# more on genetics and the historical decline of violence

the good professor harpending who, unlike me, actually knows what he’s talking about when it comes to population genetics, took a mathematical look at my suggestion (guess!) that there might have been enough time over the medieval period for genetic changes in the population to have resulted in the historical decline of violence in nw europe that pinker described in The Better Angels (see also eisner).

prof. harpending concludes that — yeah, sure — there might’ve been enough time (from the 1300s to the modern period) to effect such a genetic change. it would’ve been a bit of a push, but it could’ve happened:

“In the present case we need a response of 1/28 of a standard deviation per generation. Assuming an additive heritability of 0.5 (the true value is probably 0.8 or so from literature on the heritability of aggressive behavior in children) the selective differential must be about 1/14 or .07 standard deviations per generation. In terms of IQ this would correspond to a one point IQ advantage of parents over the population average and in terms of stature parents with a mean stature 0.2 inches greater than the population average. This would occur if the most homicidal 1.5% of the population were to fail to reproduce each generation.”

no, i didn’t understand most of that either.

i do understand that he thinks he went conservative in his calculation (i.e. using an additive heritability [<< two links there] of just 0.5 although he thinks it's probably more like 0.8), so that might mean that his calculation should actually be even more in the hbd-ist’s favor. in any case, he concludes that natural selection against “genes for violence” (or selection for “genes for nonviolence”) could explain the historical decline of violence in nw europe “if the most homicidal 1.5% of the population were to fail to reproduce each generation.” a bit of a push, maybe, but possible. (if they really did fail to reproduce.)

he suggests:

“Justice was famously brutal and harsh in Medieval and Renaissance England so this may not be an entirely meaningless exercise. In this excellent essay Peter Frost suggests that the nearly the same selection against violence occurred in the several centuries before the fall of the Roman Empire, and he provides grisly details of Roman treatment of criminals.”

that is one route to go — have the state simply remove the bad guys out of the gene pool.

i’d like to suggest another route (and this is where i’m going to start sounding like a broken record): that they got rid of clannishness in medieval nw europe.

why should getting rid of clannishness matter? because, for whatever reasons (i think the reasons are connected to inclusive fitness), clannish people are violent. blood feuds, honor killings, general obstreperosity — clannish people are just not peaceful.

why? i think it’s ’cause clannish populations are inbreeders and inbreeding alters the possible inclusive fitness payoffs. if you’re from an inbred group, you don’t have to stick your neck out for two brothers or eight cousins to increase your inclusive fitness. if your group is inbred enough, you might only have to be altruistic (in the biological sense) to just one brother or only four cousins (’cause you share that many more genes with your inbred relatives than individuals in an outbred population would, capiche?).

in an inbred population, violent clannish behaviors — which are just the flip-side of being altruistic towards one’s relatives (i.e. be really un-altruistic towards one’s un-relatives) — would/could quickly be selected for since the inclusive fitness payoffs are greater for each altruistic act. and this is exactly what wade and breden (1981) found: inbreeding can accelerate the selection for altruism genes (see also here).

so, to get rid of violence, you could get rid of clannishness. and to get rid of clannishness, you need to get rid of inbreeding. which is exactly what happened in medieval europe starting in the early part of the period. the roman catholic church, supported by secular authorities, banned cousin and other close marriages beginning in 506 (i think that’s when the first ban on cousin marriage was laid down).

enforcement of the various cousin marriage bans, which ranged from first to sixth cousins depending on what century you’re talking about, wasn’t easy — at least not in the beginning. the church, for instance, didn’t require that a marriage ceremony take place in a church until something like 1000 or 1100, so enforcement by the church in the early middle ages was probably patchy at best. however, there were LOTS of secular laws throughout nw europe banning close marriage, including very much so in anglo-saxon england. just a couple of examples: the law of wihtred from the 690s outlawed cousin marriage — and the punishment for cousin marriage in another anglo-saxon law from sometime the 900s-1000s was slavery for the perpetrators. again, difficult to know how well these laws were enforced; but that there were plenty of such laws indicates that the authorities were keen to do something about all this close marriage.

