Comment by razib on Open Thread, September 1-15, 2012 · 2012-09-07T08:33:19.177Z · score: 1 (5 votes) · LW · GW

if LW gave me dictatorial powers i would have nuked this sub-thread a long time ago, and saved a lot of people productive time they could have devoted to more edifying intellectual pursuits.

also, as a moderate diss, i don't delve deep into LW comments much anymore. but some of these remind now me of usenet in the 1990s. what i appreciate about the 'rationality' community in berkeley is that these are people who are interested in being smart, not seeming smart.

Comment by razib on Open Thread, September 1-15, 2012 · 2012-09-06T19:44:58.686Z · score: 1 (3 votes) · LW · GW

i wasn't expecting much from that thread. i was more curious about the rationale of the atheism+ proponents. i got confirmation of what i feared....

Comment by razib on Open Thread, September 1-15, 2012 · 2012-09-06T18:38:56.752Z · score: 4 (10 votes) · LW · GW

the explanation is banal. 10 hour days at my "day job" + i sleep 6 hours + and have a daughter. not much on the margin. i devote way more time to moderation of comments than a typical blogger as it is, so it shows when i cut back.

Fallout from Razib banning or driving away quality commenters has reached even here.

i don't see what that has to do with anything. LW people say stupid things all the time.

addendum: i don't have much experience on this forum, but i am friends with people associated with the berkeley/bay area LW group. as i said, LW people say stupid things all the time. but, LW people tend to not take it personally when you explain that they're being ignorant outside of domain, which is great. so my last comment wasn't really meant as negatively as it might have seemed. but the back & forth that i have/had with the LW set does not translate well onto my blog, where there is usually a domain-knowledge asymmetry (i'm pretty good at guessing the identity of commenters who know more than me, and usually excuse those from aggressive moderation, because i wouldn't know what to moderate).

Comment by razib on [SEQ RERUN] Zut Allais! · 2011-12-27T20:38:17.658Z · score: 4 (6 votes) · LW · GW

Aside from lactose tolerance (or more accurately, lactase persistence, as the "wild type" is "intolerance"), there are differences in enzyme quantities in saliva due to copy number variations between those populations which have a history of consuming carbohydrates and those which do not. There are also the various resistances to malaria. For multiple reasons, including history (e.g., malaria seems to have become endemic in the Mediterranean over the course of the Roman Empire), we know these are all new, anywhere from 6,000 to 500 years before the present. I can give other examples, but these are the most clear and distinct in the literature.

Comment by razib on BHTV: Yudkowsky / Wilkinson · 2009-01-26T05:20:13.000Z · score: 0 (0 votes) · LW · GW

the video adds a lot if it is kerry and not will hosting free will. just sayin'.... ;-)

Comment by razib on Brief Break · 2008-08-31T18:29:25.000Z · score: 0 (0 votes) · LW · GW

Hey, at least I got through over a solid year of posts without taking a vacation.

more a man than most!

Comment by razib on The Psychological Unity of Humankind · 2008-06-24T09:04:30.000Z · score: 7 (9 votes) · LW · GW


hm. there's a lot to this...but let's just say that when it comes to psychology the residual of non-universality seems really important to focus on in the area of individual differences. specifically, it seems entirely possible that frequency dependent personality morphs abound; e.g., all the stuff about drd4 that i've been posting on my weblog. of course, this is single-locus, but i think it's just the locus of biggest effect. i suspect that what evolutionary psychology misses is that after you account for the substrate of human universals you've got personality morphs which are playing "games" with each other, and in particular there are a host of low frequency morphs running around. these morphs are "complex" to my way of thinking, though perhaps not complex in the way you're implying (i know the evo psych argument against a lot of variation on traits). i happen to think epistasis might also be pretty significant in the transient.

second, i'm also getting curious about variation in traits we perceive as on-off, where those who are "off" are purely pathological. e.g., it turns out that 2% of the population might be "face blind" but have been cryptic because they develop techniques to mask this problem and don't talk about it. i don't think it is just a 2% vs. 98%, i think there are a few other steps "in between", though there's a skewness toward the "normal" facial recognition ability side. on the basis of this i'm willing to dig a little deeper into "human universals" to see what might crop up on the margins.

