New blogs

Leherensuge was replaced in October 2010 by two new blogs: For what they were... we are and For what we are... they will be. Check them out.

Wednesday, October 1, 2008

Polygynic bias in reproduction... but how?


The latest PLoS Genetics hit is
Sex-Biased Evolutionary Forces Shape Genomic Patterns of Human Diversity, by M.F. Hammer et al.

We already knew, or at least believed, that the effective population of males has generally been lower than that of females. It seems obvious considering that in principle any fertile man has the potential to engender thousands of children, provided he finds the sufficient number of mothers, while women can only have a limited number of children each (I think the highest single figure I have ever heard of is something like thirty - and one shivers at the idea of giving birth so many times, plus breastfeeding and rearing all them). Much lower figures are more normal of course but the difference of reproductive potential between sexes remains and can be very high.

But we were not certain on how that worked in reality through the generations. We typically find more mtDNA than Y-DNA diversity but these items may not be fully comparable anyhow. This paper adresses this issue by comparing the diversity at the X chromosome and at autosomal DNA. If the effective population (the people effectively reproducing) would be the same for both genders, we should find that the X-DNA diversity should be 75% of the autosomal one. What Hammer and colleagues have found is that it is not the case that X-DNA diversity is generally higher than under neutral expectations and in some cases even higher than autosomal diversity itself.

That basically means that some men have been more succesful than others and that the average woman as well. Expected indeed, as the difference in biological potential is certainly there.

But the confusion comes with the use of the term polygyny (one man married or otherwise reproducing, like in concubinage, with several women). While the conclusions are more imprecise, the term arises in the abstract: our results point to a systematic difference between the sexes in the variance in reproductive success; namely, the widespread effects of polygyny in human populations. The term is adressed at times confusingly, as normally this means formal poygyny and not mere infidelity. Both Dienekes and Kambiz emphasize the term polygyny suggesting more or less explictly that it has been a formal feature of human societies. True that part of the fault may lie on Hammer's shoulders, quoted at New Scientist saying: I don't know how long monogamy has been with us. It seems it hasn't been around long, evolutionarily.

This is far from clear, I think. Even without formal polygamy, there has always been opportunities for some men to be more succesful than average in this aspect.

The most important and revealing graph in that paper is figure 2, where the six sampled populations are shown with their respective ratios of X-DNA diversity in contrast with a hypothetical neutral 0.75 line. With the possible exception of the San (Bushmen), all clearly exceed the neutral model. When we take the point extimate (ignoring the possible error) they appear like this:

- San (Bushmen): 0.85 (13% over neutral expectation)
- Biaka Pygmies: 0.90 (20% excess)
- Mandenka (West Africans): 0.93 (24% excess)
- Han Chinese: 0.94 (25% excess)
- Melanesians: 1.04 (39% excess)
- Northern Basques: 1.05 (40% excess)

The first interesting thing is that the hunter-gatherer peoples are the ones showing lower deviation from the neutral expectation, i.e. the ones where polygyny has been less important overall. The difference between the San and the Biaka is significative but I wonder if it can reflect Bantu influences among the latter.

Excluding hunter-gatherers, there are two (geografically meaningless) groups: Han and Mandenka appear moderately polygynistic, while Melanesians and Basques appear rather extreme. It would be interesting to contrast these peoples with their culturally akin neighbours, for instance compare Chinese with Japanese, Mongols or Thais... or Basques with other Europeans. But we will have to wait for that, I guess.

Anyhow, it is the "highly polygynistic" group the most intriguing set. Melanesians (at least Papuans, I could not find which ethnicity they are specifically) are known for actually practicing polygyny with normality: they are a highly patriarchal Neolithic society. But Basques are a totally different case: there is no record of such institution ever existing at all and the monogamic (patriarchal maybe but softly so in the European context) household is a national institution. So I have been chewing on the issue for several days now because it really stroke me as very odd.

