’s life used by Gregory of Tours.  In this light it is clear that there is every scope for great war-leaders to have risen in Britannia, and for their kingdoms to have subsequently vanished, without leaving any trace.  Ambrosius Aurelianus may have been a mighty figure but, were it not for one comment in Gildas, he would be lost to history.  In this context there might equally well have been a successful war-leader or king called Artorius or Arthur.  If he fought against both Romano-British and Saxon rivals, and if his family or faction lost control of their realm or army to another group that was, or claimed, an Anglo-Saxon origin, he would have served no one’s historical purposes by the eighth century, when our narrative sources begin.  This would explain why he was barely remembered in any insular sources before the Norman conquest and why his story was evidently so little known.

What we have seen is that the factions involved in fighting for the control of what had been Britannia in the fifth century were probably a fluid mix of ‘Saxon’ and ‘Romano-British’.  It is worth remembering that the fourth-century Saxon homelands were very exposed to Roman influences and that many Saxons evidently served in the Roman army – as the official Roman belt-buckles and brooches that were cremated with them when they died testify.  Conversely, the Britons, like other provincials, were fully familiar with Saxon soldiers, and many probably fought alongside them in some of the late Roman army’s regiments.  This makes it extremely unlikely that there were drastic differences between the fighting styles of Romano-British and Saxons.  The idea that the Saxons always fought on foot – based in any case on a handful of late Anglo-Saxon references taken mostly out of context – is improbable.  The equipment from the famous bog-deposits in Denmark and northern Germany shows that warriors in that region were entirely familiar with mounted warfare.  It is most likely that, as across the early medieval West, élite warriors fought on horseback or on foot according to the needs of the situation.

The archaeological evidence from the lowlands, especially, suggests that warriors made much use of throwing weapons – in the late Roman tradition.  Throwing axes of the francisca type, heavy iron angones descended from the Roman pilum, and other throwing spears are known.  Analysis of the shields of the period suggests that they were smaller than later (although it may also be that they were taller and oval, rather than round).  There is some evidence of archery.  Sling-stones are known from the western hillforts but may equally belong to their earlier Iron Age occupation.  Armour and helmets are more or less unknown from the fifth and sixth centuries (all of the surviving examples are later) and swords are rare, but this probably stems simply from the fact that (unlike in some mainland areas) such items were not used in the funeral ritual rather than necessarily from a real scarcity or absence.  Warfare may, thus have been quite open and fast-moving with much use of missile weapons before the fight was decided in close fighting with sword and thrusting spear.

In Britain, as across the rest of Europe, archaeological and other evidence suggests that important changes took place in the decades around 600.  The practice of warfare was one of the things affected by these transformations.  It seems that fighting became larger in scale and more based upon the clash of the close-packed ‘shield-walls’ familiar from accounts of Viking warfare.  These changes buried the ‘world’ of Arthur, the historical details of any actual Arthur figure, and the type of warfare he may have engaged in for ever.

---

[1] I have used the terms Anglo-Saxon, Saxon and English interchangeably in this essay.

Gender in the Merovingian World

In Tours, the holy woman Monegund had a little garden next to her cell where she could go to be alone (VP 19.1).  One day while she was there a woman was able to watch her from a neighbouring rooftop: ‘she gazed upon her importunely, filled with worldly desires’, and consequently went blind until Monegund healed her.  This is not the easiest passage to unravel.  Who was filled with worldly cares – Monegund or the importunely watching woman – is not entirely clear; what exactly the worldly cares were and what the transgression was that robbed the woman of her sight are likewise fairly obscure.  These ambiguities, however, have surely been present to any readers of or listeners to this text and, as Derrida (1967/1997) showed, no text has a stable, originary meaning, present to itself.  This – decidedly queer – tale raises interesting and important aspects of gender in the Merovingian world.

Gender remains an understudied area of early Frankish social history and, indeed, of early medieval history in general.  Thirty years after Joan Scott’s (1986) classic paper on gender as a useful category of historical analysis, which set out so clearly the difference between women’s history and gender history, early medieval historians remain – in relation to students of other disciplines and periods within the broad church of medieval studies – comparatively unsubtle investigators of the topic. ‘Gender’ is frequently still employed as a simple placeholder, signalling token comments about women.[1]   It is hard not to paraphrase Scott: my understanding of (for example) aristocratic patronage is not changed by the knowledge that women participated in it.  Analytical confusion remains about whether one is studying women’s history or gender history (e.g. Smith 2001; Nelson 2004; Hen 2004; etc.). Many articles putatively dealing with ‘gender’ treat women as a trans-historical or ahistorical category (egregiously, Nelson 1996, the continuing wide citation of which is indicative of the general conceptual malaise).  Although now examining masculinity as well as femininity (e.g. Hadley (ed.) 1999) early medieval historians have not (e.g. Halsall 1996; 2010a) moved on sufficiently from the idea that gender is the social construction placed upon the biological category of sex. Man and woman are timeless categories; only the way people write about or envisage them changes.

But, as Judith Butler (1999; 1993) showed, it is not so simple to separate biological sex from gender.  The way we inhabit our bodies and social space, and live our lives, is something negotiated, discursively, with society, however unconsciously, from the moment we enter the world of the Symbolic (on which see, e.g., Lacan 1978, p.238-375; Evans 1996, p.201-3).  The relationships between gender, biological sex and sexuality are not straightforward and there is no meaningful point in our lives at which we experience sex separately from gender.  The fundamentally bipolar scale of biological sex (genitalia, chromosomes, etc.) and the processes of human sexual reproduction do not and cannot in themselves determine sexuality or marital or family structures, not least because their perception and representations are as enmeshed in language as are the notions of sex, marriage, family, or idealised woman- or manhood.  Butler’s famous concept of performativity came not just from the idea of performance but also from the speech-act category of the performative: phrases (e.g. ‘I pronounce you man and wife’) which create the thing they describe (Butler 1999, p.185 is quite clear on this). It is thus not the case that a somehow natural ungendered or pre-gendered body performs gender as something extrinsic to itself, something which, unlike that notionally sexed but pre-gendered body, is contingent upon social and linguistic structures.  Gender is constituted in its performance; behind the mask is nothing or – rather – only another mask (Deleuze 1994, p.27-30).  Philosophically, this is old news and has not gone unchallenged (Copjec 1994), but it is indicative of early medieval history’s resistance to ‘theory’ that this has yet to make much impact on our subject.

Smith (2004), for example, mentions Butler’s notion of performativity once (2004, p.17), but seems to have misunderstood it, to mean performance.

Identification is an ‘enacted fantasy’ (Butler 1999, p.185).  All identities are structured ultimately by fantasy and desire: a mental image of what the identity-ideal might mean in terms of its relations with people of other identities, and a desire to occupy that space.  That means behaving towards people of other identities in particular fashion, in the manner that (one thinks) people expect, as well as in ways that might be thought best to achieve one’s own ends.  The motion towards this ideal can never be complete; however close one comes, one can never simply be the identity-ideal: thus the enactment of fantasy.

Consider another story from Gregory of Tours’ gallery of the unexpected: the case of the cross-dressing Poitevin at the tribunal of the rebellious nuns of Holy Cross (Histories 10.15). Nancy Partner (1993, p.418, 439) noted that Gregory describes this person as a man in woman’s clothing, not as a woman, but are we authorised to take that as any sort of basis for discussion, any more than we should take modern Republicans’ insistence that a transgendered person is a man ‘dressed up as a woman’ (Badash 2016)?  Are we permitted to refer to the Poitevin as ‘he’ rather than ‘she’ on Gregory’s say-so?  Whether the Poitevin lived her life as a woman or lived his life dressed as a woman is impossible to say.  Gregory reports that, when questioned, the Poitevin said that s/he had made the decision to dress as a woman because s/he was incapable of manly work (opus virile) but we cannot know what to make of this because the whole text, above all its evident distinction between biological sex and material cultural gender, is soaked in a gendered discourse of power.  One of the story’s attractions is that it is so undecidable at so many levels that trying to claim what Gregory, or the Poitevin, ‘really meant’ by any of the crucial phrases is quite pointless. In noting the story’s separation of the ‘man’ from the ‘clothing’ we are however presented with a difficult course to chart between the Scylla of essentialism and the Charybdis of endless, disabling relativism. How do we investigate agency or resistance to normative views?  How might we identify the difference between a man who dressed as a woman and a biological male who lived as a woman?

