« The Status Quo in Israel | Main | On Prisons »

August 13, 2007

How Quickly Do Genes Change?

Andrew Sullivan, getting all excited about the theories in Gregory Clarks' A Farewell To Arms, enthuses:

Conservatism has long posited that human nature has no history. But what if it does? What if genetic adapation occurs more swiftly among humans than we once believed? This implies that human nature is actually more plastic than we have long thought - but generationally, not individually. It suggests that different populations may have not just different cultural but different genetic inclinations...These are my wild-eyed inferences from a book I have not yet read.

Sigh. Do we have any evidence that this happens? I know people are excited about Clark's book, but Clark is an economic historian, not a biologist or geneticist. It's one thing to say that human nature changes and evolves -- particularly when the rich reproduce at a faster rate than the poor, which is all Clark shows -- but this genetic stuff should be informed by, you know, some sort of expertise on the question of genetics. Maybe someone could ask the folks at Gene Expression, say?

August 13, 2007 | Permalink


Sounds like learned behavior to me and nothing at all to do with genetics. There is just no time for a genetic mutation to be as pervasive as he would like to believe (rough guess I would say it would take tens to hundreds of thousand years).

Maybe there is some long dormant "Rich Gene" that is activated when a person smells TEH RICH! That makes the person go all gaga and crazy to work hard, save, and be more productive. I would postulate that it works the same way as the "GOP Corruption Gene" in where Republicans in power decide that taking bribes is OK, paying to give a b***job in a rest area is the norm and shredding the Constitution is a civic duty.

Posted by: zAmboni | Aug 13, 2007 3:49:54 PM

Easily Sullivan's most obnoxious tic - far worse than his know-nothing brand of big pharma advocacy - is his excitement at the possibility of reinvigorating eugenic science.

Any time a scientist discusses human nature as both (a) natural in the sense of having genetically coded and knowable content and (b) changeable through evolutionary or technological means, Sullivan links with gusto and offers a insta-ish one-liner about the possibilities for the future and the way "the left" or "liberals" won't listen to Science when it tells them about human nature.

He doesn't actually know the science. He's probably read Pinker and doesn't know that 95% of the field thinks Pinker's wrong, and disastrously so. He still thinks that the Bell Curve was good science, and still defends publishing it as if it were an academic tract, but without any peer review.

Another scientist makes problematic claims that could support a renewed eugenics, and Sullivan is there.

Posted by: DivGuy | Aug 13, 2007 3:51:54 PM

Sounds like more of the Bell Curve that Sullivan likes to taut in his issues with race, economics etc

Posted by: akaison | Aug 13, 2007 3:54:02 PM

Not having read "A Farewell to Alms," I can't comment directly, but an op-ed/review of it in the Financial Times on Friday interpreted the thesis of the book to mean that upperclass values percolated through society as children of the rich filtered down to the lawer classes, not upper class genes. Did Clarks ever make an explicit argument about genetics in the first place?

Posted by: Tyro | Aug 13, 2007 3:57:28 PM

Div beat me to the punch. He doesn't need to be a scientist by the way. He just needs to know what the hell he's talking about. And least anyone imagine this is a left-right discussion. You have people such as queer theorist on the left who will argue that the research involving trying to find the set of genes behind sexual orientation is a social construct.

Posted by: akaison | Aug 13, 2007 3:57:38 PM

You're confusing this book with Hemingway's. Clark's book bids goodbye to alms not arms.

Posted by: eriks | Aug 13, 2007 3:58:17 PM

If I am not mistaken Tyro- and maybeyou are just talking about the book, but in case you aren't- the point is that some seem to be using this as excuse to talk of genetics.

Posted by: akaison | Aug 13, 2007 4:01:09 PM

Conservatism has long posited that human nature has no history.

Not to go off on a tangent but what the hell does this mean? I can't parse this sentence in any way that makes a lick of sense.

Posted by: NonyNony | Aug 13, 2007 4:07:31 PM

No, no, no. It's "Goodbye to Alm That."
Andrew's kind of confused, though.

