History Matters 5: Analysing Carolingian patriarchy

In my last two posts, I’ve been starting trying to do what Judith Bennett calls for in History Matters: Patriarchy and the Challenge of Feminism (University of Pennsylvania Press, 2006): historicising patriarchy (in the sense of ‘a system of social structures and practices in which men dominate, oppress and exploit women’). This is my provisional attempt to take the analysis a little further and look in a little more detail at how Carolingian patriarchy worked and how patriarchies might change (without necessarily affecting the patriarchal equilibrium).

One of the problems in doing this kind of analysis is thinking about how the components of the system break down. In my earlier attempt at periodisation, I’d mentioned religious ideology, socio-economic forms and the status of the male elite. Bennett quotes (p 58) Sylvia Walby’s theories of ‘modern England’ (from 20 years ago) talking of six patriarchal structures: ‘mode of production, paid work, the state, male violence, sexuality, culture’. ‘Paid work’ isn’t relevant for the early Middle Ages and I’d add in the church as an organisation. With that, how does Carolingian patriarchy fit together?

1) There’s a feudal mode of production. The basic unit of production is the conjugal peasant household, with men’s work and women’s work. This gives peasant women an important economic role within the household, but ties them into marriage and domesticity: they can’t easily manage economically on their own. There’s also a coercive extraction of surplus by the nobility in most cases, which means that if necessary the ‘lord’ must be able to act as a warrior (or at least control a band of warriors).

As a mode of production, ‘feudalism’ is normally seen as dominant from late antiquity to the Industrial Revolution. The importance of physical coercion to this gradually goes down, as a ‘justice’ system increases which can do the physical coercion for the landlord, but it’s still predominant until at least the early modern period.

2) The state consists of the combination of an attenuated sub-Roman state, with hierarchical offices and a rudimentary legal system, and a barbarian warrior aristocracy. From the Roman side comes the ideological presumption that office holding, legislation and the administration of justice are male preserves. From the warrior aristocracy comes the practical reality that women (and boys) are not suitable for carrying out public roles, because of the violence required, (as represented in the need for the count to keep a gallows).

The idea of public office as male-only is extremely long-lived, stretching from the classical period through to the twentieth century. The predominantly warrior nature of the aristocracy reappears (after a break) in the sub-Roman period and continues on until the early modern period (the ideology may continue even longer).

3) The institutional church has a male-only priesthood and hierarchy, but a few offices for women religious, which are kept subordinate to male offices. There is an increasing co-operation of the church and the state/ruling class in the Carolingian period. The church gets protection from this and also donations both of material and personnel (since it is not entirely self-reproducing). In return, the ruling class get ideological and administrative support (and also additional military support from monastic militias).

The institutional form of the Catholic church developed in the fourth century and hasn’t changed substantially since, although in Protestant areas of Europe there were no official positions for women between the sixteenth and the late twentieth century. The particular configuration of the relationship between the church and the ruling seems to me to be confined to the seventh to eleventh century (say Gregory I to Gregory VII), with the key factors of family monasteries with monastic oblation (which tied nobles and monasteries together) and threats from ‘pagans’ (which meant churches needed the protection of rulers more than they feared them).

4) The early Middle Ages was a violent time, and male violence was a key part of the Carolingian patriarchy. Firstly, male violence was used by the ruling classes to reinforce their rule. Secondly, women who were not under the protection of a powerful man were vulnerable to violence from other men (one of the typical images of the helpless is ‘widows and orphans’). Thirdly, a level of domestic violence was condoned (although Carolingian reformers did try to prevent uxoricide).

Although violence against women hasn’t disappeared in the modern world (and there are arguments about the extent to which domestic violence is still sanctioned), violence against women outside the home has generally declined over the centuries, with the increase of the rule of law. (Where you’d put any historical break is less clear).

5) In terms of sexuality, the Carolingian world had socially imposed universal monogamy (SIUM) for the laity and (theoretical) celibacy for religious men and women. In theory the church tried to impose a single (female) standard of sexuality on both sexes. In practice, a lot of double standards continued, (although it’s not clear how easily Carolingian women were ruined).