I cannot know how was Basque society in Prehistory (and that, with the exception of some brief accounts, lasts until the late Middle Ages) but what we know does not point in that institutional polygynistic direction at all. Wilhelm von Humboldt, who visited the country in the early 19th century, reported instead that Basques had then frequent pre-marital relations and that marriage was normally formalized when the girl got pregnant (possibly but not necesarily from the groom, he implied). From Inquisitorial accounts we also know that Basques from both sides of the mountains used to engage on pagan orgiastic celebrations (community sex and drug parties probably with religious meaning) as late as the 16th century. We also know that the status of women was rather high and that they enjoyed nearly as much freedom as men (and were specially important as leaders of the old religion: "witches"). Earlier reports, even if way too limited, do not contradict this picture in any case.

Informal polygamy (both sexes) was frequent among Basques in the past and was socially accepted, in other words. But patrilineal monogamic family was still the basic cell and that is very clear in the central importance of the undivisible household in Basque law.

So I guess that such a high X-DNA excess diversity among Northern Basques, rather than formal polygyny (unattested, quite unlikely to have ever existed in my opinion) what reflects is informal polygamy or promiscuity and the implicit ability of women to choose the father of their children to some extent in this context, moreso if some anticonceptives were known, what is probable considering that witches (sorginak for their Basque name) were primarily herborists.

That way the more desirable men would get some advantage... but not because they wielded particular power over women, as in typical polygyny, but because women would tend to choose them with preference over the rest as the biological fathers, that, as it is well known, is not necesarily the same as the legal father.

The child born inside marriage is son of the husband, Napoleon dixit. Biology shrugs, cheats and laughs last.

...

Three months later... Natura publishes something behind a paywall (authored by Keinan et al.) that finds exactly the opposite results: that in Africa the X-DNA to A-DNA ratio is c. 0.75, while in Eurasia is c. 0.65, much lower. They scratch their heads wondering what may have caused a loss of female efefctive population, what makes little sense (hey, why no massive polyandry?, it's as logical - or not - as massive polygyny in the previous one).

I don't think I can handle such a variety of contradictory results and make any sense. But, as Manjunat said, the post needed an update.

22 comments:

Manju Edangam said...

I wonder if loss of men during battles has been considered as one of the variables here. I'm sure there would have been heavy loss of men compared to women in tribal warfares. In isolated populations like Basques(considering that they maintained their isolation most of the time) the impact may be large. Even this could account of here higher X-DNA diversity. Anyway, the survivers of the warfares need not always be the most capable of men.

Maju said...

Well, what I've read of tribal warfare is that it's relatively harmless: if anyone dies, it's typically the old man or woman who takes more risks and can't run fast enough. In general tribal peoples, specially hunter-gatherers can't afford human loses, so it is likely that they retire when they suffer a casualty or two. People is really important for them, not just because they know each other very well but because it is the group what secures the survival of the individual, more than any resource.

Now, it is possible that warfare was much more violent in the Iron Age (period in which the Celts arrive at the Basque borders and evidence of armed conflict becomes common).

Basques historically have exported many men anyhow (because of the property structure: only one son could inherit the household) but I find hard to believe that in those historical times there were many unmarried mothers. The rate is comparable to actually poygynistic Melanesians, so it's not like a few exceptions could make this pattern: it must have been something systemic, and I suspect that tolerance towards promiscuity (what allowed women to choose the fathers, at least more than otherwise) could have been the main reason.

The alternative would be that Paleolithic Basques would have been polygynistic, but I know of no data that could support this hypohesis.

Manju Edangam said...

Okay. I don't have much idea about casualties in a tribal warfare.

because of the property structure: only one son could inherit the household

Is this common to whole of Europe? I think even among English only the eldest son could inherit the property and rest had to fend for themselves.

Maju said...

Is this common to whole of Europe? I think even among English only the eldest son could inherit the property and rest had to fend for themselves.

I don't know. I know that Spanish common law nearly forces equal division between the heirs and that minifundism is a major problem of Galicia (NW, Portuguese-speaing, Spain) precisely because uncontrlled divison of farmland. But in the past the "mayorazgo" system (elder son inherits all root property) also applied in parts of Spain, at least for aristocratic households (that in many cases were the only ones that could be called that way). But this was surely supressed by the burgueoise reforms in the 19th century.

I have read anyhow that cultural differences in property structure and family style caused some divide between Northern and Southern Europe in the Middle Ages, favoring uncontrolled demographic growth in the south while containing it in the North, at least when there were few economic options to found a family. In this sense the Basque system may appear more like that of Northern Europe probably but I also imagine that the N/S dvide was blurry and patchy in many cases (just a general trend).