Posted by: Dan Collins | Aug 13, 2007 4:07:51 PM

You have people such as queer theorist on the left who will argue that the research involving trying to find the set of genes behind sexual orientation is a social construct.

Well, while I think that there's obviously a link between genes and sexuality, I'm very skeptical of linking sexuality to genes. (And I should be clear that I don't think sexuality is a "choice" - that's a stupid idea - but I think there's an excluded middle here, though, in that I'd suggest that genetic proclivities are organized and practiced only through available cultural means.)

I work in ancient history, and it's quite obvious from the texts of Greco-Roman antiquity that most men were, in modern terms, functionally bisexual, while a far smaller number of women were. You can see other arrangements of sexuality in early modern Italy, medieval Japan, and so on. I refuse to accept that those people in those other cultures were either (a) wildly genetically different from contemporary Americans or (b) failing to live up to their "true" sexuality, and the vast majority of Greek men were "really" straight but had lots of sex with men because they were forced to.

I think that historical research on sexuality denaturalizes most powerfully the notion that most people are heterosexual, or that heterosexuality is a natural human thing to be. In fact, in lots of cultures in lots of times, people just weren't heterosexual, didn't think they were heterosexual, wouldn't have even really understood the concept of heterosexuality if it were explained to them. This conclusion also means that it isn't universally true that some smaller percentage of people are always and everywhere homosexual or bisexual - and that is problematic for the movements for rights of sexual minorities.

There's a case for "strategic essentialism" because it's so hard in America to explain that if something isn't genetically determined, that doesn't necessarily mean that it's a "choice" - and the false, stupid notion that sexuality, and thus homosexuality, is a "choice" has served to cause incredible pain to so many people. I understand why it's so important to be able to argue from genetics.

But in the end, I have to (a) stick with the historical evidence and (b) argue that the problem with the argument from genetics is that it makes straight the norm, the typical, the usual, and fully naturally so. The constructivist thesis - and construction is not determination, and in no way disregards the existence of genes - on the other hand, challenges the notion that people are everywhere and always usually straight, and suggests that heterosexuality is a peculiarly modern euro-American way to be, and maybe we can rearrange sexualities and bodies to forms that do less outright harm to people.

Posted by: DivGuy | Aug 13, 2007 4:25:43 PM

What I find interesting about the right's obsession with genetics is that it ignores one of the things that actually separates human beings from other animals - culture.

Culture - that is, the gradual evolution of what anthropologists call "lifeway" from one generation to the next via teaching, learning, and thinking, has essentially made genetic evolution obsolete in human terms.

The whole point of culture from an adaptive standpoint is that it allows an otherwise fragile population (for example, one hundred fifty pound hairless apes) to respond to comparatively fast environmental changes without dying out.

Cultural evolution long ago superseded the much slower and more haphazard genetic evolution in terms of its importance to the survival of the human species. The only way that will ever be reversed, paradoxically, is if there are changes to our environment that outpace cultural evolution.

What that would look like, to give an example, would be if a certain group of homo sapiens discovered that through their lifeway, they were causing extremely rapid warming of the climate, but if the culture happened to be led by people who refused to accept this and blocked the population from doing anything about it for several generations.

If that happened, of course there could be a large die-off that would bring genetic evolutionary pressures back to the fore.

I think we can all agree that's an extraordinarily far-fetched scenario, thank goodness.



Posted by: Ape Man | Aug 13, 2007 5:01:43 PM


none of which has anything to do with the science of the genetics. which is my point and the same i made to the theorists. the point is that your theories about what cultures force peo to think should not trump what the science tells us is the case. i see no difference between your argument and sullivans. both of you would have your social norms trump the science. unless there is an ethical or moral issue, i prefer the reverse

Posted by: akaison | Aug 13, 2007 5:15:05 PM

Conservatism has long posited that human nature has no history.

Not to go off on a tangent but what the hell does this mean?

I assume it means human nature doesn't change.