the law of wihtred is, i think, the earliest anglo-saxon law that i’ve come across which made cousin marriage illegal (at least in the part of england where the law of wihtred applied). so the push against inbreeding in anglo-saxon england started at least as early as 690 a.d. again, it may not have been very effective at that point, but england’s outbreeding project had begun by that point.

lorraine lancaster, still considered the authority on anglo-saxon kinship, concluded that, although its importance was beginning to wane (as indicated by a shift in who would be awarded wergeld in the event of a crime against a person, that person’s kinsmen or their guild), an individual’s extended kindred remained of importance in anglo-saxon/english society well into the 1000s. that suggests to me that “clannishness” was still around in the 1000s in england. feuding was definitely still a regular event.

the situation had changed quite a bit by the 1300s when nuclear families were all the rage and englishmen no longer relied so extensively on their extended families. people were still violent in 1300s england, but of course the shift from clannishness to non-clannishness — i.e. from violence to non-violence — would’ve taken some time. evolution doesn’t happen overnight.
_____

the state’s monopoly on violence and outbreeding don’t have to be mutually exclusive explanations for why there may have been a genetic change in nw europeans leading to a decline in violent behaviors. the answer might be both. like jayman said

“Inbreeding, and hence clannishness, can interfere with this process, because while the State is selecting for less violent people, clan conflict presents a counteracting selective pressure for people who are more violent (and can fight feuds).”

…so in places where inbreeding has not abated or did not abate as early as in england — the arab world/middle east, china (or parts of it anyway — h/t luke!), the highlands of scotland, the auvergne — the state hasn’t managed to quell violence as easily. the combo of outbreeding + an effective state seems to be a winning one. better yet if you don’t need such a very strong state (modern nw europe) and the population is just non-violent naturally.
_____

this is all just a theory, of course — theory with a small “t”. but, as cochran and harpending have said (h/t kiwiguy!):

“Whereas tests of hypotheses ought to be careful and conservative, generation of hypotheses ought to be speculative and free-ranging.”

so there! (^_^)

there ought to be a way of mathematically modelling my suggestion — i.e. that the historical decline of violence in nw europe is at least partially the result of the de-selection (if you can say that) of “genes for violence” due to a reduction in inbreeding — but since i’m pretty much numerically illiterate, i won’t be the one working up those models. i would think, though, that in addition to using the breeder’s equation in the calculation, you’d also want to factor in inbreeding/outbreeding somehow.

(note: comments do not require an email. chinese clan house.)

1. A comment over at the WestHunt blog notes that transportation to the Colonies was a primary mechanism of enforcement in early-modern England. That would lead to selection *for* violence among the colonists, and possibly selection *for* clannishness, as the last pockets of cousin marriage in Britain got cleaned out to Kentucky.

I wonder if there is a genetic propensity towards cousin marriage, or if it’s purely environmental. There is a mechanism where children raised in the same household are rather unlikely to find each other sexually attractive, but people do tend to prefer, to some degree, people who look like them (or their opposite-gender siblings, anyway). However, there’s also the sexual allure of the “exotic” which works against cousin marriage.

2. It is interesting to compare world maps of consanguinity and murder rates (see below). They obviously correlate though no doubt they correlate with other things too, most notably economic development. So which, if any, is the independent variable? I’m inclined to go with hbd chick’s favorite hypothesis: in clannish, inbred societies their tends to be one moral code governing in-group behavior and an entirely opposite one for out-groups. Robbery, blood feuds, rape, homicide, they often go together, with the latter sometimes being punished less severely (hence more tolerated) for out-group offenses if I am not mistaken.

OTH, for murderers not to reproduce assumes catching most of them before they are sires (<18?) and assumes a rate of police efficiency which did not exist before modern police forces were institutionalized. There were no police in London before 18th century, for example, and none in Rome either during the time of Caesar, or so I have read.

http://en.wikipedia.org/wiki/List_of_countries_by_intentional_homicide_rate

http://www.consang.net/index.php/Global_prevalence

3. Incidentally, someone who is proficient in computing correlation coefficients could use the country tables in the two links above to compute an actual number. Race needs to somehow be separated out since it is a confounding variable, as is multi-raciality (US) and multi-ethnicity (Balkans). So it is probably a hopeless computation.