Maju said...

As for tribal warfare, some passages from my old anthropology manual (Marvin Harris, 1988) on Northern Australian tribal warfare:

In the morning the two armies (sic) alligned in front of each other. The elder began hostilities yelling insults and accusations against individuals among enemy ranks (...). "So when individuals began throwing spears they did because of reasons based on personal arguments" (...). Precision was all but present, because most of the spears were thrown by the elderly.

It was not rare that it was some non-fighting innocent or some of the vocal crones that dribbled among combatants, yelling obscenities to everybody, and whose reflexes to avoid the spears were not as good as those of men (...) As soon as someone fell injured (...) the fight stopped immediately until both gangs evaluated the consequences of this new developement.

Even if hunter-gatherers seldom attempt to annihilate each other and often abandon the battlefield when one or two casualties have happened, the cummulative effect can be quite important.

And then it comes to ponder that between 10% (!Kung San) and 28% (Northern Australian tribes) of adult men can die in warfare. This does not mean that they would die childless anyhow, probably the greatest damage is the loss of valuable and experienced workforce but most fallen men would already have children around at the time of their death.

You may have a point anyhow, but I doubt it is the only cause. There must be some "effective poligyny" (not necesarily formal one, illegitimate children also count) as main cause, I think. Melanesians (I understand Papuans, though the paper is unclear) are both formally poligynistic and warlike, and they may serve as example for the highest rates of gender bias.

Manju Edangam said...

Thanks for the inputs. By the way, I didn't say warfare is the only cause. I was wondering whether it was considered as one of the variables(along with say polygyny). I mean your numbers 10-28% are not small. Even if these people had children the effective survival rate would be different from the men who survived and procreated more. Again this situation may also lead to polygyny. Maybe polygyny feature itself may have different reasons.

Maju said...

Well, 10% is for Bushmen, who are also the ones showing the lowest bias (13%). If warfare is the cause, it would be almost in 1:1 relation.

But consider please that there are many other ways of dying young, specially in hunter-gatherer conditions: illness, accidents, inccidents with a large animals... Additionally women have a greater risk of dying young because of birthgiving complications, what probably compensates for greater male deaths in violent circumstances.

I still think that the main factor is some sort of active poligyny, be it formal (as in patriachal "harems") or infomal (promiscuity and illegitimate children). Promiscuity is universal, while formal poligyny is less common instead and probably only exists in relatively developed societies of at least agricultural level.

Manju Edangam said...

Additionally women have a greater risk of dying young because of birthgiving complications, what probably compensates for greater male deaths in violent circumstances.

Agreed, that certainly can be the case.

Anonymous said...

Thanks for presenting the data from Hammer et al in the most accessible way I have seen.
I wonder if you could help me out with I a few questions, if you don't know the answer nobody will.

1)When a Northern Basque woman gets married does she go to live in the Husbands community?

2)If so, is this true usually, or almost always?

3) How common is it for Northern Basque women to marry someone from another town or even district?

Maju said...

I can talk of Basques in general (and I don't think there is any peculiarity in the North in this regard). The house was normally the man's property and inherited patrilinearly - so, yes the woman would go to live to the husband's house by default. Most often couples would be from the same or nearby communities anyhow, people who new each other maybe from childhood. Of course this has changed in the last century plus, as mobility has increased in general and farming has become less common.

Anonymous said...

Thanks for your time and the quick response to my questions, they were related to patrolocality and how its impact in increasing X-chromosome diversity could have been mistaken for evidence of polyginy of course.
You seem doubtful that formal polyginy was ever as widespread among the Basques as it was among the Melanesians, this gives me confidence that there is something wrong with this study.

I should have read the post more carefully because it pointed to what I was looking for: a practice unusually common among the Basque people.
The central importance of the indivisible household for Basque inheritance patterns you identified in the post seems to be something that could affect family formation. I looked it up and the Basque tradition, according to what I read, was that one, usually the oldest, son would be designated the heir to everything. The others would often tend to emigrate, these emigrants would, presumably, tend to be the sons?

