in jack goody’s The Development of the Family and Marriage in Europe, i came across two examples of hajnal mating patterns (i.e. comparatively high rates of late- and/or never-marriage) which occurred historically outside the hajnal line [pgs. 8-9]:

“The notion of the uniqueness of a late marriage for women and of frequent celibacy for both sexes may require some modification in view of the evidence from twentieth-century Tibet and from Roman Egypt, even if this is less substantial than one would like. In Egypt Hopkins writes of a ten-year post-pubertal delay for women (1980: 333) while in a survey this century among the Khams of eastern Tibet, there were numerous unmarried women and nearly 40 per cent of households had no married couple (Carrasco 1959: 69).”

goody suggests that these examples might (might) refute some researchers’ suggestions that the late-/no-marriage pattern of western europe somehow explains western europe and capitalism and all that, although goody acknowledges that further evidence would, of course, be needed [pg. 9]:

“[I]t [these examples from tibet and roman egypt] would tend to reduce the claims that this demographic regime is linked by a causal nexus with the rise of the West, that is, of Western Europe.

“While Hajnal suggested that these patterns emerged in the late sixteenth century and were possibly to be linked with the development of capitalism and Protestantism, other writers have seen these same features as present in a yet earlier period, but characterising the north-west rather than the whole of Western Europe.1 Some take the view that England was unique in these and other important respects, and Macfarlane has recently seen this singularity as including the presence of a strongly ‘individualistic’ streak, which he tentatively derives from its roots in the German woods (1978: 206) [i partly disagree w/the german woods part-h.chick]. Those who find these features present before the sixteenth century see them as predisposing factors in the rise of capitalism.”

and footnote 1:

“Hajnal himself thought that medieval villagers did not follow a ‘European marriage pattern’; Razi has given support to this idea, finding that in the pre-plague period in the village of Halesowen in the West Midlands of England, marriages took place between the ages of 18 and 22 (1980: 63; also Dyer 1980: 234); however the basis of the calculations has been criticised by Smith (1979: 112), who, like Macfarlane, leans towards the view that the late marriage of women is early and English. See also Smith’s valuable comments (1981) on Herlihy and Klapisch-Zuber (1978).”

i don’t want to get into a discussion about the marriage patterns in medieval halesowen or the ones that dyer discusses in worcester right now, but i do want to point out that both of these places are located in the west midlands and, so, are quite possibly on the edge of the core area of outbreeding/manorialism in europe/england and perhaps, therefore, hajnal’s late marriage arrived in these areas much later. dunno. i’ll come back to this some other time.

back to the tibetan and roman egyptian examples of hajnal mating patterns…

there are two reasons — well, one set of unanswered questions and one known reason — that neither of these examples is comparable to what happened in northwest europe:

1) we don’t have any idea for how long late-/never-marriage was present in either tibet or roman egypt. for tibet we have only a twentieth century survey revealing late-/never-marriage (close to 40% of kham households in eastern tibet had no married couple at all in 1949 – pg. 145) — for roman egypt we have some info about late-marriage (ten year post-pubertal for women – pg. 8), but i don’t know for how long (don’t have access to the paper). but since we’re talking about evolutionary processes, we do need some amount of time for anything to happen. in northwest europe, late-/never-marriage is at least a four hundred year old practice (altho, imho, it’s the coterminous outbreeding that’s really the key here, not the late-marriage — not if you want to explain the rise of capitalism and such things). if, for instance, late-/never-marriage was new to tibet in the twentieth century, well — that’s not going to make a whole lot of difference yet. also, wrt the roman egyptian example, the late-marriage probably only applied to a small subgroup of that society — see point 2 for more on this.

2) both of these societies — tibet and roman egypt — had, or had up until fairly recently, practiced close marriage. the roman egyptians — who, btw, were actually greeks in roman egypt — married, as everyone has heard, their siblings. (but contrary to what you might have heard, brother-sister marriage was not ever common in pharaonic or roman egypt. yes, the pharaohs practiced sibling marriage but probably not the general populace, and the historic records we have for sibling marriage in roman egypt are accounts of greeks who had settled in the kingdom who, for various reasons, mostly to do with maintaining their class status, did not want to marry in with the locals. i keep meaning to do a post on this, and i just haven’t gotten around to it yet.) wrt the kham people in tibet, i don’t know about them specifically, but in general it’s my understanding that tibetan peoples today generally avoid marriage between paternal relatives out to the seventh generation and also avoid marriage between maternal relatives out to the third generation. however, per ippolito desideri, first cousin maternal cousin marriage was common in tibet as recently as the early 1700s [pg. 192], which would fit with the general pattern of marriage in east asia (i.e. with maternal cousins, usually mother’s brother’s daughter marriage). the question is, when did the tibetans abandon this first cousin marriage? it’s sometime within the last three hundred years anyway. (btw, tibetan groups in india still regularly practice maternal first cousin marriage.)

[edit 03/26: but see slng.uls’ comment below and my response to it. thanks, slng.uls!]

so, while these are two interesting examples of hajnal mating patterns occurring outside the hajnal line, they’re really not comparable to what happened in northwest europe. the case in roman egypt really isn’t comparable since there we’re talking about a small subgroup of the population — their mating patterns would hardly have affected the larger society. and in the case of tibet, we have pretty recent cousin marriage — as recent as what probably happened in peripheral places in europe like ireland — which is to be found outside the hajnal line.

previously: big summary post on the hajnal line

(note: comments do not require an email. ippolito desideri.)