If this person was born anatomically male, that clearly did not determine the construction of his/her gender.  We do not know what was meant by the claimed incapacity to perform opus virile, or whether or how this related to her/his sexuality (Dailey 2015, p.000; Halsall forthcoming).  Thorpe’s somewhat misleading translation, upon which Partner (1993) relies too heavily, suggests that (qua man) the Poitevin was impotent.  If so, his gendering did not relate to his sexuality.  Perhaps he was married; perhaps his sexuality had little bearing on his decision to marry and try to raise a family.  Other sixth-century options visible in the story include gendering oneself as male, but refraining from all sexuality (as a monk or secular ecclesiastic), or as a married woman and similarly abstaining from sex (as a nun – married to Christ).   Even what we might suppose was the normative Frankish family unit, with a mother, father and children, was not the sole vehicle for socialisation.  The masculine child could leave his natal family and spend a long period either in another family household or in the overwhelmingly masculine society of a retinue (Halsall 2010a; 2010b).

The rebellious nuns’ accusation, that the Poitevin dressed to conceal his masculinity and thus work in a house of religious women (with no implications for his sexuality), adds yet a further dimension.  If we accept the bishops’ decision that the accusation was groundless and trust the Poitevin’s own account of the reasons for his/her costume, another unanswerable question arises: if identity is an ‘enacted fantasy’, what was the mental image or ego-ideal in question?  This story illustrates how anatomical sex, sexuality, gender, living arrangements, marriage and so on combine in a galaxy of historically-contingent ways.[2]

One advantage of studying northern Gaulish social history is, however, that while documentary sources are comparatively scarce a vast archive of material cultural data exists in hundreds of cemeteries and thousands of sixth- and seventh-century burials.[3] These have much to tell us, not least because the dead were commonly interred in an archaeologically-visible funerary costume, alongside various other artefacts – the whole (elements of clothing and the other items) referred to by the shorthand term grave-goods.   Most serious scholars have moved away from interpretations of this practice as either ‘Germanic’, ‘pagan’, or both (for critique, see Effros 2003; Brather 2004a).  Where skeletal data survive, intact burials allow us to compare a range of treatments of the body with their biological age and sex.  These allow us to separate (from our perspective) the Merovingian social treatment and experience of the body from its anatomical sex, physiological development (whether or not ever experienced separately from socio-linguistic categories) and so on.  This does not undermine Butler’s point but rather provides a way of further interrogating it in a past context. It does not solve the problems identified earlier but adds another critical perspective and an additional data-set.

As a further example, let us take Grave 32 at the cemetery of Ennery, in which a body sexed anthropologically as male was interred with objects – most notably a necklace – otherwise only found with biologically female skeletons (Simmer 1993, p.46-47; Halsall 1995, p.78-94; 2010b).  We cannot say whether the person known as ‘Ennery 32’ thought of himself as a man who dressed as a woman, or of herself as a woman, or was thought of as a woman by the community or as a man who dressed, or lived life, as a woman, or what their preference was (if any) for sexual partners (if any), or whether any link existed between that and the funerary costume.  It does not allow us to identify dominant and subversive readings (and say which was which), or by whom such opinions were held.  This case and others like it do, however, demonstrate that, while necessary to correct many fundamental errors at the time, the 1990s debate over whether the skeletal or the material evidence was ‘correct’ (e.g. Lucy 1997, Effros 2000) was theoretically misconceived.  Equally, Halsall’s (2010c) attempt to read ‘discrepancies’ between the two forms of data in terms of transgression is too unsubtle.

The sixth-century furnished burial ritual was a public event (Halsall 2010d, p.203-60).  Communal norms were important in determining the appropriate numbers and forms of goods deposited with the dead of different genders at particular stages of the life-cycle.  Thus, if the physical anthropology of Ennery 32 is correct (and there seems no good reason to doubt it), the implication is seemingly that someone born with male physical attributes was interred publicly in a way that permitted his/her identity and lifestyle to be reflected in how s/he was laid to rest before an audience.  That leaves many questions unanswerable but challenges modern ideas about what is ‘natural’, ‘traditional’ or ‘normal’ or that ground their references to the ‘natural’ and the ‘normal’ in claims about long-term European history, and does so without idealising the medieval past.  The funerary data also permit the exploration of change through time; the quantity of material recovered allows the dating of burials to roughly twenty-five-year periods (Périn 1980; Legoux, Périn & Vallet 2009).

The construction of gender had changed since the late Roman period.  Classical Roman gender construction had largely revolved around the notion of the civic Roman male, a single set of model behaviours concerned with moderation, self-control and so on.  Not only the female but also the barbarian and the animal revolved around this central focus, praised for closeness to it, derided for distance from it (Foxhall 2013 for excellent recent overview).  Whilst a barbarian might be able to perform masculine Romanness so effectively that his non-Roman origins were effaced, a woman (by which I mean anyone living a life as a woman) was apparently prevented by her sex from ever fully occupying that central position.  It would be wrong, however, to suppose that that central position was fixed or attainable even by men.  Classical civic masculinity concerned the citation of an always-unreachable ideal.  The implication of the fact that the single gendered ideal was masculine and performative is that its enactment distinguished the social actor from a set of characteristics that were gendered feminine.  In an inversion of de Beauvoir’s (2011, p.293) famous dictum one was not born a man but only ‘became’ one.  Implicit in Roman thought is the idea that (in modern terms) the default, ‘pre-social’ gender was feminine or, in terms more easily assimilable with Roman concepts, that outside this performance, while not everyone was a woman, everyone was ‘womanly’.

The cemetery archaeology’s implications are that this situation had changed by the early Merovingian period (see below).  The roots of this transformation should be sought in later fourth- and fifth-century shifts.  After the formal separation of the different branches of imperial service during the Tetrarchic period, a new, martial form of Roman masculinity seems to have emerged.  Late imperial military culture strongly suggests that the army began to develop new identities, stressing the opposite of the traditional model Roman male: barbarian and even animal (Halsall 2007, p.102-10).  However, this new masculinity required the older one to exist, to round out its symbolic context.  Both forms were, in the final analysis, based on Roman imperial legitimation.  The ‘martial model’ inverted civic masculine ideals but the efficaciousness of that strategy depended squarely upon those traditional ideals’ authority and prestige.  This ‘barbarizing’ identity, and the legitimacy of military and civil titles, also relied upon a link to the emperor who embodied military and civil office and both models of masculinity.  While (in a late Roman context) some could emphasize martial masculinity’s wild, fierce, animal, barbarian traits to imply a weakness in civic manhood, it always had to be remembered that anyone with an education would view such characteristics as lesser, uncivilized and womanly.  The traditional or orthodox concepts thus remained in place even as an alternative ‘reading’ of its symbols and bases (its citations, in Butler’s terms) emerged, which flagged up their implicit blind spots.

Employing Simon Critchley’s (2014) insights into Derridian philosophy, I have called this a deconstruction of classical gender (Halsall, forthcoming).  As stated, identity is a constant movement towards an ideal, so these renegotiations, oscillations and redefinitions of what such an unattainable object of identification was were crucially important in the lived gender of late and post-imperial people.  The situation I have described endured for a couple of generations beyond the deposition of Romulus ‘Augustulus’ because the post-imperial western kings continued to occupy, however notionally, positions within the imperial hierarchy and, like the emperor, to embody points at which civic and martial masculinities came together.

I would suggest that, by playing with and valorising the traditional civic male ideal’s ‘constitutive other’, this ‘oscillation’ within the masculine ideal enabled more actively idealised feminine traits to emerge, based more strongly around sex, the body and reproduction and thus not simply dependent upon the emulation of the male.  This, I suggest, is visible in the furnished inhumations, where two distinct artefact sets were employed to denote masculinity and femininity (Halsall 1995, p.79-83; Effros 2003, p.99-100, 154-63 for notes of caution; Halsall 2010b for response).

