one of the preeminent historians of medieval france was georges duby whose work was primarily focused on feudalism, but he also wrote quite a bit on medieval french family structures as well. his main research area was the mâconnais district of burgundy in central france, but he also dealt with other regions of france including the northeast which at one time was part of what was known as austrasia (see also here).

duby’s major finding related to the medieval french family was that, around ca. 1000, there was a titantic shift in family structures in northeastern and central france (and possibly other areas — i’m not sure) from kindreds to lineages, at least amongst the aristocracy, although obviously at some point the commoners followed suit — there are no kindreds in france today. here’s what he had to say about it in The Chivalrous Society [pgs. 146-147]:

“I want to conclude by drawing attention to a point which seems to me essential and by formulating in this connection a hypothesis for research. In this part of western Europe the genealogical recollections of men living at the end of the twelfth century seem, indeed, to reach back according to the rank which they held. At the level of the smaller knights, it goes back towards the mid-eleventh century, in castellan families to the region of the year 1000, in the families of counts as far as the beginning of the tenth century. These thresholds, beyond which the ancestral remembrance was lost, were the more remote the higher placed was the lineage in the political and social hierarchy. This need not surprise us. But it is interesting to observe that the three chronological points appear to be exactly those reached by the researches of present-day scholar trying to recontruct the real blood relationships of families. Moreover, researches cannot reach any point earlier than these. Thus in the society of the Mâconnais, I have been able to uncover kinships in the lineages of knights up to the first half of the eleventh century, the lineages of castellans to the end of the tenth century, and the lineages of counts down to about 920. Beyond these dates I have found it impossible to discover who was the father of the earliest known ancestor. The obstacle is not in the documentation which changes neither in nature nor quantity. We might therefore think this obstacle … resulted from the transformation of the very structure of kinship. Indications of patrilineal blood relationshps disappear from written sources at the very point at which research, going back in time, steps across these chronological thresholds. This reflects a lessening in the importance of these blood relationship in the family consciousness at these dates. In the documents at our disposal it appears as if, at different levels in the aristocracy, the kinship structure was gradually transformed between the beginning of the tenth century and the mid-eleventh century. Before those dates there was no lineage, nor awareness of genealogy properly speaking, and no coherent remembrance of ancestors. A member of the aristocracy considered his family, if I may use the phrase, a horizontal group, spread out in the present, with no precise or fixed limits…. At a later date an individual felt himself, on the contrary, to be part of a family group with a much more rigid structure, centred on agnatic consanguinity and its vertical links.

duby put this shift down to the effects of feudalism (and the related rise of primogeniture which, duby says, was connected to the changing agricultural production methods [see mitterauer]), and i’m sure he’s correct, but i also (of course) think that this shift was connected to changing mating patterns. feudalism can’t be the entire answer since, for example, the early medieval irish had a sort of feudalism — they had a fief system anyway (see Cattle Lords & Clansmen: The Social Structure of Early Ireland) — but unlike the burgundians, the irish hung on to their extended families/clans until very late (into the modern period). where the early irish differed from the burgundians and other germanic populations was that they 1) didn’t have manorialism (until much later when the normans partially introduced it) and 2) they kept on marrying close cousins right through the medieval period.

historians are in agreement that the earlier germanic populations — the franks and the visigoths, etc. — married close cousins to some degree or another in late antiquity/the early medieval period — enough that, for whatever reasons, the roman catholic church and tptb bothered to ban the practice/pass laws against it specifically beginning in the early medieval period. i don’t know whether or not the early medieval lex burgundionum had any regulations regarding cousin marriage, but the burgundians do seem to have converted to roman catholicism (from arianism) by about the year 500, so, like the franks, they may have been some of the earliest of the north europeans to start enjoying the church’s cousin marriage bans (not that the bans were necessarily well-enforced at this early point in time, but the push against cousin marriage had begun by then).

and don’t forget that along with this shift from kindreds to lineages, there was also a shift towards nuclear families.

i think that the broadening of the mating patterns in medieval france and other areas of nw europe (i.e. from close relatives to more distant ones, or even to unrelated individuals) resulted in the shrinking of the family structures (i.e. from broad kindreds to narrower lineages and nuclear families).

here is a little more on duby’s findings from frances and joseph gies’ Marriage and the Family in the Middle Ages [pgs. 124-26, 129 – kindle edition]:

“Around the millennium, by a mechanism that is not well understood, a profound change took place in family dynamics….