On the issue of change and continuity on this one, it depends a lot on what scale you look. In one respect you can see Christian marriage as a continuity for almost 2000 years (until secularism in the 1960s – and even in the US, ideas of marriage have changed substantially, with much easier divorce). On another, there is a big change when marriage is no longer predominantly an alliance between two families via the exchange of women (a pattern which shows continuities between prehistoric times and the nineteenth century for the upper classes). On another level, you could see the big break as at the Reformation, when celibacy became a marginal option.

I’d argue in this case, however, for a shorter periodization, covering around 450-900, as an era which both extended monogamous marriage and valorised it. Kate Cooper has shown how in the late antique world, after the fourth century ructions over priestly celibacy, other texts tried to recreate an ideal of the good Christian marriage. The Merovingian period again exalted virginity, but it also began to extend marriage to slaves’ relationships. In the Carolingian period there is further protection for slave marriage and a general valorisation of monogamous marriage (as discussed by Pierre Toubert and Katrien Heene’s The Legacy of Paradise: Marriage, Motherhood and Woman in Carolingian Edifying Literature, which is unfortunately very hard to get hold of). After 900, the Cluniacs and others again start denigrating marriage and exalting the superiority of celibacy as writers like Jerome had done. I would therefore see the late antique/Carolingian period as one of several ideological periods of marriage, which while it has parallels to other periods, is distinctive.

6) Any discussion of culture can always get bogged down in this baggy monster, but I want to stress two particular bits of ideology that I haven’t mentioned before, which are visible in the Carolingian period, but also in many other eras. One is the ideology that women at all social level should perform one particular kind of ‘women’s work’ – varying forms of cloth production/needlework, whether it’s Roman women winding wool, Charlemagne’s daughters spinning and weaving or Victorian ladies embroidering. Such activities may be largely nominal, but they do symbolically stress the domestic role of all women. Secondly, there is the Christian ideology of womanhood. This is a varied mix, which can be used to stress either the equal humanity or the inferiority of women, but it always entails the subordination of wives to husbands.

Overall, I’d see the Carolingian period as the developed form of an early medieval type patriarchy. Although there are many long-term continuities, I’d see significant discontinuities with the end of the Roman world and after the Gregorian Reformation. In contrast to these periods, I’d see the early medieval patriarchy as marked by relatively low ideology/misogyny. Instead, male violence played the crucial role – government required violence to suppress violence and those unable to wield violence were vulnerable. Ideology was only needed to justify power-holding when holding power didn’t mean wielding a sword. (This meant that women who were able to wield violence could hold power).

The prime motor for these changing forms of patriarchy, meanwhile, doesn’t seem to be women. It’s hard to see how the social, economic and legal position of women changes substantially between the fall of the Roman empire and the late medieval development of women’s independent economic activity which Bennett talks about. The patriarchal equilibrium seems to be as much about ensuring that changes in male activity (and in particular the changing relationships between elite male groups) doesn’t allow any new opportunities for women to rise, rather than responding to a rise that women have already made. A model of patriarchies that seems them as purely reactive (which the metaphor of patriarchal equilibrium may encourage) may give us a misleading view of how they have historically operated.


10 thoughts on “History Matters 5: Analysing Carolingian patriarchy

  1. A fascinating analysis. However, as a historian who focuses on the later Middle Ages, I would quibble slightly with your periodization of feudalism. It might interest you, in this context, to look at the November 2008 issue of “Gender and History” (which you may already know, and if so, apologies!). It’s devoted to periodization, and in this context I am thinking particularly of Martha Howell’s article, “The Gender of Europe’s Commercial Economy, 1200-1700.”


    • I realise that I’ve been inconsistent in this: I did mention Martha Howell’s work in a previous attempt at periodization, but in this post I was unconsciously reverting to conventional broad periodizations of economic history and not taking account of her work. One of the problems I’m finding with trying to look at these very long-term continuities/changes is that I’m running up against the limits of my own historical knowledge (particularly on socio-economic history). I often have a vague sense that some particular idea or organisation changed after 1500, but no more. Maybe one of the things we need is more in depth collaborative work by women’s historians of different periods on some of these themes, so they can actually try and pin down what changes when, rather than talking past one another.