It is immediately clear that the feminine cluster of artefacts is composed, overwhelmingly, of costume or bodily adornments: brooches, earrings, hairpins, dress-pins, bracelets, necklaces.  Items that might be representative of female work – bread-cutters, weaving batons, spindle-whorls – are quite rare in a Frankish context and occur in what seem to be quite prestigious contexts.  It is also important to note that – with the important exception of weaponry – most earlier Merovingian technical and decorative expertise (especially in metalwork) was invested in items of female adornment.  Archaeologically-visible elements of sixth-century masculine costume, are usually relatively plain.  Written sources allude to jewelled or otherwise decorated belts in élite contexts, but we rarely find archaeological examples; the belt evidently did not need decoration for its symbolic weight.

The ontology of gender can be investigated through the material construction of the life-cycle, as revealed through grave-goods deposition, and the written sources’ sparse information read in the light of these more plentiful data.  The basic account can be fairly brief, and relates primarily to burial in sixth-century northern Gaul.[4]   In early childhood, the furnishing of burials made little recognition of the child’s sex.  The lack of stress in communal relations produced by a small child’s death (which should emphatically not be understood as equating with a lack of grief) meant that these burials were rarely the occasions for the expenditure of resources.  It is not unusual to find feminine artefacts in the burials of slightly older children, possibly reflecting the betrothal of the deceased child (betrothal as young as eight is attested in written sources: Venantius, Carmina 4.26).  The gendering (and indeed sexing) of the female child could then be brought about by the demands for marriage alliances that lay at the core of communal politics.  (In line with the foregoing discussion, the use of the terms male and female throughout the remainder of this chapter implies nothing about the anatomy of the person in question.) What has been called ‘interpellation’ into the gendered world (Butler 1993, p.xvii) may not have been decisive in the child’s earliest years.  It is worth noting that masculine items remain scarce in burials of subjects who died before puberty.  The exceptions invariably concern unusual and lavish burials of members of the élite (e.g. Werner 1964).  It might appear, on that basis, that aristocratic male children could be seen as sexed/gendered earlier than those of other levels of society.  This would have meant a very real difference in the experience of gender at different social levels.  It might, however, equally be that different demands of burial ritual, a need to mark a distinction from other social strata and different stresses caused by the death of a male heir, caused a young boy’s body to be gendered in his funeral in a way not necessarily reflected during his life.

The first major change in the construction of masculine and feminine identity occurred at puberty, when adolescent women were frequently interred with a wide range of feminine objects, most notably the jewellery mentioned above.  The Pactus Legis Salicae notes that the wergild of a woman of child-bearing age was three times that of a typical free, Frankish ingenuus (cp. PLS 24.8, 41.1).  Two clauses imply that this increase in legal value was triggered by the woman’s having begun to bear (or breast-feed) children (PLS 24.8: postquam coeperit infantes habere; 41.16: postquam ceperit nutrire), whereas another appears to see the woman’s value in terms of the victim being of an age where there was potential child-bearing, which it sees as beginning at twelve (PLS 65e). This evidence, alongside written references (including inscriptions), suggests that the onset of the menarche brought the immediate sexualisation of the female body, marriage and potential child-bearing.  The interpellation of a child into a fully sexed/gendered female role was sudden and brutal.  The comparatively high investment in the funeral display for women of this age presumably relates to the tension within community relations that their deaths might cause, undoing fairly recently-made marriage alliances when any children produced would have been very young.

The funerary record suggests that the masculine body remained ungendered at this age.  Weaponry is rarely found with the bodies of anatomical males in their teens, although other masculine items are occasionally present.  The difference from the feminine teenager is underlined by the fact that the Pactus marks a point of transition for males at twelve years.  It is the boy under twelve years who has the higher, 600-solidus wergild but who is also considered legally incapable (PLS 24.7).  Clearly, the law dealt with a young boy’s murder in terms of his potential to become a man. The superficial discrepancy between the legal and archaeological evidence is probably explained by the fact that the wergild relates to compensation for the damage done to a family by a killing, whereas the burial data are explained by the rift in social relationships between families caused by a death.  The transition to legal responsibility, seemingly at twelve, was marked by the first cutting of the boy’s hair – apparently making the male body visibly different from the female for the first time (PLS 24.2).  It is important to note, nonetheless, that the acquisition of legal responsibility and a non-feminine hair-style does not, according to the archaeologically-visible data at least, appear to have resulted in the full recognition of a masculine identity.  Even in the context of public ritual display, when we might expect gender distinctions and attendant costumes to have been made more visible, the deceased male was not strongly gendered.  Yet, young men’s sexual desires were well recognised by contemporaries.  Cassian (Institutes, 6.1) said that the spirit of fornication, against which a monk must strive, commenced its attacks with puberty.  It is noteworthy, however, that young male sexuality was less clear-cut.  Before twenty, sexual explorations between ‘boys’ could be treated as ‘games’ (ludi) and treated less severely than sexual relations between older men (Cummean Penitential, 10; see also Penitential of Theodore 1.2.4).  The funerary record is suggestive here too.  Double interments from the sixth-century cemetery of Ennery (‘graves’ 6 and 8: Simmer 1993, p.33, 37) and the seventh-century necropolis of Audun-le-Tiche (grave 103: Simmer 1988, p.50-53, 55), contain two males with their arms laid deliberately on top of each other (whether interlocked is difficult to say).  The subjects were all younger than about twenty.  Perhaps the somewhat ambivalent semiotics of the burial display (Halsall 2010c, p.347-9) were considered acceptable among men of this age-group.

Once into his twenties a male could expect to be buried with masculine items, including weaponry.  A more clearly gendered masculine identity could now be ascribed to the dead.  A male might by now have been serving in another household for some time, creating bonds between his family and the head of that household (Halsall 2010b).  It is likely, too, that he was betrothed.  As with younger women, his death could threaten a range of inter-familial relationships requiring their maintenance through the burial ritual.  Women dying at this stage of the life-cycle might, however, be buried with a larger overall number of grave-goods but with a less lavish display of bodily and costume adornments (see also Stauch 2008).  This stage of the female life-cycle appears to have lasted until about forty.  Again, the investment of resources in their burials should be seen in terms of the stress placed upon inter-familial alliances brought about by their deaths.  Even at forty, a woman’s eldest male children, and younger daughters, might still be unmarried, raising complicated questions about inheritance between the families involved (PLS 101, 110; Decretio Childeberti 1.1).

From about thirty, some males were buried with the full panoply of weaponry.  Evidence for male age of marriage is vague but on balance it seems that it was typically in the late twenties (Halsall 2010a).  Males dying between about thirty and sixty were typically the members of the early Merovingian community who received the most lavish interments, with numerous masculine items and other grave goods.   Explanation should again be sought in communal politics and in the synchronicity of the different generations’ life-cycles.  A man dying in his thirties or forties was probably married but his children, especially sons, would still be minors.  Even in his late forties, his eldest daughters would likely only recently have married and his eldest sons would be unlikely to be much older than twenty, well short of male marital age.  Given that marriage and fictive kin alliances around his daughters and sons were doubtless intended to secure his support or allegiance, his death questioned such ties.  Perhaps more importantly, his death left open the problem of succession as his eldest sons would not have established themselves sufficiently to inherit his communal standing.  The issues of his widow’s property and remarriage (PLS 44, 100) also pertained.  It is unsurprising, therefore, that the deaths of men of this age-group could necessitate the fullest ritual attention in feasting, gift-giving and the display of the family’s ability to lay their dead to rest in the most appropriate fashion.[5]  Importantly, the distinctions in the lavishness of male grave-furnishing become sharpest at this stage in the life-cycle.