4. One reason penalties for minor offenses were so draconian in early modern England was that so few of the offenders were apprehended: they served as a deterrent by terror.

5. @anthony – “A comment over at the WestHunt blog notes that transportation to the Colonies was a primary mechanism of enforcement in early-modern England. That would lead to selection *for* violence among the colonists, and possibly selection *for* clannishness, as the last pockets of cousin marriage in Britain got cleaned out to Kentucky.”

the ancestors of a lot of the cousin marriers in kentucky emigrated of their own accord, afaik. most of the transportees from britain to the new world were actually sent to new england!

the interesting question is australia — apparently three times as many criminals were shipped there as to the u.s./canada. — and today australia is a pretty peaceful place. there were, of course, way more regular immigrants to australia than criminal “immigrants.” i’d be curious to know if the immigrants (both regular and criminal) to australia came from any particular region(s) of britain as they seem to have done in the u.s. (i thought hackett fischer was planning to follow up Albion’s Seed with a book on australia/new zealand. anybody know if that ever happened?) — because, of course, i think the different regions of britain had different degrees of clannishness: england not so much (except for cornwall and the border regions) — scottish highlands, wales, border regions much more clannish.

of course, circumstances always matter, too. we are creatures of both nature and our environment (i really do know that!). if you take a bunch of clannish criminals out of an environment where there’s lots of competition for resources (old blighty in the 1700s-1800s) and ship them off to a new, practically empty environment with more resources available for one and all (except for the unfortunate aborigines), then the levels of clannish fighting would probably decrease. if there’s nothing to fight over, why fight? thing is — the propensity for violence might still be there should conditions change. (mind you, like i said above, i don’t know if the australian immigrants were clannish in the first place. if they were mostly english and not from the borderlands, then they probably weren’t.)

6. @anthony – “I wonder if there is a genetic propensity towards cousin marriage, or if it’s purely environmental.”

yes, that’s been brought up a few times in the comments here (lord knows where now!). rushton argued that there is assortative mating and that that indicates that, back in the mists of time before people moved around so much hither and thither, we must’ve been attracted to, and mated with, our close-ish relatives in the past.

it would seem to make logical sense to do so. if (if) the prime directive in life is to reproduce as many copies of your own genes as possible, it would make sense to want to pair yours with those of close-ish relatives ’cause then you’ll get offspring with extra copies of your genes. if you were to always mate with total strangers, your genes will be watered down in the lineage eventually (well, that’ll probably happen to all of our genes anyway … eventually). then again, a little distance is good so as to not build up too many deleterious genes (alleles). thus, yes…

“There is a mechanism where children raised in the same household are rather unlikely to find each other sexually attractive, but people do tend to prefer, to some degree, people who look like them (or their opposite-gender siblings, anyway).”

…the westermarck effect.

some populations who practice first cousin marriage are clever about this westermarck thing and make sure that cousins do not get to play with one another as kids (i think maybe the moroccans traditionally, but i’d have to double check that). then they’re not grossed out when they’re made to marry one another as adults. in fact, they are quite attracted to one another.

7. @luke – “It is interesting to compare world maps of consanguinity and murder rates…. Incidentally, someone who is proficient in computing correlation coefficients could use the country tables in the two links above to compute an actual number.”

you must be a mind-reader! i was thinking of doing that very thing. (^_^)

the problem is — or one of the problems is — that, as i think you already know, i don’t think that the consang.net data are all that reliable.

don’t get me wrong! it’s a great set of data in many ways, and i think it’s fantastic that the consang folks have compiled it, but there are a lot of holes in it, i think. like the numbers for china for instance — underestimated probably.

plus, i also think that, ideally, more historic data is needed (if possible) — not just the consanguinity rates for the last 50 years or so. we’re talking about evolution here (see post), after all, so we need a little time depth.

still, it’s pretty much the only data set we’ve got on consanguinity rates, so it’s worth a shot, i guess.

another “problem” is that — well, i think a very strong state is going to reduce violence. if the punishment for murder in saudi arabia is beheading, that might be quite a deterrent! otoh, the average prison term for homicide in england — where there is no capital punishment — is 14 years. how much of a deterrent is that? which begs the question: do they not need as big of a deterrent in england as in saudi arabia? what if saudi arabia had england’s punishment for murder? what would happen to their homicide rates? or vice versa for that matter.

