Tuesday, March 06, 2007

Evolution of "Religion"

I'm going to use this post over at Gene Expression as a convenient foil for something that irritates me about the debate over the evolution of religion. Namely, you could just as easily describe the substance of the debate as the evolution of philosophy, of science, or of abstract thinking. Until the very recent past, there hasn't been a difference, and whatever your side is on the culture war, there's little need for this muddle.

The question, for Gene Expression and others, is "Why do people cling to ideas that are manifestly implausible?" It's an odd definition for two reasons, first because most religious people make all sorts of claims that are not merely plausible, but based on observation (think of the various ancient astronomical sights, or nineteenth-century naturalists documenting every variety of flora and fauna in part because they were testimony of God's greatness), and also because "plausibility" is necessarily judged after the fact. Plenty of good ideas at first seemed wholly implausible, and the Newtons of the world frequently interlaced their most lasting discoveries with ideas that now seem totally insane.
The point is, cognitive activity does not easily separate into modern categories of "religion," "economics," and so forth. While I find the debate fascinating, casting it in such terms leads to bad science, to explanations based on social darwinism rather than Darwinism, and to political claims with little connection to the issues at hand.


switters said...


I've been bouncing about a thought or 2 in the old bean about Music Theory. I haven't spelled it out thoroughly enough for a toppost, but I think my recent ruminations are at least tangential to your concerns.

Music Theory is an after-the-fact phenomenon. Oh, sure: Palestrina wrote a text on Counterpoint at about the time just before Bach hit the scene. But even that text was hindsight.

Forms arose out of the analysis of music, in particular Bach's, only because of their consistency. Same with Mozart, Beethoven, Brahms. Symphonies had very specific rules that composers had to follow to the letter, until following them became tedious. So they each in turn bent them, broke them, what have you.

It's easier to talk about rules and the evolution of jazz because the rules in jazz are a lot more straightforward and approachable than they are in so-called serious music. So I may approach it that way.

Suffice it to say that constant evolution, when it comes to most forms of music, is, ultimately, bound by and determined by its inherent consistency.

I don't know if you know where I'm going with that, but, for me, it's both a liberating notion with regard to music as art form, and a terrifying idea with regard to music as state propaganda.

I'll flesh it out more one of these days.

Keifus said...

I'm a little confused by your question. Are you posing it to Gene Expression, or are you saying that GE and others seem to pose it to teh world?

As far as scientific theory goes, it took history a while to get there, but modernly, it's predicated on falsifiability. "Manifestly implausible?" If so, then it's not science. Manifest the implausibility, and it's not a good theory. (And I don't think religion is allowed to be forsaken when implausibility is manifest, not when it's mature anyway.)

I like your points on it implausibility after the fact. The actual science that managed to get done back in the days when theology was considered a science is something I find amazing.

As far as cognitive bases for religion, I say glibly* that you can throw observable human cognitive elements like "face recognition" and "speech" into the pot and come out with something closely resembling religion, much like Aristotle could doctor earth with a little fire to make gold...


*because I don't have good tools to argue it sincerely

twiffer said...

why does anyone believe anything? or, more to the heart of it, why do those people not believe the same things i do? because, of course, one's personal beliefs are the correct ones.

my personal take on religion is that it is a social construct, like any other. the purpose of religion is to set rules for society and to act as an authority. i think the question, though, that is asked is why do people cling to beliefs that have been reasonably disproved*? perhaps it's little more than comfort. acertaining what convinces someone of the truth is a difficult task. defining truth is hard enough in and of itself.

*yes, nothing can actually be "proved". take it as meaning proved as well as we can. everything is fluid anyway.

august said...

Sorry for my poor phrasing. My argument in a sentence: until very recently, "religion" is not distinguishable from other areas of human endeavor, so any argument about the evolution of "religion" is an argument about the evolution of all sorts of other things we now distinguish from religion.

