The creation of Carolingian homosociality

I’ve recently been thinking about masculinity again and was reading with interest Rachel Moss’ discussions of her new project on medieval homosociality. Rachel states:

I will argue that homosociality was a cultural mindset that, by privileging bonds between men, made it possible to create networks of socially-codified relationships that supported hegemonic norms and facilitated the structuring of patriarchy.

I also discussed the concept briefly with her on Twitter, when she was explaining that she saw homosociality as “a set of socio-cultural mechanisms of which socialisation is an element”.

Rachel works on the late medieval period, so most of the key sources she used in her previous book on fatherhood aren’t available for early medievalists or only in very small quantities: conduct literature written by the laity, gentry letters and chivalric romances. In fact there’s probably only one Carolingian status group about whose socialisation and male bonding we know in detail; that’s monks, whose homosociality is likely to be considerably different from that of other men.

What I started thinking about instead is what type of consciously-created homosocial  structures exist in the premodern West (and in the Carolingian period specifically) and when and why they are created. I’m focusing on deliberately-created structures because there’s one immediate difference between the twenty-first century West and most earlier periods. Both work and education are now substantially less sex-segregated than they used to be. Educating boys and girls together (past their early years) isn’t the historic norm and while officially mixed-sex working places have existed, they’ve normally been less common than single-sex ones, from the weaving shed to the regiment. So the question becomes why homosociality in the premodern world wasn’t just left to arise naturally, as a by-product of such segregation. Why was it made to happen? (I realise that you can have socio-cultural mechanisms for homosociality that arise incidentally from institutions/cultural patterns created for other ends, but “mechanism” does imply at least some purpose).

These are my four suggestions of where and why such (male) homosocial structures were created:

1) Team-building

Leaving (male) bonding to chance is fine in most environments, but in ones where a group has to work together effectively, they may not be adequate. The obvious examples here are warfare and sport (which at least in the medieval period, often shade into one another). The kind of formal and informal bonding required provides the context for the best known disaster of Carolingian homosociality, as reported in the Annals of St-Bertin for 864:

The Young Charles, whom his father [King Charles the Bald] had recently received from Aquitaine and taken with him to Compiègne, was returning one night from hunting in the forest of Cuise. While he meant only to enjoy some horseplay with some other young men of his own age, by the devil’s action he was struck in the head with a sword by a youth called Albuin. The blow penetrated almost as far as the brain, reaching from his left temple to his right cheekbone and jaw.

(Translation from Janet Nelson, The Annals of St-Bertin, Manchester, 1991)

Charles never recovered properly, suffered from epilepsy and died a couple of years later, aged around 19. Mostly, however, bonding via horseplay was less fatal and the training involved might perhaps have saved these young men’s lives in future combat. Homosocial practices for creating strong teams may also have been important for those involved in such dangerous occupations as fishing or mining.

2) Creating a group identity: support and control

It may be an obvious point, but homosociality is never about friendships between all men: it’s about particular groups of men who also share some other common characteristics (race, class, age, occupation etc). Some homosocial practices are as much about creating a group based on these other factors as on a specifically male one. A modern example is how working-class boys may create groups/gangs in reaction to their schools, creating their own form of masculinity by rejecting the vision offered by middle-class educators. Homosocial groups of this sort can potentially be very small, such as a pair of men becoming blood-brothers/sworn friends.

There are obvious overlaps between group identity and team-building, but there are also attempts to create group identity where teamwork isn’t a concern or participants may normally be relatively isolated. I’m thinking in particular here of efforts to encourage bonding between early medieval priests, whether it’s simply rural priests having monthly meals together on the Kalends or the kind of communal living for city priests suggested by Chrodegang of Metz. The example of priests also shows that homosocial structures promoting group identities can emerge from below, or they can be imposed from above. The shared meals of Carolingian priests were probably a custom that they developed themselves, and which bishops, such as Hincmar of Rheims, either tried to suppress or convert to their own ends in order to control their diocese better.

Another possible Carolingian example of creating a homosocial group identity is the early medieval guild. Unfortunately, we don’t know the details about them that we do for late medieval guilds, which clearly did consciously create homosociality and ideals of masculinity. Otto Gerhard Oexle thought that women as well as men were members of Carolingian guilds, though he was relying on a condemnation (by Hincmar of course!) of riotous communal meals, which might be guild-related, but might not be. Such guilds provided local support networks, so a group identity (whether specifically homosocial or not), was obviously important.