Women older than about forty and men above sixty received far less attention in their grave-furnishing, mostly receiving interments that, like children’s, were ‘neutral’ in gendered terms.  While the burials of women above forty were sparsely furnished and rarely contained jewellery, male graves, though poorly furnished overall, occasionally included a weapon or other masculine items.  One should, however, hesitate before assuming a low esteem for old people on this basis.  The lack of investment in the archaeologically-visible elements of burials, as explained above, relates primarily to the relative lack of tension caused by deaths at this age, when children had reached maturity and established households and communal standing.  It is significant, nonetheless, that gender was more frequently marked in masculine burials than in female.

It is immediately apparent from this account that the experience of time itself was gendered.  Masculine and feminine life-cycles were constructed differently, with socialisation and sexualisation running at significantly different speeds.  It is noteworthy, too, how much more sexualised was the feminine life-course and how much more violently it was punctuated by gendered social expectations or by interpellation into the sex/gender system (on which, classically, Rubin 1975).  Feminine socialisation and the sexualisation of the female body were apparently simultaneous and rapid processes, supposedly occurring at the onset of the menarche.  This apparently physiological rule is immediately destabilised, however, by the Pactus’ displacement of the grounds for higher female legal status from the actuality of child-birth or breast-feeding into the potential for child-bearing, determined according to age-group (PLS 65.e).  The bearing of this upon female sexuality is unclear.  Same sex relations posed no threat to the legitimacy of children and a marriage was possibly considered to remain valid even if the woman ceased to cohabit with her husband.  Gregory (Histories 5.32) makes it clear that it was the rumour of a wife’s sexual relations with another man that brought about the violent dispute between two Parisian families; her leaving her husband is relegated to an ablative absolute sub-clause.

The female bodily adornment suggested by the grave-goods merits further reflection.  The indications are that the forms of clothing not normally archaeologically visible could also be highly decorated.  It is interesting to note the location of these items, mostly associated with the hand and arms, the breast and the hair (especially if tied up or covered, signifying marriage), the female bodily areas that Frankish law penalised touching (PLS 20.1-4, 104).  The unavailable female body is not concealed but highlighted.  Significantly, too, the costume which most emphasized these aspects began to be worn at puberty and was mostly set aside after forty: during the life-cycle stage wherein a woman’s wergild was three times higher than a freeman’s.  The law saw this as the age of the (legitimately) sexually-active woman.  The costume in which women were interred might be interpreted as symbolizing feminine ideals: the beautiful, chaste, good wife and mother.  Merovingian female status and identity were based solidly on reproductive sex and marriage.

Male socialisation took much longer, between puberty and about thirty (Halsall 2010b) and appears to have established a man’s right to marry and start a household.  It seems likely too that in the sixth century it was closely related to the acquisition of an ethnic identity, Roman or Frankish.  The Pactus (e.g. PLS 41) only assigns ethnicity to adult males.  Before about twenty it might also be that, as noted, male sexuality was given some latitude.  Although, by the sixth century, the martial, Frankish model was generally considered more dominant, various ideals of identity still existed.  The variation between burials with weapons and those without possibly relates to this and the divergence between lavishly- and poorly-furnished graves among males at this stage of the life-cycle could reflect the varying success with which males managed to achieve a fully gendered identity as the head of a household.  It is even possible that Gregory’s Poitevin chose his/her particular lifestyle as a result of the demands of that earlier phase of the life-cycle.  The crucial ‘citational’ points in the masculine life-cycle were, furthermore, much less (notionally) dependent upon physiological development and reproductive sexual practice. Again, these physiological or bodily developments should not be seen as neutral or pre-social canvasses upon which society inscribed gendered identities (Butler, 1993).

Significantly, before marriage, males were referred to as pueri (boys) regardless of physiological age (Halsall 2010b).  That this phase of the life-cycle could produce various models of masculinity is important.  As noted, the difference between mature adult males buried with weapons and those buried without might relate to ethnicity: Frankish identity was closely related to military service (Halsall 2003, p.46-48).  If the martial (Frankish) model of masculinity increasingly dominated, it nevertheless continued to relate primarily to the civic ideal represented by the free Roman male.  Traditionally, as described above, civic masculinity had been differentiated from a backdrop of ‘womanly’ characteristics, which the martial forms had in turn played upon and valorised.  Consequently, the two gendered ‘poles’ most visible in the archaeological record, the martial masculine and the feminine, were constructed not as binary oppositions but in relation to the civic masculine ideal, now weakening in its social importance.  This seemingly more ‘bipolar’ construction of gender, I suggest, opened up the space within which the anonymous Poitevin/e of Gregory’s Histories or Ennery 28 operated (Halsall, forthcoming).  The feminine was no longer necessarily the constitutive outside of the masculine.  Young males’ performance and citation of gender seems to have been much less ‘teleological’ than in female experience (Freeman 2010:4, 8, for teleological time).

Fifth-century developments in Christian attitudes to gender have been well discussed (e.g. Cooper & Leyser 2001).  The debate over, and trend towards, the absolute repudiation of sex was perhaps as yet less resolved in Gaul than elsewhere and the emphasis upon chastity and abstinence produced a convergence of masculine and feminine ideals that resembled classical gender-construction. The Church might have been a repository for many of the old ideals of civic masculinity and classical gender as the martial model became more dominant in secular life.  Many sixth-century bishops entered the church after a secular career, marriage and children.  A good example is Gregory of Langres who entered the church after a long career as a comes (VP 7.1; Wemple 1985, p.132-6 for a useful survey of the earlier Merovingian church’s ambivalent attitude to married clergy).  Unsurprisingly, the principal virtues in claiming ecclesiastical and religious authority were moderation and continence rather than asceticism. Gregory of Langres only had sexual relations with his wife for the purpose of procreation, according to his great-grandson, biographer and namesake (VP 7.1). 148 out of 293 recorded sixth-century saints are bishops (James 1982, p.55).  Although asceticism was well-known, the trend towards extreme, competitive, public renunciation noted elsewhere is far from dominant in early Merovingian religious life.  Seclusion, secrecy, modesty, and an especial worry about the vainglory that might beset ascetic virtuosi seems more heavily emphasised.  Gregory of Langres again furnishes an example. After his ordination he performed his asceticism in secret (VP 7.2).  The punishment of vainglorious (or potentially vainglorious) holy men is recorded (VP 15.2; Histories 4.34) and, famously, Vulfolaic’s attempt to emulate Syrian stylites in the Ardennes was quickly stifled by the local episcopate (Histories 8.15).  The flaws of those condemned as ‘false’ holy men doubtless included ‘showy’ asceticism (Histories, 9.6; 10.25).  Even the recluse Lupicinus’ excessive and ultimately fatal self-mortification was performed beyond public view, while walled up in a cave (VP 13.1). Some male holy men only decided to abandon the worldly possibilities of marriage and family during the long phase of pueritia.  Bracchio, Venantius, Patroclus and Leobardus all, according to Gregory, chose the religious life when faced with the social expectation that they would marry, during or at the end of this phase of the life-cycle, sometimes on their father’s death (VP 9.1, 12.2, 16.1, 20.1).  This further underlines the variety of models of gender and sexuality that were seemingly available to males at this age.

Many of the known sixth-century religious women (most of whom had royal connections) similarly had family lives before entering their religious vocation: Chlothild, Radegund, Ingitrude and Monegund.  Female monasticism was in its infancy and it was uncommon for young people to enter or be given to monasteries (De Jong 1996). Many holy women, like Ingitrude and Monegund, lived as individual religious or formed small communities of like-minded women near the great urban shrines, like that of Saint Martin, Tours.  Many such communities, like Ingitrude’s, did not survive their founder’s death and even large and prestigious houses like St John, Arles, or Holy Cross, Poitiers, could have chequered histories.  This was not, however, so different from contemporaneous male monasticism.  Early Merovingian models for female religious life were rarely provided by cloistered virgins.  The author of Saint Genovefa’s life, it has been pointed out (Wood 1988, p.378), found difficulty in finding an exemplar for a female vita, so that Genovefa appears rather episcopal (the bishop of Paris is a figure significantly absent from the text).  Over half of Venantius’ Vita Radegundae concerns her life before she secluded herself in her nunnery of Holy Cross.  Even the cult of the Virgin Mary seems not to have flourished in sixth-century Gaul. Gregory’s Glory of the Martyrs, for instance, contains very few miracles associated with her veneration (GM 8-10, 18-19).  Something like the single, masculine gendered ideal known in classical thought might have persisted in this sphere.  In the prologue to his life of Monegund (VP 19), Gregory tells us that the Lord provides holy examples of how to live: not just men but also women, who struggle ‘not feebly but manfully’ (non segniter sed viriliter), presumably accounting for Monegund’s inclusion among the ‘Fathers’.