Keif -- the question is my paraphrase of what Gene Expression seems to be asking. I think there are interesting arguments to be made about the distinction of religion and science, but that the differentiation does not happen at the level of genes (which is the only level that natural selection works).

Sorry, not a lot of time right now. more specific responses later.

Gregor Samsa said...

August: You misunderstand the purpose. The Times article gives a clearer explanation of the basics than the GE link.

First, evolutionary theory seeks to explain not particulars and differences between cultures, but commonalities, especially universals. Theism is much more pervasive than monotheism, so it is the former which is likely to have a stronger evolutionary basis. Look out for specific features which lie at the intersection of all world religions (theism, particularly with an anthropocentric bent, belief in afterlife and efficacy of prayer/ritual, etc.). You’re right that “religion” doesn’t work too well as a category in itself, but the unit of analysis (“mental modules”) is usually smaller and more focused. It’s a separate issue whether these units can be viewed as constituents of what may be called the “religious mindset” (need not be exclusive).

Second, evolutionary theorists love a challenge, so the game is to make sense of psychological propensities that are at first glance a disadvantage for propagation. Lust is no evolutionary mystery, but love is (not really, but requires a closer look). There are mental propensities which confer a clear survival advantage (e.g. sense of language or geometry), even if their application is not always free of error (bad science, incoherent speech). Those elements of universal religion are interesting which, ipso facto, (apparently) handicap the gene’s interest in spreading (e.g. by inducing suicide-bombing, or the reluctance to covet thy neighbor’s wife). Generally speaking, religion is curious insofar as it tends to instill in us a wrong model of the material world and disincentivizes the pursuit of genetic replication within it.

august said...


The point is that if you look at the details at what the article claims is being selected for, it is either

a. "group selection" which I think is bogus

b. a kind of unintended consequence argument. what i object to here is the idea that the primary unintended consequence is theism.

As I say, I know I need to fill this out, and have suddenly been slammed with more pressing real life things. But you could as easily say that science is curious in that it has frequently offered wrong models of the material world and carried results (nuclear waste, concerns about overpopulation) that seem to handicap the gene's ability to/interest in spreading.

I'm accusing scientists of having a poor understanding of the history of science prior to about 1500. Put it differently -- if we looked at Pythagoras, then deism becomes indistinguishable from mathematics.

august said...

"unintended consequence" very poor phrasing, sorry -- I'll spell out the times article when I get a chance and revise the top post.

Gregor Samsa said...

(a) Group selection is bogus if taken as a primitive (i.e. the starting point of explanations). It is not bogus when (something like) it can be shown as a derivative of genetic level selection in specific contexts. Some of the most interesting work in neo-Darwinian theory has shown how altruistic and sacrificial behavior in animals (with more liberties in theorizing, humans) can be sensible from a gene’s narrower point of view. However imperfectly, we have evolved as moral animals, haven’t we? Dawkins loves trashing group selection (and he is right, under the proper interpretation), but his rhetoric can mislead first time readers. Keep an eye out for terms like kin selection and reciprocal altruism.

(b) The unintended consequence argument has been used to explain many, many things besides theism (e.g. comedy and obesity). The “evolution of religion” project is in many ways old hammers looking for new nails.

Regarding science v religion, we are in a semantic trap, I’m afraid. It’s best to peel away these labels and read without prejudice the tighter mechanisms being described (e.g. the evolution of agent detection mechanisms), and then see to what extent they help you understand various existing or historical institutions (science, religion, nations, tabloids…). To state the obvious, what bundle of activities we choose to call religion, science, politics, etc. is somewhat arbitrary, but that’s merely a distraction. Evolution may help us understand the basis of the Pythagoreans’ spatial aptitudes and their spiritual leanings, and whether they should be viewed as occultists or mathematicians is a problem of modern cultural taxonomy, not evolutionary theory.

august said...