DivGuy, I'm sure you've been over this before, but it seems to me that there's a difference between what you refer to as functional sexuality and a particular sexual orientation in the current sense. Ancient Greek men, for example, who had nothing against having sex with young men, recognized, and made fun of, men who were only attracted to men, who would be true homosexuals. I think men would have sex with just about anything if it were fashionable, and learn to rhapsodize about it, but it's not quite the same thing as sexual orientation.

I also think heterosexuality is by far the most common sexual orientation through history, possibly even in cases where it might seem otherwise.

This isn't to say whether sexual orientation is genetically based or not. Most experts in the biological side of this feel it's likely based in some mix of factors including genetics, and not including choice.

Posted by: Sanpete | Aug 13, 2007 5:17:00 PM

Sanpete -

You've managed, predictably, to glide past DivGuy's point, which isn't that in Socratic Athens a whole bunch of guys were homosexual or a whole bunch of guys had what we'd consider a bisexual orientation, but rather that our concept of sexual orientations does not map onto that culture. Our way of thinking about and arranging the spectrum of human sexuality is just that: our way, not the only way, and a study of other (past) societies indicates as much. Imagine a citizen of ancient Athens who never had sex with any males, regardless of their age, and stuck entirely with women. While he looks, from our perspective, exactly like a "normal" heterosexual, he isn't; we're just anachronistically projecting our picture of the world (of sexuality, in this case) onto theirs.

Posted by: Quarterican | Aug 13, 2007 5:31:03 PM

"Human nature has no history" is a quote from a Glenn Loury article.

What is means, basically, is that human culture doesn't evolve; only technology does (which I guess according to conservatives exists outside of culture somehow.)

I would hesitate to describe the point further as it is hard to do justice to an idea that you think is wrong, and I think this idea is basically wrong, unless you construe the terms "human nature" so narrowly as to render it essentially meaningless.


Posted by: Ape Man | Aug 13, 2007 5:31:53 PM

I think men would have sex with just about anything if it were fashionable, and learn to rhapsodize about it, but it's not quite the same thing as sexual orientation.

I also think heterosexuality is by far the most common sexual orientation through history, possibly even in cases where it might seem otherwise.

These two statements are contradictory. Either most men are happily functionally bisexual, or they aren't. There is no definition of heterosexual that means anything under which the Greeks were heterosexual.

You still need a false consciousness theory to explain how sexuality can be natural and genetic but also highly culturally variable. You still need to argue that the vast majority of self-described and practicing straight men in America are lying to themselves about their desires for men, or that Greeks were lying to themselves about their true desires for women.

The power of the constructivist thesis is that I don't need to presume false consciousness in situations where absolutely no evidence of such exists.

Posted by: DivGuy | Aug 13, 2007 5:32:20 PM

Oh, and DivGuy, that was one of the best explanations of the problem with the genetic path of argument in the sexual orientation equality struggle I've ever read. I literally could not have said it better myself - in fact I couldn't have said it half that well.

Do additional writings of yours on this subject exist, or are there others exploring this vein that I could take a look at? I'm very interested in these ideas.


Posted by: Ape Man | Aug 13, 2007 5:37:02 PM

I think men would have sex with just about anything if it were fashionable, and learn to rhapsodize about it, but it's not quite the same thing as sexual orientation.

Seems like this is the basic source of disagreement. If sexual orientation has a cultural component, it is the same thing as sexual orientation.

It seems to me that the set of people I desire to have sex with defines my sexual orientation, and that set is probably influenced by cultural factors.