Comment by razib on A Failed Just-So Story · 2008-01-06T01:16:03.000Z · score: 1 (1 votes) · LW · GW

mike, for interdemic selection (one form of group selecction) there are models where altruists help the group grow but decrease in frequency over time, but then there is a disturbance where the groups reassort. at this point the altruists are at less an advantage and the groups without altruists go extinct, and the process begins anew as altruist frequencies decrease globally. see d.s. wilson's Unto Others for exposition (or an intro pop genetics book).

Comment by razib on A Failed Just-So Story · 2008-01-05T17:36:20.000Z · score: 2 (2 votes) · LW · GW

There are similar problems with getting Tit-for-Tat strategies to accumulate in repeated ecological Prisoner's Dilemma tests. If there isn't a subgroup population of sufficient size, it never comes to dominate the population.

Sometimes particles tunnel out of very deep potential wells, is all I can say.

lt's separate cultural and genetic dynamics here. on the fine scale level between group genetic variance is simply often weak tea to within group variance, and as eliezer implies 'cheater' strategies are pretty good. in meta-population dynamics when one group exterminates another is it usually total genocide? my reading of the ethnography suggests that usually men kill men and boys. they keep women who are fertile. this means that between group competition results in genetic assimilation of losers via their female lineages. but, it often implies cultural annihilation. the descendants of the sabine women carried their genes, but not their culture (or at least proportionately less of their culture, and no identity as sabines).

how is this relevant to religion? i think it suggests that the case of cultural group selection for religion is viable, but genetic group selection is less so (that is, selection for religiosity on a biological level because of its functional benefits). the conversion of african tribe after tribe to christianity or islam en masse isn't one of genetic replacement by muslims or christians, it is one of flips of group identity.

Comment by razib on A Failed Just-So Story · 2008-01-05T17:29:36.000Z · score: 2 (2 votes) · LW · GW

It's only comparatively recently that religion has been separate from nationhood or tribe, and its central tenets questioned. The fact that it's still around in a weakened state could simply be analogous to, say, our fight or flight mechanism. Or an appendix.

this is due to the fusion of religion with philosophy and the emerge of super-states which required super-religions. the first religious skeptics emerge in the record precisely when religion shifted from being informal implicit paganism toward a more formal structured system. e.g., the seeds of puranic hinduism, buddhism, jainism gave rise to the carvakas. the mystery religions were countered by lucretius and the epicureans. the religious enthusiasm of the mohists and some daoist groups was countered by the philosophical naturalism of xun zi.

It's only group selection when it's religious groups competing against nonreligious groups. It's individual selection when it's individuals in groups competing against loners.

no. david s. wilson connects group selection with functionalism. so the character of a religion counts, and the character of another religion counts. as a case study think of the syncretistic abangan islam of java, and the more orthodox santri islam of java. the latter is rising in frequency in relation to the former. why? rational choice theorists would argue that the latter is a more appealing product, it offers more goods & services. group selectionists or functionalists would assert that santri islam is better suited to the anomie and dislocation which post-village life imposes upon the typical javanese. if, for example, santri mosques served as a better social insurance system than the village institutions of the abangan one can imagine that over time group level effects would aid the santri. wilson would argue that the santri are a more integrated and efficient 'social organism.' rodney stark, a rational choice theorist, has made the same sort of argument for ancient christianity, suggesting that its self-help ideology and normatively enforced altruism & humanism (e.g., no exposure of babies, helping the sick, etc.) resulted in greater natural increase vis-a-vis pagans. finally, do note that cognitive psychologists tend to record no deep level differences in how religionists conceptualize their supernatural agent of choice, so that's the deeper substrate layer.

Comment by razib on A Failed Just-So Story · 2008-01-05T10:12:17.000Z · score: 8 (8 votes) · LW · GW

to be more evolutionarily precise, i'm saying that propensity to religion might be a correlated response of selection for other traits. e.g., agency detection and theory of mind. these are very nifty tools which we humans are pretty strongly hard-wired for. but i think there's a good deal of circumstantial evidence that they breed in us a tendency to imagine spooks all over the place.