3) Creating patronage networks

One point of consciously creating homosocial institutions/groups is to bring together men who might not otherwise meet one another. The classic examples of this are the masonic lodge and the late medieval guild. What’s important to note is that these don’t simply have horizontal bonds for support. They also encourage vertical bonds, providing access to networks of favour and patronage.

One of the interesting questions is the extent to which chivalry (both the ethos itself and specific chivalric orders) can be considered as a patronage and favour network. Certainly, when knights spare other knights in battle, while slaughtering their inferiors, that’s an important result of a homosocial bond.

For the Carolingian period charter witnessing clearly is a homosocial patronage network of this kind – there are a few female witnesses, but they’re exceptional. The more difficult question is to what extent early medieval assemblies and the court functioned as predominantly homosocial patronage networks.

Early medieval assemblies had, so far as we know, overwhelmingly male participants. While we don’t have much sense of how they functioned culturally, Hincmar’s discussion of general assemblies in De ordine palatii, with the king gathering information from men coming from all the regions and greater and lesser men coming together certainly suggests opportunities for networking and patronage.

The issue in assessing Carolingian courts as homosocial institutions is that we don’t know how sex-segregated they were. There are two possible models here: one is what you might call the Beowulf model, with the king and his warriors in the mead hall and the queen as the only elite woman present. (Presumably there are always both male and female underlings doing the hard work). The alternative is the late medieval and early modern court which are mixed-sex groups, with elite women present in attendance on the queen. Such mixed-sex courts may have times when they split into single-sex groups (such as the bathing pool at Aachen), but more time is spent together.

Unfortunately we don’t really know how many women there were at Carolingian courts and who they were (apart from Charlemagne’s female relatives). At some point I need to trawl the sources and see if I can find any examples of non-royal women there. The evidence so far suggests Charlemagne’s court is nearer the Beowulf model, but that’s still a bit uncertain.

4) Teaching homosocial practice

My final suggested reason for the deliberate creation of homosocial institutions is in order to ensure that homosociality is done right. One of the recurring research questions about homosociality is how it relates to homoeroticism. In most Western societies, some forms of homoeroticism have been disapproved of, even in societies, such as the Roman empire, where same-sex sexual acts might be acceptable. There’s therefore always been a need to draw boundaries between the right kind of (male) homosocial relationships and the deviant kind, even if these boundaries may be drawn in very different places in particular societies.

There’s a chicken and egg question here: do controls over homosocial activity develop to prevent ‘deviance’ or is ‘deviance’ invented as a method to control the socially-important area of homosocial activity? (The latter is what Eve Sedgwick suggests). In a few cases, however, you can see homosocial institutions that are developed specifically to avoid the wrong kind of male-male encounter. For example, there’s the monastic communal dormitory, whose development in late antiquity Albrecht Diem discussed.

The dormitory is at once a device for preventing homosexual activity between the monks, by allowing their surveillance, and a way of bringing them together as a group (rather than having them sleep in separate cells). It’d be interesting to look at other homosocial institutions to see if some of them have the same double purpose.


As I said at the start, I’m not sure it’s possible to look at Carolingian homosocial institutions in much detail (in the way that I presume Rachel Moss will for late medieval homosociality). But an overview like this does suggest that there are fewer homosocial institutions in the early Middle Ages than later. Part of this is the result of living in a less complex society: there are no universities, for example, which form an important site for homosociality from the twelfth century onwards. In contrast, almost all of the homosocial institutions that exist in the earlier Middle Ages continue into later periods.

There is one important exception to this: homosociality seems to become less important in court culture, with women playing a more prominent role there. But this may indicate more of a shift in homosociality than its decline. To what extent is chivalry a way of replacing the male-dominated court of the early Middle Ages with a ‘separate sphere’ for men within mixed-sex courts? My suspicion is that homosociality (or at least its conscious cultivation) became more important later in the Middle Ages, although I’m not sure of a good way of measuring that. But just thinking about such possible changes in homosociality over the medieval period (as well as the many continuities visible) may be useful.

2 thoughts on “The creation of Carolingian homosociality

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s