Your comments are very helpful, thanks. I'll try to think harder about group selection. A paragraph in the Times article that annoyed me was this one:

"There might be clear costs to taking on a role analogous to the sentry bird — a person who stands up to authority, for instance, risks losing his job, going to jail or getting beaten by the police — but in humans, these local costs might be outweighed by long-distance benefits. If a particular selfless trait enhances a person’s reputation, spread through the written and spoken word, it might give him an advantage in many of life’s challenges, like finding a mate. One way that reputation is enhanced is by being ostentatiously religious."

That doesn't sound to me like reciprocal altruism. It sounds to me like circular reasoning. But perhaps I am wrong to single out that part of the article. I'd certainly defer to you on the issue.

I think your last paragraph says exactly what I was trying to say ("The point is, cognitive activity does not easily separate into modern categories of 'religion,' 'economics,' and so forth."), but more elegantly. I meant that even given that no modern word is going to be a perfect fit, the choice of describing the processes as the "evolution of religion" is deliberate and misleading.

At any rate, my impression is that we agree, in that "reading without predjudice the tighter mechanisms being described" is what I would advocate. I don't deny that such mechanisms will be linked to a variety of human activities (including what we now call "religious"), but I do hope that when it comes time to make those links, a certain precision is brought to bear about what is being talked about.

Thanks again. I'm a neophyte, and so it's possible that I'm still misreading (you and the article). If my impression of agreement is misplaced, I trust you'll let me know.

Gregor Samsa said...

No problem. I hope I managed to be more engaged than didactic. Seeing you display a serious interest in evolution (saw your reading list earlier) triggers in me the religious instinct of seeking recruits.

I think we agree in terms of separating the evolutionary psychology aspect from the sociological/taxonomic aspect, which was the point I wished to press (like everything else, that too is artificial beyond a point, but still a useful dichotomy in my view). I am afraid we’ll probably still disagree a lot within each domain. I don’t find your extended quote disagreeable; just too general to be helpful (which may be why you found it circular). But here’s the bombshell. I do think the stuff they are talking about can aid our understanding of what we conventionally (20th century, western) view as religion, though the evolutionary part of it is quite speculative. Talk about a bait-and-switch, but I’ll entertain complaints (and arguments) some other time perhaps.

august said...

Gregor - I'll look forward to it.

Swit -- if I'm reading you right (not likely) I think you are saying two different things.

On the one and you are talking about people (musicologists?) looking back at what happened and drawing a straight, deterministic line across ground that, at the time the musicians were composing, was open and fluid.

On the other hand, you seem to also feel that musical innovation does follow some kind of internal logic. So (again, trying to follow you), its fluidity is not absolute. Bach had nearly infinite options for innovation, but he was still going to work in a certain historically determined mode, not pump out "Macarena" or even "Suite Judy Blue Eyes."

After that, I have to do a little filling in to get to your last paragraph. I guess the state propaganda notion is frightening because people tend to work within a common enough musical idiom (probably a bad word) that the forms, and therefore the listeners, can be manipulated. And liberating, I suppose, because what makes a great artist great is precisely the capacity to innovate using available resources.

I don't know if that's where you are going, but perhaps my attempt at paraphrase will help you along the way.

Anonymous said...

Gregor quite weak on a). Excellent explication I know of kin selection and reciprocal altruism in Selfish Gene (later editions - but still long long ago, and probably in august's). last two chapters (as I recall), and particularly "nice guys finish first". I think Gregor's use of "group selection" is very confusing here, although probably not confused (kin selection insanely not to be linked with group selec, and not recip altruism either [less colloquially insane]). Also, for the record, Dawkins' own position on true group selection considerably more nuanced than Gregor's (i.e., bogus) - certain conditions, certain levels, certain constraints, but generally not.