Returning to the secular sphere, sixth-century masculine objects, apart from the – largely undecorated – belt, symbolised things that men did to or with things or other people: weapons, tools, flints, strike-a-lights, knives.  These have many implications for identity and personhood (see Cohen 2003).  By comparison, the highlighting of the female body suggested by the feminine grave-goods implies that woman was the object of the gaze.   Female identification was performed bodily, publicly, in the gaze of the community.  The ‘barbarian’ term Venantius used to describe the bejewelled costume that the saint laid aside when she renounced the world (and adequately describing that revealed by sixth-century grave-goods: ‘composita sermone ut loquar barbaro stapione’, Vita Radegundae 13.30) seems to mean ‘stepping out’ costume (McNamara, Halborg & Whatley (trans.) 1992, p.76, n.55).  Such visibility was doubtless not purely feminine; especially in northern Gaul, sixth-century society was very visible or ‘optic’, in which resources were heavily invested in public display and in which public ritual was crucial to the operation of society.  The local community, its politics and ‘structures’ were performed in front of an audience. The big cemeteries, the foci for the burial ritual discussed earlier, are one part of this.  The Pactus Legis Salicae stresses public performance as a strong component of legal procedure (PLS 46, 58, 60).  The investment of wealth in costume and display must go some way towards explaining the ephemeral traces of earlier Merovingian rural settlement.[6]    Although the sixth-century northern Gallic rural community might have been quite large, comprising several small settlements, it was nevertheless a very local arena (Halsall 2012).

This argument does not imply female passivity or a lack of status.  Although institutions or formal codes might be established by and for a masculine social élite, the everyday inhabitation of social spaces could create, in the interstices of the boundaries set out on such a ‘social map’, social standing and power (Halsall 1996, p.22-23).  Nevertheless, the evidence suggests that the motion of feminine identification in particular required performance in the gaze of others.  Ontologically, to be woman was inseparable from that visible occupation of a point in public space, in the communal gaze.  The implication of that might be that the female gaze differed subtly from the male, with the woman, instead of just looking, ‘seeing herself being seen’.  This would result in a form of bio-political control which could easily be imposed by women on other women.  In comparative perspective none of this is surprising, but it applied particularly in sixth-century Gaul.  It is interesting to compare that conclusion with the invisibility of women in so many of our documents.

Here we may return to Monegund, the Poitevin/e of the nuns’ tribunal and ‘Ennery 32’.  Whatever is going on in the story of Monegund and her neighbour, part of the problem is that the woman is looking. Whatever lay behind Gregory’s text, it is also clear that looking at a woman was almost impossible without some sort of thinking about sex or ‘worldly desires’ taking place.  This only underlines why female religious had to be taken out of the normal public arenas and enclosed (Dailey 2014).  This is in some ways underscored by cases like the Poitevin/e and the person buried at Ennery.  Both instances, as they are known to us, come from very public contexts: a tribunal in a civitas-capital (the Poitevin/e stepped forward coram omnibus: ‘in front of everyone’) and a burial ritual.  Lacking the overt references to the sexual body, Ennery 32’s costume would possibly be appropriate for a man who lived life dressed as a woman (Halsall 2010c, p.342-3).  Perhaps.  But by concentrating too heavily on a ‘transgressive’ interpretation, that reading ignores the important, if seemingly obvious, point that follows from Butler’s philosophy of gender: whatever the biological sex of the deceased’s bodily remains, someone who lived life as a woman and died between the ages of forty and sixty was, simply enough, buried in a costume appropriate for a woman aged between forty and sixty.  However, the deceased didn’t dress themselves for their funerals.  Ennery 32’s family (I assume) prepared and clothed the body for its burial, which must have meant that – assuming that the anthropological sexing is correct – they were well enough aware of the disjunction between the physical body and its social skin and were prepared to represent the latter in public display.  What does the expenditure of resources on this funeral display have to say to thesis that the lavishness of burial is related to the stress in society?  All we can do is throw out a series of questions and possibilities, and mistrust any attempt to close down or (hetero-)normalise them.

Important changes occurred around 600, including very significant transformations to the burial ritual (Halsall 1995, p.262-9).  Grave-goods declined and the relative investment in more permanent, above-ground commemoration increased; cemeteries became more numerous as earlier large sites ceased to be used by multiple settlements; the audience present at funerals probably declined commensurately.  These changes, surely related to the social hierarchy’s increased stability and the local élite’s security, clearly affect our evidence’s ability to answer questions about gender in the same way as before.  Nonetheless several issues deserve our attention.

The first is the shift in grave-goods to the masculine.  There is usually a marked relative decline in the number of burials marked by feminine grave-goods and in the number of types of grave-goods used to signify feminine identity (Halsall 1995, p.110-63).  Those that remain are often rather less ornate than before (Legoux, Périn & Vallet 2009).  By contrast, though weapons, the objects employed to signify at least one – probably dominant – form of masculinity in the sixth century, decline in numbers and variety, the percentage of adult male burials with at least a token weapon increases (e.g. Halsall 1995, p.134).  These aspects are surely interlinked.  More importantly for current purposes, at least some males begin to be buried in a more archaeologically visible costume, notably the great decorated plate-buckles that dominate the seventh-century Gaulish artefactual record (Lorren 2001; Legoux, Périn & Vallet 2009, p.32-35). Further, the most elaborate brooch-forms of the century, disc brooches with cabochons and filigree, can be found in masculine as well as feminine graves (e.g. Liéger, Marguet & Guillaume 1984).  Did seventh-century feminine attire really become as plain as the cemetery data suggest?  Maybe not.  Embroidered clothing, for instance, is rarely archaeologically visible (Vierck 1978; Effros 1996).  Women were possibly no longer buried in a costume that bore much relationship to formal attire in life.  This would nonetheless be significant and the change is not easy to explain away, even if it is yet more difficult to explain or describe with certainty.

Some points can nevertheless be made. The greater solidification of the social hierarchy and, compared with the sixth century, a consequent rise in the relative importance of descent and of male family- or lineage-heads seems to lie at the heart of the issue. We might, nevertheless, focus again on the issues of visibility and performance.  The later phase of the cemetery of Lavoye (Meuse) suggests that in the seventh century only one of the three different groups that had earlier used the site continued to do so for all of its burials (Halsall 1995, p.138-9, 141-2).  The others apparently interred mainly adult males at the site.  This might imply that, as the everyday social arena contracted to the still more local level of the individual settlement, political gatherings, whether at the level of the old community or higher, became more exclusively masculine affairs.  A possible development in seventh-century northern Gaul is the fragmentation of the former civitates into smaller pagus communities at a level between the old villae and the civitas (Halsall 2012).  Politics may have been played out, furthermore, to a greater extent away from the old public spaces of the cities and more in the private, or privatised, spaces of royal or aristocratic palaces or villas/estate centres, or rural monasteries.  This might have increased the political invisibility of most women.  Perhaps the masculine body now became something more crucially in the political gaze, and masculinity something performed rather differently than before.  We might consider the location of the greatest foci of display in masculine costume, the ever-larger, often lavishly-adorned plate-buckles and counter-plates, across the lower stomach.  Like the high, narrow-waisted doublets and full breeches of the late sixteenth and earlier seventeenth century, they might be read as highlighting virility.