Posted by: Ape Man | Aug 13, 2007 5:41:43 PM


Thanks! Hopefully within the next couple years I'll have some writing on the subject, but the text that has most influenced my thinking on these issues of sexuality and history is probably David Halperin's Saint Foucault. He doesn't explicitly address genetic arguments, but he makes clear how a generally constructivist theory denaturalizes heterosexuality, and how the notion of natural sexual orientations is problematic for the political goals of sexual minorities.

akaison -

You say that I have a theory about what cultures force people to think. What I'm arguing is that we both have a theory of how cultures play a consititutive role in though, but I like mine better. You seem to be disregarding the brute historical facts of Greco-Roman male sexuality. If you accept these facts, and you argue that sexuality is fully genetically determined, then you have to argue that most Greek men were "really" heterosexual, but their culture forced them to have sex with men, and they didn't really want to, and this lack of desire showed up nowhere in the historical record because... hey, look, it's Halley's Comet! False consciousness theory makes a mockery of history, ignoring evidence and forcing all people of all cultures and all times to be essentially the same as modern Americans and Europeans. I do admit that I think that we are constrained by our cultural and historical contexts in terms of what we can think and what we can desire. But your theory of the purely genetic basis of sexuality also implies a theory of social construction, and that theory is badly flawed.

Posted by: DivGuy | Aug 13, 2007 5:53:11 PM

our concept of sexual orientations does not map onto that culture

Not sure how predictable it was, Quarterican, but you've managed to glide past my point, that DivGuy's premises, and yours, don't imply that conclusion. It's entirely possible, and a standard view, to see the ancient Greeks as heterosexually oriented men who were functionally bisexual.

Imagine a citizen of ancient Athens who never had sex with any males, regardless of their age, and stuck entirely with women. While he looks, from our perspective, exactly like a "normal" heterosexual, he isn't; we're just anachronistically projecting our picture of the world (of sexuality, in this case) onto theirs.

Again, that doesn't follow. It's possible, but it's also possible that he was indeed a heterosexual in our sense.

DivGuy, a common intuitive view of sexual orientation is that it's the primary orientation that would manifest in conditions where all options are open. On that view, that when options are constrained most men are amazingly flexible about what they can find sexual satisfaction in makes it more difficult to verify what men's sexual orientation is, if any. But it does distinguish functional sexuality, in the sense of actual sexuality, from sexual orientation, without the need for false consciousness (as far as I can tell, not knowing all you might have in mind). That men don't have actual desires for other men wouldn't imply that they couldn't if constrained to. That's not to say that there isn't some false consciousness in this regard anyway.

One strong reason to suppose that such orientation exists in some cases is that people with sexual orientations that aren't reinforced, and are in fact made exceedingly difficult, still find they have that orientation, even if they conform behaviorally to the cultural norm. I don't think this is due to false consciousness. Possibly only some functionally heterosexual men are also so fixed in orientation, but not as many as think they are.

The ancient Greeks did have some cultural false consciousness in regard to women and their desirability as partners, of course (women are children, etc.), and that was associated with their explanations of their homosexual behavior, but I don't think that's required for men to be functionally bisexual.

I think there are good reasons based in basic biology, including animal studies where culture is less of an issue, to suppose that there are some biological bases for heterosexual orientation, and certainly it has dominated functionally through history.

If sexual orientation has a cultural component, it is the same thing as sexual orientation.

I don't see why that would be, but it wouldn't apply the kind of underlying sexual orientation at issue, in any case. The cultural element could account for some of the variation in functional orientation.

I don't think all of this is as important as many others probably do. I don't think it has implications for how issues regarding homosexuality ought to be handled.

Posted by: Sanpete | Aug 13, 2007 6:55:19 PM

hm. ok, i run gene expression. i'm going to ignore some of the comments above. here is my "short" take

1) greg clark should read d.s falconer's introduction to quantitative genetics. a few of my fiends who he is in touch with are already pushing him in that direction. it would had more formal precision to his argument.