also, as pinker pointed out, some of the homicide rates out there might be kinda fudged. is china telling the whole truth about their homicide rates? is saudi arabia? do we even know what it is for the dpr of congo?

still, i’ll give it a shot anyway — see what i find! (^_^)

8. @luke – “They obviously correlate though no doubt they correlate with other things too, most notably economic development.”

yes. what i keep thinking, though, is that it’s probably not economic development leads to less inbreeding (although it might do that sometimes), but rather that less inbreeding leads to economic development.

or a certain kind of economic development anyway — that based upon independent players, corporations (not family ones), a free market based upon the trust of strangers (imagine how weird that is!).

where’d all that arise in the world? nw europe, mainly england + the netherlands. where’d outbreeding happen the most in the world? well, amongst the bushmen apparently, amongst the semai probably, and amongst nw europeans — especially the english perhaps.

@luke – “I’m inclined to go with hbd chick’s favorite hypothesis….”

me, too! (~_^)

@luke – “OTH, for murderers not to reproduce assumes catching most of them before they are sires (<18?) and assumes a rate of police efficiency which did not exist before modern police forces were institutionalized. There were no police in London before 18th century, for example, and none in Rome either during the time of Caesar, or so I have read."

tricky business catching criminals and getting them out of the gene pool quickly enough. to be efficient, one should really take a more yahweh-ian attitude and eliminate the kids of killers, too, for the sins of their fathers:

“I the LORD thy God am a jealous God, visiting the iniquity of the fathers upon the children unto the third and fourth generation of them that hate me….”

(no, that’s not a policy that i advocate! just sayin’ — it would be more efficient.)

9. @luke – “It is interesting to compare world maps of consanguinity and murder rates….”

what i’d like is a list of feuding rates for different populations around the world. keep an eye out for that for me! (^_^)

10. Just read this, from a new blog recommended by Steve Sailer:

“Blood feuds make evolutionary sense. However, if common law can evolve so that justice is swift and sure, then the obligation can be passed over to the State. England achieved this in the Middle Ages, with an independent judiciary, trial by jury, and a common law which applied to most citizens. As a consequence, violence was far lower than on the Continent.”

The author, a psychologist in London, offers no supporting evidence about police efficiency however when it came to the common people. The Magna Carta was between the nobility and the king and the newly arising merchant class, a tiny percent of the total population. But, hey, I’m not an expert. What do I know?

11. Semi-off-topic: One of the recurrent issues (with me at least) is the extent to which inclusive fitness behavior (nepotism and corruption for example) are purely a result of actual relatedness together with cultural and sociological factors, and to what extent they are further amplified by increases in gene frequencies favoring these types of behavior which are brought about as a result of many generations of inbreeding. I think I have a way to test:

The Chinese are a corrupt, relatively inbred society with a culture that accepts, tolerates, and even celebrates this type of behavior (guanxi). However many Chinese are immigrating to the United States and are integrating into our society. So the question is, are second and third generation Chinese immigrants who are not embedded in Chinese enclaves more corrupt than the general population. My impression is that they are not based on published statistics.

12. @luke – “My impression is that they are not based on published statistics.”

which published statistics (for example)? inquiring minds want to know! (^_^)

13. @luke – “‘Blood feuds make evolutionary sense. However, if common law can evolve so that justice is swift and sure, then the obligation can be passed over to the State. England achieved this in the Middle Ages, with an independent judiciary, trial by jury, and a common law which applied to most citizens. As a consequence, violence was far lower than on the Continent.'”

yeah. interesting. and nicely descriptive. but, like so many “explanations” of how the english/nw europeans made such a success of it, it doesn’t actually explain how or why the english would suddenly latch on to common law and why their society should continue to work and evolve along those lines. it just happened, i guess!

14. LukeLea –

So the question is, are second and third generation Chinese immigrants who are not embedded in Chinese enclaves more corrupt than the general population. My impression is that they are not based on published statistics.

Does living in San Francisco count as “embedded in a Chinese enclave”? Because the political behavior of the Chinese here is rather corrupt.