Other than that: sure.

twiffer said...

so, i was watching a nature show last night*. one where they show orca hunting sea lion pups, by racing up onto shore with a breaking wave. and i thought, to the sea lions, orca must seem as demons. extend this to the concept of animism and, perhaps, the genesis of religion is in hunting and being hunted.

just a thought.

*okay, i was a little baked at the time.

Gregor Samsa said...

“Group selection” was the term invoked by august; I was merely echoing it but with the purpose of not linking but delinking the various strands. The NYT piece talks about one guy using group selectionist arguments (David Sloan Wilson) which neatly segues into another fellow (Richard Sosis) using gene-centric commitment arguments (submitting to costly religious rituals signal of commitment to communal yada). I can only conclude that august was conflating the two and putting them all in one box (the quote he pulls out is taken from the second strand, not the first), and tried to warn him off it. Clumsy no doubt, but try doing it in a few words.

Dawkins may discuss academic group selectionists in measured tones, but he is acerbic when addressing the vulgarized pop Darwinism which often (and obliviously) takes it as axiomatic. The rhetoric isn’t even carefully confined; the NYT piece quotes him as describing Wilson’s arguments as “sheer, wanton, head-in-bag perversity”!

august said...


I can't judge Wilson outside what is presented in the article. In the article, the paragraph I cited is pretty clearly in the Wilson camp. I thought it was circular because the starting point (the gene) was "belief in god(s), " Wilson seems to say that selection for belief in god(s) results in selflessness, and we know theat there is a correlation because "one's reputation is enhanced by being obviously religious." The link between the original gene and what Wilson is talking about is assumed rather than demonstrated, and (in the article's next paragraph) when he starts talking about an "adverserial relationship between deists and atheists" that "keeps social groups on an even keel" -- I can only assume he's talking about the selfless and the selfish, but again the correlation between those two traits and "deism" or "atheism" is not clear. Wilson's the one who studies religion because he thinks it tells us something about group selection. If he were just saying that there are selfish and cooperative strategies, and that communication makes cooperative strategies more likely to work -- then that would be the same as Dawkins.

All this strikes me as the opposite of the tact you advised, of being very precise about the genes being selected.

All that being said, I happily confess that I don't get Wilson, and welcome enlightenment. Sorry to reopen this after we had come to a happy equilibrium (roughly -- kicking the can down the road). My continued patter is sheer, wanton, head-in-bag perversity, and by way of apology I'm happy for you to have the last word if you want it.

Twif-- I'd actually be sympathetic to an argument that said religion results from getting baked.

Heliogabalus said...

"is an information pattern, held in an individual's memory, which is capable of being copied to another individual's memory".

Definition of Memetics.

Gregor Samsa said...

“Last word” suggests an adversarial framing which wasn’t my intention, so apologies if necessary. Let me back up and try to be clearer.

The article divides evolutionary theories of religion into two broad groups (describing, not endorsing):

(1) Byproduct theories. These postulate that evolution selected psychological traits (e.g. hypersensitive agent detection faculties) that manifested in lots of survival enhancing behavior (e.g. fleeing predators) but also others (e.g. ascribing an intentional agent behind storms and floods) whose contribution to survival prospects is neutral or negative. Natural selection would favor these “mental modules” if the value of the beneficial behaviors generated outweighed the costs of wasteful ones, at least in the ancestral environment.

(2) Adaptive theories. These share the view that the religious expression of these mental modules could be useful (fitness-wise) in itself, and not a useless relic or harmful side effect. The question is, useful to whom?

Wilson is of the opinion that it was useful to clans or tribes (by promoting intra-group cooperation via the putative motive of serving a common god) in the competition for resources with other tribes, and that is why they were selected. This line of reasoning takes the group as the unit over which natural selection acts, and isn’t the favored framework of mainstream evolutionary biology.

Sosis’ arguments, on the other hand, try to show (forget the merit of its particulars) that religiosity could be fitness optimal for individuals within groups. It may also make the group stronger in the process, but that is incidental to his logic, which takes the individual (more accurately, the gene) as the unit of selection and is thereby very much within neo-Darwinian orthodoxy.