So if things were as I am outlining historically; the sole heir to a farm would have very good marriage prospects and there were an excess of women. If a woman died in chidbirth for example the owner of a farm might easily be able to remarry. Historically was female servants bearing illegitimate children ever common or accepted?

Maju said...

What you mentions makes some sense but I understand that it would not explain the extremely high deviation from neutral expectations. There must be something more, I think.

Also, the non-heir sons did not necesarily emigrate: they looked for fortune in other business than farming, sometimes abroad (America specially, but that only in Modern Age) but at home as well. When possible, a second farm was founded or acquired, while sailing/fishing/privateering, manufacture (specially metallurgy/weapon-making but also ship-building) and trade were important alternatives. They could also became servants of richer people or priests/monks - and these were also alternatives for women as well. Not everybody emigrated (and not always the farm went to the oldest son: that's variable) and emigrants often came back with some earnings to estabilish a family back at home (in fact often the richest ones were returning emigrants: the "Americans" or "Indians").

In any case, while the household farm (family root property) could not be divided, secondary, possiby smaller, farms could be acquired (or founded) for the remaining sons. Farming could be or not a well-to-do activity also, it was ownership of a house what gave the stability to found a family (and the right to vote, btw).

So if things were as I am outlining historically; the sole heir to a farm would have very good marriage prospects and there were an excess of women. If a woman died in chidbirth for example the owner of a farm might easily be able to remarry.

Guess so - up to a point. Women were free to choose their husbands, so not only wealth would matter, certainly. AFAIK marriages with big age difference were not common, except in the more culturally Castilian areas, like Enkarterriak (Encartaciones). In the typical rural Basque Country, I know of no accounts of such interest marriages, though they may have existed anyhow.

Historically was female servants bearing illegitimate children ever common or accepted?

Not recently: Christian morality penetrated very strongly in Modern Age. Unmarried women, anyhow, would most likely stay at home with their parents or become nuns. Female servants, in the few rich households where they existed, most likely were married on their own, so guess that your explanation is not satisfactory, specially as feudalism was rare or even totally non-existent.

I doubt tha all these normal biases can cause Basques (or any other Europeans - for whom we don't have any data, and would be interesting to compare with) to have a deviation that is higher than, say, the Chinese or the Mandinka. There must be some other reason and I suspect that past tolerance for promiscuity (or even institutionalized one - as in akelarreak/sabbats) may have been an additional reason.

Of course, it would be a lot easier to judge if we had this kind of data for other European populations. In principle I'd expect more bias for mainstream feudalist European nations, where legal abuses as "first night right" (and a-legal ones as everyday abuse and rape by feudal lords) were common. Such privilege structure was almost non-existent among Basques and (considering the apparent reasons behind Basque Medieval survival, as rebel stronghold against late Roman Feudalism), it would seem that have never been an important part of social structure.

So my guess is that, whe data for other European and Mediterranean peoples is available, we may still see a high unusual devioation among Basques. But not because of mere patriarchal bias, which should yield (in absence of formal and widespread polygyny) rates more like the Chinese and Mandinka (both are patriarchal and could historically be polygynic) but because of relatively free promiscuity.

Anonymous said...

Thanks for another fast and comprehensive response. You made some excellent points that I never thought of concerning "first night 'right'" and outright rape by lords in other countries' feudal systems.

Maju said...

Yeah. I don't know which is the actual weight of such things but it's obvious that feudal nobility (and any kind of raider or plunderer) had their (brutal) privileges. Of course power may also be seductive... but, if it's not, it can be imperative anyhow.

Basque society was/is patriarchal but softly so and was much less hierarchical than any neighbours. But we do not have data to compare with French, Castilians, English, Germans or Italians - who were in general much more strongly affected by these matters of hierarchy and patriarchy.

Manju Edangam said...

Do you need to update your post based on this study?

Maju said...

Well, more than an update, I'd write something but as the article is behind a paywall, I don't have access to the data, so hard to uderstand.

Whatever the case, that study basically means that the conclusions of the one I dealt with here should be taken with extreme caution, because the data and results are almost exactly the opposite.

Anonymous said...