Plate-buckles are not common in sixth-century graves and are found overwhelmingly in masculine burials (Lorren 2001, p.197).  When women began to wear these objects, they were initially small and plain.  Masculine versions then gradually became larger and more ornate as women begin to wear decorated plate-buckles.  This is an interesting dynamic.  One might suggest that as a hitherto masculine symbol was adopted by women, men redefined it, and so on.  Such a dynamic has many parallels. One is the way in which, as women began to be depicted in Roman funerary art with symbols of learning earlier only used to depict the model of a good father and husband, male sarcophagi began to show such things less and stress pastimes like hunting instead (Huskinson 1999).

Why did women start wearing hitherto masculine objects?  Perhaps this is another indication of the shift suggested earlier, back to a more ‘monopolar’ construction of gender.  The good wife, the good woman, has virtue demonstrated partly through masculine artefacts.  One might be able to move from there to consider the most lavish disc brooches in a similar light.  Rather than simply being a feminine accessory, these objects possibly transcend the normal plainness of female dress adjuncts to be as decorated as masculine brooches.  Grave 147 at Audun-le-Tiche buried in a clearly prestigious location and possibly even a founder-burial, might take on a different aspect (Simmer 1988, p.65-6, 68). This is one of the most lavish female graves of the cemetery, with unusual displays of jewellery. What is surprising, after a study of sixth-century graves, is that its occupant was an old woman.  But, if the postulated redefinition of gender around 600 is correct, it might be that the display here is not fundamentally primarily related to sex, reproduction and the family, as would have been the case earlier, and perhaps not an example either of how distinction in seventh-century funerary display was made by consciously breaking communal rules about the correct grave-goods or of the more straightforward display of familial wealth and standing (as argued by Halsall 1995, p.264-5).  The grave’s dimensions are exceptional, requiring more effort than usual for their construction.  Perhaps the recognition of female status had come to be based on other issues than marriage, sex and child-bearing.

Have we returned to something more like the Roman construction of gender?  The clearest difference between the two situations is that the martial model of masculinity is now dominant. Various factors including the spread of Frankish ethnicity had led to a shift in the ways of raising the army away from the ‘ethnically-based’ force of the sixth-century to an army based more on aristocratic retinues in the seventh (Halsall 2003, p.53-56).  This model was now, in important regards, more socially restricted, its performance confined to more select political gatherings of men.  Few if any women – indeed only a minority of men – could approach this ideal. In this connection it is noteworthy that the distinctly ‘virile’ figure of the Neustrian Queen Fredegund, leading her army to battle, belongs not to the sixth century, when she lived, but to the early eighth (LHF 36).  That was profoundly different from the classical situation.

However, another model, religious and consciously asexual, stood outside reproduction and the family.  The shifts that occurred in this area of Merovingian Gaul were subtle, but, at the risk of over-schematization, some suggestions are possible.  One aspect of the greater entrenchment of local aristocratic power around 600 was, as is well known, the increased focus on more organised rural monasticism, often, and too-simplistically, associated with the Irish holy man Columbanus (Fox 2014).  Away from towns and frequently less subject to effective episcopal domination, these houses became the foci for much of seventh-century Gaul’s secular as well as religious politics.  With grants of immunity, abbots could become almost as significant religious and political figures as bishops (Rosenwein 1999, p.59-96), while the estates with which their houses were endowed sometimes put them on a course towards equally land-owning importance.  Female rural monasteries often had a male house attached, not least to provide the nuns with officiants at mass.  Consequently, these ‘double-houses’ were often governed by women.

This flourishing and comparatively more stable monasticism possibly led to a greater level of entry into religious life during childhood or adolescence.  Some saints of traditional type are nevertheless recorded.  Saint Arnulf is unusual for having led a successful life as a warrior as well as a politician before becoming bishop of Metz (Vita Arnulfi 4-5) and eventually retiring to be a recluse in the Vosges. Nonetheless, many bishops, as before, entered the church after secular service: Eligius, Audoin and Desiderius of Cahors are famous examples.  Yet, marriage is rarely mentioned and, as with some sixth-century holy men, the crucial point seems to have come at the end of their period of unmarried apprenticeship: their pueritia.  Most of the famous abbots seem to have entered the church quite early.  If the period of pueritia had become more teleological in its gendered outcome, as the non-martial form of masculinity now dominated, sexual renunciation and the ecclesiastical life was perhaps the only alternative for those who could choose.  This limitation of gendered models was probably yet more acute for females if the suggested shift towards a more monopolar construction of gender is correct.  Chaste widows entering the religious life are attested in the seventh century (most famously with Saint Balthild) but the period’s saints’ lives are dominated by women who dedicated themselves to a virginal life early on, so that the struggles that occurred when their families’ demanded that they marry, become a common feature of the vitae.  Obviously, as more successful female houses were established and as many emerging Frankish noble families listed nuns and abbesses among their number, the availability of this gendered model increased exponentially.  It would be interesting to examine whether this was accompanied by a change in the significance of the cult of the Virgin, or whether the vitae of Roman virgin-martyrs circulated more than before.  Crucially, then, while the idealised ‘centre’ of the gendered secular world could never truly be occupied other than by a male, in the religious sphere, such a central role could be occupied, effectively, by a woman as well as a man.  It is small surprise that the seventh century was an age of great abbesses.

The roles and possibilities opened up by the performance of an identity can never remain fixed. The impossibility of stasis stems not only from the demands of, in Bourdieu’s terms, the habitus, of everyday coexistence, or the constant renegotiation of roles and statuses in Giddens’ theory of structuration, although this is clearly a major element.  It stems similarly from the fact that identities are not entities, but idealised, unattainable objects and so can never have a fixed, stable, authentic meaning.  As Derrida said in his classic (1988) discussion of performatives, the communication of such signifieds in daily performance always, through their iterability, risks miscommunication and slippage (see also Butler 1993).  There was no such single thing as medieval gender or a finite number of genders, and there was always scope for change, active renegotiation, play, transgression and, yes, repression (in all senses).  The everyday lived existence of the people of Merovingian northern Gaul between say 500 and 650 was constituted by gender and its performance.  Gender, reciprocally, delineated performance and went far beyond the simple construction of men and women.

More than that, the close analysis of Merovingian gender, in all its surprising malleability, changeability through time, and stubborn refusal to conform to the patterns that modern readers expect of ‘the medieval’, is a valuable political resource in the present, when so many politicians appeal to a mythical past ‘normality’ and ‘naturalness’ in order to attempt to fix modern sexuality and gender-relations in a particular mode.  The Merovingian case study tells us that there is nothing timeless and ‘natural’ about past gender and sexuality, other than the impossibility of keeping them in their place.

 

Endnotes

[1] Recently, e.g. Fox (2014, p.13): ‘The question of gender is another component of the same question. … [A]fter [Columbanus’] death women came to occupy an increasingly important place in the leadership of Columbanian communities’.

[2] This example illustrates the profound error committed by those who wish to label the chaste clergy a ‘third gender’, as though there are otherwise only two. McNamara (2002).

[3] For introductions in English to Merovingian cemeteries and their study, see Dierkens & Périn (1997); Périn (2002); Effros (2003).

[4] For fuller accounts, see Brather (2004b); (2005); (2008); Halsall (1995); (1996); (2010b); Lohrke (2004); Stauch (2008).

[5] Many of the famous early Merovingian ‘tombes de chef’ are of males of this age group such as the occupant of the famous grave 319 at Lavoye (Meuse) and the two ‘chefs’ (graves 11 and 13) recently excavated at Saint-Dizier: Joffroy (1974, pp.95-101); Truc & Paresys (2008).

[6] See, above all, Van Ossel (1992) and Peytremann (2003). Excellent overviews can be found in Burnouf et al., (2009), p.95-153, and Catteddu (2009), esp.p.25-87. In English, see Périn (2002); (2004); Zadora Rio (2003).