2) the rate of change of a continuous phenotypic character is proportion to two things:

a) the extent of heritablity (the proportion of the population wide variation due to genetic variation)

b) the selection coefficient

height is a continuous character. it seems about 80% heritable in modern populations (where nutrition is less of an issue in generating variation). in other words, about 4/5 of the variance you see in a modern population is due to variance in genes. so, for example, assume that you have a population where the median height is 70 inches (5'10). assume that a genetic dictator selects only individuals who are 75 inches to be the parents of the next generation. the prediction equation would tell you

offspring difference = (heritability) X (selection difference)


4 = 0.8 X 5

in other words, the offspring of the selected population would be 4 inches taller than 5'10. now, note that the parents were 5 inches taller, but, only 80% of the variance was genetic, so the other 20% is non-heritable.

so that's the sort of evolutionary dynamic greg clark is talking about. you take a fuzzy quantitative trait with some underlying genetic component, and start ratcheting selection pressure, and change will happen.

some people here are talking about culture as if it is distinct from genetics. it isn't. here is my article on lactase persistence in eurasia. the cliff notes are this:

about 8,000 years ago all humans were like all mammals, we couldn't digest lactose as adult. today 95% of northern europeans, and large numbers in other populations (see the post) are lactose tolerant. extraction of DNA from europeans around 7,000 years ago doesn't show this gene, so it was probably more recent. the area of the genome around LCT (the locus in question) is the largest block of homogenized DNA in the typical european's sequence. in other words, it is like a selection sledgehammer was taken to the gene pool within the last 10,000 years. and, this was due to cultural change.

Posted by: razib | Aug 13, 2007 7:08:30 PM

How Quickly Do Genes Change?

oh, shorter answer is genes change in proportion to the selection pressure exerted. if a behavioral type has a correlation with particular genes the underlying genetic architecture will shift in direct proportion to the selection & proportion of the phenotype it controls (in other words, if there is a far weaker gene-behavior correlation the strong selection will have a smaller effect because it is less biased).

Posted by: razib | Aug 13, 2007 7:11:40 PM

p.s., one of the genes which controls for light skin in europeans (explains 25-40% of the color diff. between africans and europeans) emerged within the last 10,000 years. the gene (OCA2) which seems to control 3/4 of the eye color difference between blue and brown eyes in europeans also seems to have emerged to prominence in the last 10,000 years. i could go on. clark's thesis isn't theoretically impossible, it all depends on the parameters. i think one would be safe to be cautious about the details, but the general model is valid.

Posted by: razib | Aug 13, 2007 7:15:25 PM

That's a very interesting point, razib. Thanks for bringing it to our attention.

One thing I'd like to clarify is that I'm not trying to assert that culture is completely distinct from genetics - obviously cultural changes can have clear consequences to the genome. The point, rather, is that human culture is adaptive in a much shorter time-scale than is human genetics.


Posted by: Ape Man | Aug 13, 2007 7:26:56 PM

btw, some people might complain that the above examples are physical.

1) selection sees fitness differences. whether it is an outcome of different skin color or a new behavior is irrelevant.

2) whether selection results in evolutionary change is simply a function of the genetic component of variation. behavior genetic studies show many traits which have moderate heritabilities.

here are some behavior genetic morphs of interest:
Monoamine oxidase:
In 2006, a New Zealand researcher, Dr Rod Lea said that a particular variant (or genotype) was over-represented in Māori. This supported earlier studies finding different proportions of variants in different ethnic groups. This is the case for many genetic variants, with 33% White/Non-Hispanic, 61% Asian/Pacific Islanders having the low-activity MAO-A promoter variant.

A specific allele of this gene (known as the 'DRD4 long' variant) has been loosely linked to a susceptibility for developing ADHD [1] and a variety of other psychological traits and disorders, like autism.

the frequency of variants of DRD4 and MAO vary between populations.

of course, this doesn't mean that i accept clark's specific model. but i wouldn't be surprised if something like what he is talking about exists. we know that people from high density cultures (think eurasian hubs) are pathogenically deadly to many from low density cultures (andaman islanders have had to be separated from mainland indian settlers for their own good). i wouldn't be surprised if high density culture results in different behavioral morphs coming to prominence (10% of central asians are direct male line descendants of genghis khan, this sort of prolific reproduction wouldn't be possible in hunter-gatherer societies obviously).

Posted by: razib | Aug 13, 2007 7:30:23 PM

The comments to this entry are closed.