Now the quote you produced appears in the middle of the bit on Wilson, but its reasoning clearly hinges on individual benefits (acting through enhanced personal reputation in the community, etc.), and seems closer in spirit to Sosis. Either Wilson’s own reasoning is a cocktail, or the reporter is jumping back and forth. What you objected to specifically is in “Wilson’s camp” in the sense of being an adaptationist view, but decidedly not so in the sense of being a group selectionist argument. Your apparent interpretation of the entire adaptationist line as group selection is faulty (and invoking social Darwinism is even more misplaced, if only for the implied naturalistic fallacy). Highlighting this distinction was my point.

That’s about it (hopefully clearer), but I’m tempted to add a point here which is more relevant to your top post. Sosis’ arguments are really for the evolution of a kind of tribalist instinct, mediated through a shared belief system and rewards/punishments to enforce/screen loyalty to it. It “explains religion” insofar as tribal loyalty is a core feature of practiced religion. The phenomenon is clearly not unique to religion (say as defined today), because many other institutions exploit it too (nations, warring cultures, scientific paradigms, bulletin boards [fray v dailykos], etc.) So what? A viral explanation of the common cold isn’t undermined by the fact that many other ailments have one too, and it’s hasty to conclude that every disease has one (bacterial, not microbe based, etc.).

The article’s main weakness lies in its oppositional portrait of a bunch of proposed mechanisms, whereas many of them are likely to have acted in complementary ways. For example, a belief in a jealous, cosmic master may act as a stronger glue to hold together tribes than belief in abstract, depersonalized notions (exploiting agent detection hypersensitivity as well as tribalism). Thus, religion’s success may lie in having harnessing the power of many more evolved instincts than less durable institutions, like bridge clubs or debating societies.

august said...


"Last word" was meant to be a way of bringing the conversation to a close, mainly because I haven't finished the Dawkins book yet, and also because I'm just too swamped at the moment to really think through the words I'm using, or to go back and give a more thorough revision of my top post. I meant "social Darwinism" mainly in response to the Gene Expression post, which has some bizarre argument about competition among religions (not a genetic argument).

I see the analogy with a virus. But two points. First, if there is research on viruses in general, and the headline is about AIDS, the headline is accurate but still partial. Second, the use of the term "religion" in the article is downright sloppy. I don't have time to go through and show the different individual phenomena it points to, but it goes well beyond belief in god.

Let me try to take the time to explain what I still don't get about Wilson.

If I have a group of birds, and a sentry gene appears in a subset of that group, and if there is a real cost to being a sentry -- then I would expect that a gene in which a critter either was or was not a sentry would not work very well. If it makes the bird even slightly more likely to die, then the gene would disappear over time. What I think would work is a prisoner's dillema type game in which all the birds (or some large subset) could act as sentries, and would be willing to take the risk if other birds were as well. I'm sure you could work out mathematically the optimal amount of time that a bird would want to be a sentry, the number of cheats (birds refusing to be sentries) you could have and still have a stable population, and so forth.

If I understand Wilson, he's arguing that groups of birds with sentries do better than groups of birds without. I don't quite understand his genetic argument for this. I guess I could imagine that there is a sentry gene among some number of birds and a "be nice to sentries" gene among the others. The sentries guard, and the nice birds -- I don't know --do something to ensure that sentry genes don't die out. Maybe they mate with sentries. I suppose they could adopt sentry babies, and if they did that would certainly hurt Dawkin's argument.

So that's my understanding of the birds.

Rats, I have to go. Maybe smartest if I just drop the whole religion angle. Is that a fair characterization of how Wilson and Dawkins disagree?

twiffer said...

well, i think mushrooms figure heavily too. so, a working theory that religion stems from ingesting hallugenogens, combined with the threat of being eaten by orca. or lions.