References

Primary Sources
Cassian, Institutes: Patrologia Latina 49 ed., J.-P. Migne (Paris, 1874): John Cassian: The Institutes (Ancient Christian Writers 58), trans. B. Ramsay (New York, 2000).
Cummean, Penitential: The Irish Penitentials ed. L. Bieler with D.A. Binchy (Dublin, 1963), pp.108-35.
Decretio Childeberti: MGH Legum, Section 1, Vol.4, Pt 1, Pactus Legis Salicae ed. K.-A. Eckhardt (Hanover, 1962), pp.267-9; The Laws of the Salian Franks trans. K. Fischer Drew (Philadelphia, 1991), pp.156-9.
GM: Gregory of Tours, Gloria Martyrum: MGH SRM 1.2 ed. B. Krusch & W. Levison (Hanover, 1969), pp.34-111. Gregory of Tours. Glory of the Martyrs, trans. R. Van Dam (Liverpool, 1988)
Histories: Gregory of Tours, Historiarum Libri Decem: MGH SRM 1.1 ed. B. Krusch & W. Levison (Hanover, 1951). Gregory of Tours. History of the Franks, trans. L. Thorpe (Harmondsworth, 1974).
VP: Gregory of Tours, Vita Patrum: MGH SRM 1.2 ed. B. Krusch & W. Levison (Hanover, 1969), pp.211-94; Gregory of Tours. Life of the Fathers, trans. E. James (2nd edition; Liverpool, 1991); Gregory of Tours. Lives and Miracles (Dumbarton Oaks Medieval Library 39), ed. & trans. G. de Nie (Cambridge, MA, 2015), pp.1-297
LHF (Liber Historiae Francorum): La Geste des Rois des Francs. Liber Historiae Francorum ed. B. Krusch, & trans. S. Lebecq (Paris, 2015).
PLS: Pactus Legis Salicae: MGH Legum, Section 1, Vol.4, Pt 1, Pactus Legis Salicae ed. K.-A. Eckhardt (Hanover, 1962). The Laws of the Salian Franks trans. K. Fischer Drew (Philadelphia, 1991)
Penitential of Theodore: Die Bussordnungen der abendländischen Kirche nebst einer rechtsgeschichtlichen Einleitung, ed. F.W.H. Wasserschleben (Halle, 1851), pp.182-219; J.T. McNeill & H.M. Gamer (trans.), Medieval Handbooks of Penance. A translation of the Principal Libri Poenitentiales (New York, 1990 [originally 1938]), pp.179-215
Venantius, Carmina: Venance Fortunat. Poèmes. ed. & trans. M. Reydellet 3 vols. (Paris, 2002-4)
Venantius, Vita Radegundae: MGH SRM 2, ed. B. Krusch (Hanover, 1888), pp.358-77; Sainted Women of the Dark Ages, trans.  McNamara, J.-A., Halborg, J.E., & Whatley, E.G., (Durham NC, 1992), pp.70-86.
Vita Arnulfi: MGH SRM 2, ed. B. Krusch (Hannover, 1888), pp.426-46
Vita Genovefae: MGH SRM 3, ed. B. Krusch (Hanover, 1896), pp.204-38; Sainted Women of the Dark Ages, trans.  McNamara, J.-A., Halborg, J.E., & Whatley, E.G., (Durham NC, 1992), pp.19-37.