Comment by razib on A Failed Just-So Story · 2008-01-05T10:08:53.000Z · score: 3 (3 votes) · LW · GW

people mean different things things by religion. also, for those looking up the arguments, david s. wilson is the major proponent of group selection-functionalism right now. scott atran and pascal boyer are the major proponents of the cognitive byproduct thesis. i tend to favor the latter. not only do i (and they) think it is a byproduct, but i suspect that religion is produced by the functional constraints of modal human psychology, just like a running engine produces heat. you can stop the heat production if it is turning off the engine. also, i don't think that the byproduct thesis precludes some element of functionalism overlain on top of it...but part of the issue here is that i think religion in terms of cognitive operation is a very different entity than religion as a set of beliefs hooked into an institutional framework.

Comment by razib on Fake Morality · 2007-11-08T21:42:55.000Z · score: 1 (1 votes) · LW · GW

God, say the religious fundamentalists, is the source of all morality; there can be no morality without a Judge who rewards and punishes. If we did not fear hell and yearn for heaven, then what would stop people from murdering each other left and right?

many (most officially i believe) believe that we are justified by faith alone and that divine grace comes without our own action. this is why calvinists generally accept predestination. there are some really byzantine logics which allow believers in this to still live lives which accept a de facto element of free will, but the point is that officially most protestants belief that your place in heaven or hell is not contingent upon your moral behavior, but rather your face in the savior.

now, there is a difference between the "official" party line, and how people act and process their ideas. but a rational attempt to discuss this probably needs to start with explicit concepts.

Comment by razib on Beware of Stephen J. Gould · 2007-11-06T06:10:14.000Z · score: 8 (10 votes) · LW · GW

He wrote as if the entire Williams revolution had never occurred! Gould attacked, as if they were still current views, romantic notions that no serious biologist had put forth since the 1960s.

he pulled the same trick with mismeasure of man.

But some readers may have to take my word for this, since the names of eminent scientists are often less well-known to the general public than the names of fast-talking scoundrels such as Uri Geller or Stephen J. Gould.

actually, two things: 1) do a literature citation search.

2) appeal to another authority of some note, e.g., paul krugman saying the exact same thing re: gould & j.m. smith.

Comment by razib on Doublethink (Choosing to be Biased) · 2007-09-16T05:49:56.000Z · score: 0 (0 votes) · LW · GW

This is an ancient and thoroughly discredited idea. See George Williams's "Adaptation and Natural Selection."

i am generally skeptical of group selectionist arguments, but we are probably on the cusp of a renaissance in this area. it will be spearheaded by e.o. wilson, who has always been a "believer," but who now believes that group selection (or at least multi-level selection) has the empirical and analytical firepower to make a comeback. i am cautiously skeptical, but in the interests of honesty i think that "ancient and thoroughly discredited" is probably a better description for group selection circa 1995 than 2007. most evolutionary biologists are probably pretty skeptical of group selectionist arguments, but in large part it is because the models presented (which tend to avoid the pitfalls of the earlier arguments) are hard to test and seem analytically intractable beyond the simplest formulations.

Comment by razib on Doublethink (Choosing to be Biased) · 2007-09-16T05:44:40.000Z · score: 2 (2 votes) · LW · GW

Imagine a gene that caused 9/10 of the humans who have it to be twice as fertility and attractiveness as the population that did not have it, while 1/10 of the humans who have it can't reproduce at all.

this is means that the allele (genetic variant) increase fitness by a factor of 1.8. this is not a "species level" benefit in anything but a tautological way. higher levels of selection or dynamic processes are only interesting if they can not be reduced down to a lower level. e.g., you can increase the fitness of the group by simply increasing the fitness of individuals which compose the group. this increases the fitness of the group, but it is easily reduced toward increasing the fitness of individuals. in other cases you can not decompose the group fitness to individuals and so there is grounds for saying that the excess fitness which is gained by having a group, or evaluating a group, is something that is "for the good of the group."

to use a sports analogy, if you brought together an all-star team you'd get a better team, not because of the team dynamics but because the individual players are so much better. in contrast, ther are teams which are very good because of group dynamics where utility players can specialize in their roles and synergistically perform far better than they might as individuals.