      • That is, of course, one of the real difficulties of doing this kind of thing, but I agree that it’s necessary. There are a couple of interesting articles on the long-term periodization of women’s history against the longterm economic periodization of European history–Judith Bennett, again, and Olwen Hufton spring to mind. If you are interested, I can supply the references.


  2. I would, just as I did last time, object to your classification of the Carolingian economy as feudal-mode. The operational state and its successful demands for service make this problematic. It rests on a lot of assumptions about how military service is organised, about what happens to peasant surplus and so on that really only demonstrably work in royal heartlands in the North, and in extreme cases, on the estates of Corbie. Again, Chris Wickham’s Land and Power deserves a mention.

    Also only really true if you define your area very tightly: women unable to hold office (countesses, abbesses, both of whom sometimes give justice even though they can’t dispense violence) or power (any woman with landed property who therefore has labouring dependents); and women doing only women’s work, whereas in many areas then and now peasant women work in the fields alongside men for some tasks. In both of these cases, it’s obviously often-to-mostly true what you say even in less masculinised areas, but the existence of exceptions makes a generalisation of this size quite misleading I think. Much of the rest is solid and useful and deserves to be rescued from this foundation on a monostructural demesne economy.


    • My argument doesn’t rest on presuming a demesne economy (indeed I didn’t mention that). I’m using feudal mode only in Chris Wickham’s very general sense of surplus being taken from a subsistence peasant. On this, I’ve now had a look at some of ‘Land and Power’, in which Chris seems to be changing his mind on the topic (particularly in the afterwords to Chapter 2 ‘The uniqueness of the East’). He’s now, I think, back at a situation in which he sees there to be only three exploitative modes of production (slave mode, feudal mode, capitalist mode), while the superstructure above this can vary, such as in whether the surplus is taken as rent or tax. The Carolingian period is definitely feudal mode, except where you still have free peasants (Chris’ non-exploitative ‘peasant mode’). (From a feminist view, there’s an interesting question about peasant patriarchy: if peasants get to keep their surplus by working less or consuming more, is this shared equally by the whole household, or is it preferentially peasant men who get to benefit?)

      Even if women’s work and men’s work aren’t mutually exclusive categories, as long as you have some work that only one sex ‘can’ do (and there is clear evidence of this: find me women ploughing or men spinning) then you’ve got a domestic economy that can only be based on a mixed-sex couple.

      I didn’t spell out the significance of warriors as fully as I should have done. If you want to exploit peasants (whether on your land or free neighbours), you ultimately have to rely on your own supply of coercive force, because there isn’t a state one you can turn to. (A legal right or a judgement is no good unless it can be enforced). The supply of coercive force is, in this period, probably a gang of men on horses. You’re not going to use women as your muscle, because only exceptional women have enough muscle. So to the extent that social rise is possible as a soldier/warrior, women are almost completely excluded de facto from that.

      You’re right that women can control men on horses (and abbesses and some royal women have their own vassi). But I think there’s a question here about ‘civilian’ control of military forces. At one end, you have the modern military in which, when Margaret Thatcher or Bill Clinton says ‘go and fight’, it’s irrelevant whether they’ve ever been soldiers or not. At the other end, the leader of a warband may simply be the toughest fighter of the lot, who stays leader only till one of his followers/rivals overthrows him. I think in the early medieval period we’re more at the warband end, and that any woman who controlled a military force had a potential credibility problem. (The same is true for any boy or any unwarlike man). Why should her men fight for her rather than defecting to someone harder (who might be able to terrorise more people and thus offer richer rewards)? There were powerful independent women who were able to inspire enough fear/respect to wield power, but it was always more difficult for them than for men.

      As for the specific point on office: I would class abbesses as among the subordinate offices of the church. Their power may be pretty impressive compared to some ‘ordinary’ men, but it’s a lot less than abbots. On the secular side, I am less and less convinced that queenship (let alone countess-ship) is an office until relatively late in the Middle Ages (if then). (I will try and say more on that if/when I blog on Charlotte Cartwright’s paper at the IHR).