Secondary Sources
Badash, D., 2016. http://www.thenewcivilrightsmovement.com/davidbadash/watch_ted_cruz_tells_transgender_women_to_only_use_the_bathroom_at_home accessed 2 August, 2016
Beauvoir, S. de (2011). The Second Sex (C. Borde & S. Malovany-Chevallier trans.) London: Random House.
Brather, S. (2004a). Ethnische Interpretationen in der frühgeschichtlichen Archäologie: Geschichte, Grundlagen und Alternativen (Ergänzungsbände zum Reallexikon der Germanischen Altertumskunde 42) Berlin: de Greuter.
Brather, S. (2004b). Kleidung und Identität im Grab: Gruppierungen innerhalb der Bevölkerung Pleidelsheims zur Merowingerzeit. Zeitschrift für Archäologie des Mittelalters 32, 1-58.
Brather, S. (2005). Alter und Geschlecht zur Merowingerzeit. Soziale Strukturen und frühmittelalterlichen Reihengräberfelder. In J. Müller (Ed.) Alter und Geschlecht in ur- und frühgeschichtlichen Gesellschaften (pp.157-78) Bonn: Habelt.
Brather, S. (2008). Kleidung, Bestattung, Identität: die Präsentation sozialer Rollen im frühen Mittelalter. In Brather (Ed.) (2008), pp.237-73.
Brather, S. (Ed.) (2008). Zwischen Spätantike und Frühmittelalter: Archäologie des 4. bis 7. Jahrhunderts im Westen (Erganzungsbände zum Reallexikon der germanischen Altertumskunde 57) Berlin: de Greuter.
Brubaker, L., & Smith, J.M.H. (Eds.) (2004). Gender in the Early Medieval World: East and West, 300-900. Cambridge: Cambridge University Press.
Burnouf, J., Arribet-Deroin, D., Desachy, B., Journot, F., & Nissen-Jaubert, A. (2009). Manuel d’Archéologie Médiévale et Moderne. Paris: Collin
Butler, J. (1993). Bodies that Matter. On the Discursive Limits of “Sex”. London: Routledge
Butler, J. (1999). Gender Trouble. Feminism and the Subversion of Identity. 2nd edn. London: Routledge
Catteddu, I. (2009). Archéologie médiévale en France: Le premier Moyen Âge (Ve-XIe siècle). Paris: La Découverte
Cohen, J.J. (2003). Medieval Identity Machines. Minneapolis: University of Minnesota Press.
Cooper, K., & Leyser, C. (2001). The gender of grace: impotence, servitude and manliness in the fifth-century west. In Stafford & Mulder-Bakker (Eds.) 2001, pp.6-21.
Copjec, J. (1994). Read My Desire. Lacan against the Historicists. Boston: . Repr. 2015 London: Verso.
Critchley, S. (2014). The Ethics of Deconstruction. Derrida and Levinas. 3rd edn. Edinburgh: Edinburgh University Press.
Dailey, E.T. (2014). Confinement and exclusion in the monasteries of sixth-century Gaul. Early Medieval Europe 22.3, 304-35.
Dailey, E.T. (2015). Queens, Consorts, Concubines: Gregory of Tours and Women of the Merovingian Elite. Leiden: Brill
de Jong, M. (1996). In Samuel’s Image: Child Oblation in the Early Medieval West. Leiden: Brill
Deleuze, G. (1994). Difference and Repetition (P. Patton, Trans.) London: Bloomsbury.
Derrida, J. (1967). De la Grammatologie Paris: Minuit
Derrida, J. (1988). Limited Inc. (A. Bass & S. Weber, Trans.) Evanston, Ill: Northwestern University Press
Derrida, J. (1997). Of Grammatology (G. Chakravorty Spivak, Trans.) 2nd edn. Baltimore: Johns Hopkins University Press
Dierkens, A., & Périn, P. (1997). Death and burial in Gaul and Germania, 4th-8th century. In L. Webster, & M. Brown (Eds.) The Transformation of the Roman World, AD 400-900 (pp.79-95). London: British Museum
Effros, B. (1996). Symbolic expressions of sanctity: Gertrude of Nivelles in the context of Merovingian mortuary custom. Viator 27, 1-10
Effros, B. (2000). Skeletal sex and gender in Merovingian mortuary archaeology. Antiquity 74, 632-9
Effros, B., (2003).  Merovingian Mortuary Archaeology and the Making of the Early Middle Ages. Berkeley: University of California Press
Evans, D. (1996). An Introductory Dictionary of Lacanian Psychoanalysis. Hove: Routledge.
Freeman, E., (2010). Time Binds. Queer Temporalities, Queer Histories. Durham NC: Duke University Press
Fox, Y. (2014). Power and Religion in Merovingian Gaul: Columbian Monasticism and the Frankish Elites. Cambridge: Cambridge University Press
Foxhall, L. (2013). Studying Gender in Classical Antiquity. Cambridge: Cambridge University Press
Hadley, D.M. (Ed.) (1999). Masculinity in Medieval Europe. London: Longman
Halsall, G. (1995). Settlement and Social Organization. The Merovingian Region of Metz. Cambridge: Cambridge University Press
Halsall, G. (1996). Female status and power in early Merovingian central Austrasia: the burial evidence. Early Medieval Europe 5.1, 1-24. Repr. in Halsall (2010d), pp.289-314.
Halsall, G. (2003). Warfare and Society in the Barbarian West, 450-900. London: Routledge
Halsall, G. (2007). Barbarian Migrations and the Roman West, 376-568. Cambridge: Cambridge University Press
Halsall, G. (2010a). Merovingian masculinities. In Halsall (2010d), pp.357-81
Halsall, G. (2010b). Growing up in Merovingian Gaul. In Halsall (2010d), pp.383-412
Halsall, G. (2010c). Material culture, sex, gender, sexuality and transgression in sixth-century Gaul. In Halsall (2010d), pp.323-55
Halsall, G. (2010d). Cemeteries and Society in Merovingian Gaul: Selected Studies in History and Archaeology, 1992-2009. Leiden: Brill
Halsall, G. (2012). ‘From Roman fundus to Carolingian Grand Domaine: crucial ruptures between late antiquity and the middle ages.’ Revue Belge de Philologie et d’Histoire 90, 273-98
Halsall, G. (forthcoming). ‘Classical gender in deconstruction’, in R. Le Jan & S. Joye (Ed.), Gender and Elites. Paris
Hen, Y., (2004). Gender and patronage of culture in Merovingian Gaul. In Brubaker & Smith (Eds.) (2004), pp.217-33.
Huskinson, J. (1999). Women and learning: gender and identity in scenes of intellectual life on late Roman sarcophagi. In R. Miles (Ed.) Constructing Identities in Late Antiquity (pp.190-213). London: Routledge
James, E., (1982). The Origins of France.  From Clovis to the Capetians, 500-1000. London: MacMillans.
Joffroy, R., (1974). Le Cimetière de Lavoye. Nécropole Mérovingienne Paris: Picard
Lacan, J. (1978). Le Séminaire, Livre II. Le Moi dans la théorie de Freud et dans la technique de la psychanalyse. J.-A. Miller (Ed.) Paris: Seuil.
Legoux, R., Périn, P., & Vallet, F. (2009). Chronologie Normalisée du Mobilier Funéraire Mérovingien entre Manche et Lorraine. 3rd edition. Condé-sur-Noireau: Association Française d’Archéologie Mérovingienne.
Liéger, A., Marguet, R., & Guillaume, J. (1984). Sépultures Mérovingiens de l’abbaye to Saint-Evre à Toul (Meurthe-et-Moselle). Revue Archéologique de l’Est et de Centre-Est 35, pp.301-17
Lohrke, B., (2004). Kinder in der Merowingerzeit. Gräber von Mädchen und Jungen in der Alemannia (Freiburger Beiträge zur Archäologie und Geschichte des ersten Jahrtausends 9) Rahden: Leidorf
Lorren, C. (2001). Fibules et plaques-boucles à l'époque mérovingienne en Normandie : Contribution à l'étude du peuplement, des échanges et des influences, de la fin du Ve au début du VIIIe siècle. Condé-sur-Noireau: Association Française d’Archéologie Mérovingienne
Lucy, S.J., (1997). Housewives, warriors and slaves? Sex and gender in Anglo-Saxon burials. In E. Scott and J. Moore (Ed.), Invisible People and Processes.  Writing Gender and Sexuality into European Archaeology (pp.150-68). London: Leicester University Press.
McNamara, J.-A., (2002). Chastity as a third gender in the history and hagiography of Gregory of Tours. In Mitchell & Wood (Eds.) (2002), pp.199-209.
McNamara, J.-A., Halborg, J.E., & Whatley, E.G., (Trans.) (1992). Sainted Women of the Dark Ages. Durham NC: Duke University Press
Mitchell, K., & Wood, I.N. (Eds.) (2002). The World of Gregory of Tours. Leiden: Brill.
Nelson, J.L., (1996). Gender and genre in women historians of the early middle ages. In ead., The Frankish World, 750-900 (pp.183-97). London: Hambledon.
Nelson, J.L. (2004). Gendering courts in the early medieval West. In Brubaker & Smith (Eds.) 2004, pp.185-97
Partner, N.F. (1993).  No sex, no gender. Speculum 68, pp.419-43
Périn, P., (with contribution by R. Legoux), (1980). La Datation des Tombes Mérovingiennes. Historique – Méthodes- Applications. Paris & Geneva: Droz.
Périn, P. (2002). Settlements and cemeteries in Merovingian Gaul. In Mitchell & Wood (Eds.) (2002), pp.67-99.
Périn, P. (2004). ‘The origin of the village in early medieval Gaul. In N. Christie (Ed.), Landscapes of Change: Rural Evolutions in Late Antiquity and the Early Middle Ages (p.255-78). Aldershot: Ashgate.
Peytremann, E. (2003). Archéologie de l’Habitat Rural dans le Nord de la France du IVe au XIIe Siécle (2 vols.). Condé sur Noireau: Association Française d’Archéologie Mérovingienne
Rosenwein, B. (1999). Negotiating Space: Power, Restraint and Privileges of Immunity in Early Medieval Europe. Manchester: Manchester University Press.
Rubin, G. (1975). The traffic in women: notes on the “political economy” of sex. In R.R. Reiter (Ed.),  Toward an Anthropology of Women, (pp.157-210). New York: Monthly Review Press.
Scott, J. Wallach (1986). Gender: A useful category of historical analysis. American Historical Review 91, pp.1053-75
Simmer, A. (1988). Le Cimetière Mérovingien d’Audun-le-Tiche: Archéologie d’Aujourd’hui No.1 Paris: l’Association Française d’Archéologie Mérovingienne.
Simmer, A. (1993). La Nécropole Mérovingienne d’Ennery (Moselle). Fouilles d’Emile Delort (1941) Woippy; l’Association Française d’Archéologie Mérovingienne.
Smith, J.M.H. (2001). Did women have a transformation of the Roman world? In Stafford & Mulder-Bakker (Eds.) 2001, pp.22-41.
Smith, J.M.H. (2004). ‘Introduction: Gendering the early medieval world’ in Brubaker & Smith (Eds.) 2004, pp.1-19
Stafford, P., & Mulder-Bakker, A.B., (Eds.) (2001). Gendering the Middle Ages. Oxford: Blackwells (= Gender and History 12.3 [2000]).
Stauch, E. (2008). Alter ist Silber, Jugend ist Gold! Zur altersdifferenzierten Analyse frühgeschichtlicher Bestattungen. In Brather (Ed.) 2008, pp.275-95.
Truc, M.-C., & Peresys, P., (2008). Trois tombes d’exception a Saint-Dizier. In Varéon, C., (Ed.), Nos ancêtres les barbares. Voyage autour de trois tombes de chefs francs (pp.50-69) Saint-Dizier: Somogy.
Van Ossel, P. (1992)., Établissements Ruraux de l’Antiquité Tardive dans le Nord de la Gaule (51e Supplément à Gallia). Paris: Éditions du CNRS
Vierck, H.E.F. (1978).  La ‘Chemise de Sainte-Bathilde’ a Chelles et l'influence byzantine sur l'art de cour merovingien au VIle siècle. In Actes du Colloque international d'archéologie: centenaire de I'abbé Cochet (pp.521-570). Rouen: Musée départemental des Antiquités de Seine-Maritime
Wemple, S.F. (1985). Women in Frankish Society. Marriage and the Cloister. Philadelphia: University of Pennsylvania Press.
Werner, J. (1964). Frankish Royal Tombs in the Cathedrals of Cologne and Saint-Denis. Antiquity 38, pp.201-16
Wood, I.N., (1988). Forgery in Merovingian hagiography. In MGH Schriften 33: Fälschungen im Mittelalter. Internationaler Kongress der MGH, München 16-19 Sept., 1986 (V: Fingierte Briefe. Frömmigkeit und Fälschung. Realienfälschung) (pp.369-84). Hannover: Hahn
Zadora-Rio, E. (2003). The making of churchyards and parish territories in the early-medieval landscape of France and England in the 7th-12th centuries: A reconsideration. Medieval Archaeology 47, pp.1-19