“The most significant discernible element in the change was a shift from partible to impartible inheritance. Among the minor nobility in the Mâcon region, the *frérèche*, the association of brothers in joint ownership, previously limited to a few families, became the rule. One son, not necessarily the eldest, was designated to succeed the father in managing the family estates and representing the family in the outside world. Marriage was restricted to this son and at most one other. Households were large. The typical household of the minor aristocracy of the time, as described by Duby, contained perhaps a dozen family members: parents, one brother with his wife and children, and brothers and sisters who remained unmarried, with some of the unmarried brothers often groomed to follow in the footsteps of an uncle who was a church official. The young men lived under the control of their parents, or, when the parents died, of the brother who became head of the family. The share of each in the enterprise was modest, but together they could afford to equip and maintain one or two of the brothers as knights.

“At the top of the hierarchy, and moving steadily down the social ladder in the eleventh century, a different form of impartible inheritance made its appearance, the succession of a single son, usually the eldest: primogeniture….

“The change in the shape of the family was signaled by an element that made its historic first appearance in the documents of the time: the surname or patronymic, passed down in the paternal line. This development was entirely original, bearing little resemblance either to the complex Roman system of nomenclature or to the naming system of the early Middle Ages, in which the individual was designated only by a first name chosen from a short family list….

“Deeds recorded in the Mâcon region before the year 1000 list no family surnames. In the next thirty-five years a few surnames appear, the number increasing throughout the eleventh century….

The progress of the family revolution varied from region to region with the political and economic situation. Local studies by different scholars disagree as to when it principally occurred, from the late ninth to the eleventh century. But an overwhelming consensus exists that sometime within this period a radical change took place in the structure and self-perception of noble families. Previously the fluid horizontal kindred was grouped around a member who held royal office. It practiced partible inheritance and gave equal weight to maternal and paternal forebears. It identified itself merely by distinctive family first names. Now the family assumed a vertical dimension, firmly seated on an estate, a patrimony which descended from father to one son and which gave the family its new, unique surname.”

i think this shift from kindreds to lineages (and nuclear families) in burgundy — and further to the northeast in france, too, if i understand it correctly — is connected to the shifting mating patterns in this part of europe over the course of the medieval period.

kindreds and clans also disappeared from other parts of northwest europe to be replaced by nuclear families, but on a different timeline than central/northeastern france and on different trajectories, the latter thanks to differing economic/agricultural systems:

– independent nuclear families were well in place by the early 1300s in the east midlands in england. the anglo-saxons in england converted to christianity slightly later than the franks/burgundians, so they would’ve headed down the outbreeding road later than those groups. (the franks were even enforcing spiritual kinship marriage bans, i.e. kinship that came about via baptismal relationships between an individual and his godparents, by the 750s, so i’m sure they were concerned about actual relatedness, too, at that point — again, probably mainly amongst the aristocracy.)

east anglia and (eastern?) kent had joint families (not, i imagine, unlike the *frérèche* of pre-1000 burgundy) in the 1300s, but nuclear families by sometime in the 1500-1800s. mating patterns may have remained close for longer in east anglia since it was a remote, swampy area — like frisia and dithmarsian, both areas which displayed strong “clannishness” until comparatively late — but i don’t know that for certain. i need to check on that.

– anglo-saxon/briton populations further away from southeastern/central england seem to have had strong extended familiy/”clannish” connections (even though they may have lived in nuclear family units) until much later, for instance into the 1600s. it may be that, because they were both 1) farther removed from southern areas of anglo-saxon-dominated england where cousin marriage bans were in place from comparatively early on (compared to, say, highland scotland or ireland anyway), and 2) living in upland areas (mountaineers tend to marry closely), these border populations practiced close cousin marriage for longer than other areas of england (they certainly seem to have done so up in cumbria). again, i need to find this out for sure.

– the irish barely gave up their extended families/clans even into the 1700-1800s. they seem to have continued to mate very closely up to at least the 1500s.

furthermore, i think that much of what we see in the reformation and the radical reformation is a set of reactions by northern europeans who were becoming more and more outbred over time and, so, more individualistic and more universalistic behaviors and sentiments were being selected for in these populations. but northern european populations were all over the place in terms of the timing and extent of that outbreeding and the trajectories that their family structures were on. these changes to family and social structures were probably all over and done with in northeastern/central france — and likely parts of the low countries — by the time of the reformation in europe, because, as we saw above, these processes had already begun in these areas by the eleventh century — because they had converted to christianity earlier than other north european groups AND because this is the area of europe where manorialism began.

footnote: interestingly, in modern times burgundy is one of the regions of france with some of the lowest cousin marriage rates.

previously: medieval germanic kindreds…and the ditmarsians and what about the franks?

(note: comments do not require an email. burgundy.)