      I’m not denying that in practice there was some leeway about women holding ‘state’ power, but the ideology was of it being male. I don’t think the argument (e.g. by Jo Ann McNamara,‘Canossa and the ungendering of the public man. In Medieval religion: new approaches’, edited by C. H. Berman. New York: Routledge) that there was a big change in the eleventh/twelfth century and noblewomen lost power then is convincing. I think right through the Middle Ages there were the simultaneous beliefs that only men should hold secular office and that, nevertheless, in some particular case it was OK that a woman was a ‘lord’. In other words, I don’t think the practice of secular female power ever became a precedent justifying it in general: each case had to be renegotiated anew.


  3. I would have to agree with you about that last paragraph, certainly, because that’s exactly what I’ve argued was happening when my favourite abbess won her case for jurisdiction over her abbey against her comital brother; in other words it’s a Scheinprozess to emplace her as fiscal ruler in her domains. But she, and others (especially widowed countesses and viscountesses, in my area, it seems) did wield the sort of control of resources that allowed her to retain followers. It may just be that Catalonia is weird, of course, but I think that these woman did hold power in the same way as their husbands or brothers did, albeit probably with a seneschal or similar for the actual war-leadership. That must get easier to do as the courtly culture develops I suppose.

    As for holding office, well, does anyone hold office outside the peak hours of the Carolingian Empire, as opposed I mean to title, honour and patrimony? Is an eleventh-century count or abbot in Tuscany, the south of France or Saxony an `officer’ in some way that his widow or his equivalent in a nunnery then isn’t? To whom? I don’t buy this one. I will agree that the Carolingian concept of officer only encompasses men (though I wonder if Rosamond or Jinty would dismiss abbesses so readily) but I think this is clerical courtier theory more than what actually happens, and we just have that patch in the early ninth century where the two meet confusing things. But if the Carolingian success teaches us anything, surely it’s that there are ways to be a credible leader above and beyond simply having the largest and most unscrupulous warband, and one of them, for example, is speaking for a saint’s cult or even a lay familia, which need not necessarily be gendered.

    Lastly, in what economic formulation other than our current post-Industrial one is production not fundamentally shaped like “surplus being taken from a subsistence peasant”? That characterises ancient, oriental, feudal and tributary modes, so if that’s the definition you’re using it’s no wonder we have a disagreement over terminology! I agree that Chris has shifted his ground over the years, but not to the extent that the differences between these formulations can be elided like that!


    • At this point, I think I need to turn the questions back on you ;-). Firstly, from the patriarchy side, there is the empirical fact that vastly more men than women hold secular power in the Middle Ages. If it’s not due to differences in (supposed) military capability or to the ideology of office, what is it due to? If it is just based on a view that husbands should rule wives, then why don’t we see lots of secular women choosing to be celibate in exchange for wielding power (as clerics did)? If it is due solely to discriminatory inheritance rules (daughters only inherit if they have no brothers) then why aren’t there any countesses before around 900, and why does the Empress Matilda have such problems?

      As for the economic history side, what mode of economic production do you see dominant in the early Middle Ages and what’s your definition of that mode? (You have to have a dominant economic mode, or you’ve abandoned systematic economic analysis and that would make Chris Wickham very, very sad.) And how does your preferred mode affect my argument about the patriarchal household as the basis of production?


  4. Starting at the end, I don’t have to have a dominant economic mode, not for the whole early Middle Ages, because I think it’s apparent that control of production is not located in the same place in all the polities of the Middle Ages. This is one reason why, while I talk the talk, I am no Marxist (and Chris manages to survive the disappointment). Both al-Andalus and Byzantium look like tributary mode to me, in fact some have said that parts of al-Andalus are built in ways that would look like ancient mode and some are outright feudal. When we can get multiple modes in one polity we need some redefinition somewhere. Likewise, do Corbie’s estates and the iorn-working villages in Chur really belong to the same mode? But their rulers answer to the same king, at times at least. So at least one reason I’m quarrelling with you about modes is that I think their shortcuts will not help you make a valid case.

    However, I do find that the maxim of ‘look for the producers’ leads me to look in the right places. I don’t think this will work for you, though. In Marxist terms are peasant households really patriarchal, primarily? Both husband and wife labour, and their labour is partly gendered yes, but in Marxist terms they are both working for the same someone else. So that’s how whatever my preferred mode is, it affects your argument.