Opportunity for Natural Selection in a Basque Population and Its Secular Trend: Evolutionary Implications of Epidemic Mortality (page 10)

The extent to which and the manner in which various segments of a population are affected during population crises are further points to be considered when examining the potential genetic consequences of a high mortality period. The findings of this study indicate that the time when the most severe epidemic mortality was detected (1830-1919) was also the time when the highest values were recorded for both the total selection index (I^sub t^) and the mortality component (I^sub m^). But the evolutionary implications of mortality crises derive basically from the differential repercussions of epidemic mortality on certain age groups. Age at the time of death from infectious disease is important, because only diseases causing death before or during the early years of the reproductive period will have significant evolutionary effects (Motulsky 1989; Lasker and Kaplan 1995).

In this study we found consistent and significant differential incidence of mortality by sex and age group during population crises: Females were more affected than males, and the relative frequency of deaths from the epidemics was significantly higher in the age groups of 1-15 and 16-45 years. With respect to differential mortality by sex, no clear pattern has been reported in the relevant literature, at least in European populations [see Scott et al. (1996)]. Nevertheless, it is not hard to deduce that the selective implications of differential mortality by sex are related to a reduction in the effective population size (N^sub e^), as a result of a marked male-female imbalance in the most fertile age group (16-35 years); overmortality of women during a crisis could give rise to a surplus of men, and the chances of remarriage for widowers would be reduced (Bittles et al. 1986; Mielke and Pitkänen 1989), which evidently can cause an increase in the variance of the family size (offspring number).

Maju said...

It's an interesting paper maybe but will hardly explain anything. The study is limited to a very localized community of Arabako Errioxa (hardly representative of the whole country) and, anyhow:

The four crises in question were of moderate intensity, with absolute annual rates ranging from 65 (1813) to 86 (1849).

With the normal mortality rate being of 43 deaths per year. Even the strongest crisis only doubles this figure, hardly any bottleneck.

Manju Edangam said...

Nature Genetics 41, 66 - 70 (2008)
Published online: 21 December 2008

Accelerated genetic drift on chromosome X during the human dispersal
out of Africa

Alon Keinan, James C Mullikin, Nick Patterson & David Reich

Comparisons of chromosome X and the autosomes can illuminate
differences in the histories of males and females as well as shed
light on the forces of natural selection. We compared the patterns of
variation in these parts of the genome using two datasets that we
assembled for this study that are both genomic in scale. Three
independent analyses show that around the time of the dispersal of
modern humans out of Africa, chromosome X experienced much more
genetic drift than is expected from the pattern on the autosomes.
This is not predicted by known episodes of demographic history, and
we found no similar patterns associated with the dispersals into East
Asia and Europe. We conclude that a sex-biased process that reduced
the female effective population size, or an episode of natural
selection unusually affecting chromosome X, was associated with the
founding of non-African populations.

Department of Genetics, Harvard Medical School, Boston, Massachusetts
02115, USA.
Broad Institute of Harvard and Massachusetts Institute of Technology,
Cambridge, Massachusetts 02142, USA.
Genome Technology Branch, National Human Genome Research Institute,
National Institutes of Health, Bethesda, Maryland 20892, USA.

Link

Does that mean early migration have left no influence on the later stagnant societies outside Africa?

Manju Edangam said...

Of course, the other study (where the results are opposite) could be followed from this study.

Maju said...

It is the same study you mentioned before but in form of press release. It's ironical that the press release is more apparently informative than the paywalled abstract of the original paper but well.

Does that mean early migration have left no influence on the later stagnant societies outside Africa?

Why? The SD article suggests that migrants were, as usual, more men than women, causing some minor but meaningful female lineage bottleneck early on.

Obviously this could only happen if in spite women being less, promiscuity was the rule, and they had children in a non-monogamic fashion. That's an aspect that is not explained to us, maybe because it's like saying that our ancestors were all "bastards" - under typical moral standards (JCM, HBJ or suely even in Vudun).

But well, guess we are all bastards, so to say, yes.

Of course, the other study (where the results are opposite) could be followed from this study.

I think that the matter simply needs more research, much more. Each time the X chromosome comes to light in the Old World, we get puzlling results, even contradictory ones.

That's why I'm not posting on this last paper separately. Because I'm in the "watch and see" mode, rather than in the "eureka" one.

Manju Edangam said...

I definitely need to take a break :-)!