    Now the main question. I don’t mean to seem as if I’m saying that military capability and a general ideology that rule is male are not dominant; they certainly are. However, what I was saying is that there are ways through this, for a few women, which are not seen as illegitimate. I realise that some women get clerics hating on them for being too masculine in office (Merovingian princesses can’t all be like the nunnery rebels though) but plenty of men get criticism too, and though it’s not gendered I wonder if that’s not more to do with picking a perceived deficiency to get audience agreement like playground taunting. There are woman who are allowed to play the game, is all, so your statements about ideology need to include these exceptions somehow.

    Are there really no countesses before 900, though? I’d never realised that. Mine start with the bulk documents in about 880 (Guifré the Hairy’s wife) and there’s Gisèle the foundress of Cysoing, who looks like a countess to me in the 850s, though she doesn’t get called that. So this is a semantic argument?


    • (A rather belated response, I’m afraid, and I’m not going to discusses countesses just now. I will try to later). On the mode of production argument, the main distinction I want to draw is between the Carolingian situation and a mode of production that hasn’t got mentioned so far: the slave mode. That mode doesn’t get much mention now, because it’s been shown to be relatively economically unimportant even in the Roman world. But thinking about patriarchies, it is significant, because slavery is about the disconnection of reproduction from the family and the domestic economy.

      Slaves cannot marry: that is the classic Roman (and New World) view. As an owner you can either actively breed them like animals or passively leave them to breed like animals, but you would no more hesitate to break up slave couples or parent-child bonds than you would to sell off a horse or a litter of puppies. And slaves don’t support their families: each individual is fed/clothed/housed by the owner.

      I think that even though the slave mode of production may have disappeared by the second century, some of the ideology and the effects lingered on for much longer. It’s only in the Merovingian period that slaves’ relationships start being referred to as marriages, for example, and bans on separating slave families are first announced in Carolingian capitularies (although the practice clearly continued). David Herlihy also argued that it’s only in the Carolingian period that you start to get comparable households at both ends of the social scale: the difference between a lordly familia and a peasant family is reducing, with fewer directly maintained dependents in a lord’s household (or at least fewer permanent dependents in the household, rather than those present for part of a lifecycle, like warriors).

      So I think it’s the decline of the slave mode (both in practical and ideological terms) that puts the lowest class women back into a domestic economy. At the same time, Christian ideology is stressing the patriarchal rule of all households, regardless of class (whereas the classical ‘economics’ of the household is all written for free citizen-farmers). The result is the new principle of the slave/serf as patriarch.

      As for the difference that the tributary mode and the feudal mode might make to patriarchies, I think that’s about the possibility of civilian aristocracies. It’s only with a tax-collecting state that it’s possible to have a paid army and a civilian aristocracy/ruling class which is also supported by that state via salaries/profits of office etc. Otherwise, you have to have a militarised ruling class, who extract surplus themselves by force. (I think there are tributary modes in which the ruling class remains militarized (Manchu China?), but they’re less common).

      This makes a big difference to patriarchal ideologies. If you have a civilian ruling class, you need a lot more ideology explaining why women can’t be Roman senators, Chinese scholar-bureaucrats, landed gentry etc (because they’re mentally/morally inferior, is the usual answer). If you have a militarized aristocracy, the simple answer is ‘women can’t fight’ and you don’t need much more justification than that for why they aren’t ruling (This lack of ideology means that exceptional women can therefore rule, because there’s nothing specific written down saying they can’t).


      • Not much time myself, either, but let me just say that I think when I say `ancient mode’ I mean the same thing that you mean by `slave mode’. And while I agree that slave status had a long long social tail (Bonnassie would be with you but would put the fundamental shifts later, with fuller Christianization) I also think there were large parts of Europe where that style of agriculture never pertained. Some of them of course are alleged hotbeds of matriarchy, so the distinction might still be valid, but I think we need some allowance for the way ideology flows, especially when transmitted by the Church to areas where the slave mode never ran.

        I am fully swayed by the last paragraph’s distinction, however, that’s really nice and seems to resolve the difficulty I was having with what you were saying above.


Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s