UK paperback

Natural selection

Bear necessities

Spare a thought for the polar bear. Not only is it slandered as “one of nature’s most vicious beasts” by global-warming “sceptic” Brendan O’Neill ((Who, on the matter of global warming, cites the business-friendly political scientist Björn Lomborg, always a revealing sign.)), it is also now at the centre of a long-running philosophical/scientific debate occurring in the august pages of the London Review of Books.

Jerry Fodor, of whose writing on the philosophy of mind I have long been an admirer, there published an article on evolution last October called Why Pigs Don’t Have Wings. After reading it a couple of times I found myself scratching my head, wondering what exactly he was getting at. I felt slightly better after a variety of biologists and philosophers and other responders to the LRB letters pages in subsequent issues also signalled their uncertainty as to what exactly he was getting at. But we can be sure that it involved polar bears.

As he explains in the original article, Fodor thinks there is a conceptual problem with the theory of natural selection. The problem, in a nutshell, is as follows:

The crucial test is whether one’s pet theory can distinguish between selection for trait A and selection for trait B when A and B are coextensive: were polar bears selected for being white or for matching their environment? Search me; and search any kind of adaptationism I’ve heard of.

“But,” you may respond (as I did, mentally), “for polar bears in the wild, ‘being white’ and ‘matching their environment’ are not two traits that we may label trait A and trait B while allowing that they are ‘coextensive’; they are the same trait.” ((As Tim Lewens’s letter subsequently pointed out: “[P]olar bears are camouflaged in virtue of being white.”)) A polar bear could not match its environment without being white; a polar bear could not be white and fail to match its environment. For a polar bear, being white just is matching its environment. Well, that is what I thought, and still do. Just because you can describe it in two ways doesn’t make it two different things. So there is no challenge for “adaptationism” here. Demanding that it be able to distinguish between one and the same thing seems a little unfair.

Let us leave the polar bears there for a moment, as they swim (I hope not too far) to the next ice floe. Fodor further argues that there is a problem with the implications of the vocabulary, often used in evolutionary theory, of an animal’s traits having been “selected for”. He writes:

It couldn’t, for example, be literally true that the traits selected for are the ones Mother Nature has in mind when she does the selecting; nor can it be literally true that they are the traits one’s selfish genes have in mind when they undertake to reproduce themselves. There is, after all, no Mother Nature, and genes don’t have, or lack, personality defects. Metaphors are fine things; science probably couldn’t be done without them. But they are supposed to be the sort of things that can, in a pinch, be cashed. Lacking a serious and literal construal of ‘selection for’, adaptationism founders on this methodological truism.

Now, up until the last sentence, this is all perfectly true, and a good illustration of how the idea of “natural selection” might in careless hands be seen as Unspeak, smuggling in an idea of purposiveness or even design. But of course, such language has long been understood as being merely a convenient shorthand for the mechanism that the theory actually proposes. Darwin himself made sure to point this out very clearly, at the end of Chapter 4 of On the Origin of Species:

But if variations useful to any organic being do occur, assuredly individuals thus characterised will have the best chance of being preserved in the struggle for life; and from the strong principle of inheritance they will tend to produce offspring similarly characterised. This principle of preservation, I have called, for the sake of brevity, Natural Selection.

There you go: for the sake of brevity — and, despite Fodor’s somewhat wilful display of bafflement, there is not much mystery as to what it is short for. This is explained patiently in a letter in the latest LRB from Simon Blackburn, Jerry Coyne, Philip Kitcher, Tim Lewens and Steven Rose:

In large numbers of articles and books, published from 1859 to the present, evolutionary biologists use the following style of explanation. A characteristic of an organism (the colour of an animal’s coat, say) is as it is because of a historical process. In some ancestral population there was a variant type that differed from the rest in ways that enhanced reproductive success. (White polar bears, for example, more camouflaged than their brown confrères, were better at sneaking up on seals, were better fed and left more offspring.) If the variant has a genetic basis, its frequency increases in the next generation.

So far, so good. But look, the polar bears are coming back!:

Is this incoherent? Nothing Fodor says bears on that question. Instead, he opposes a very particular way of presenting the explanation. Some people think we can talk of ‘selection for’ a characteristic, and identify rather precisely the traits that have been ‘selected for’. Fodor tries to argue that this is wrong: that there is no single correct answer (whether we know it or not) to the question of whether it was the whiteness of polar bears or their blending in with their surroundings that was ‘selected for’. Whether he is right is a philosophical issue about which people can disagree, but it has nothing to do with the coherence of Darwinian explanation. Natural selection proceeds if three elements are in place: variation in a trait, an effect of the variation on reproductive success, and some means by which the trait is inherited. Both the whiteness and the environmental blending emerged from the historical process that the selection explanation describes.

I admit that I don’t understand why Blackburn et al concede to Fodor his distinction between polar bears’ being white and their blending in with their surroundings, since, for the reasons given, I don’t believe these are separate things. In any case, the already soupy plot thickens and changes hue, as Fodor replies to their letter thusly:

Since the hypotheses that the bears were selected for being white and that they were selected for matching their environments support different counterfactuals (what would have happened if their environment had been orange?) they can perfectly well be distinguished in (for example, experimental) environments in which one trait is instantiated and the other one isn’t. I don’t claim that locally coextensive properties are indistinguishable in principle. I claim that, since the theory of natural selection fails to distinguish them, there must be something wrong with the theory.

I am not an evolutionary biologist, but I suppose that no one would seriously propose that predators in an orange environment would be selected for having white fur. Fodor only arrives at this artificially absurd possibility because he has split the single observed quality of polar bears being white-and-camouflaged into two properties, one with a potential to be universally applied (white), and one relative to a particular environment (camouflaged), supposing that the different verbalizations now point to two different things in the world. Let us hypothesize that polar bears in an orange environment would, ceteris paribus, at length become orange. Now, in an orange environment, being orange and being camouflaged are not two “locally coextensive properties”, they are the same property; just as in a white environment, being white and being camouflaged are the same property. What Fodor sneakily calls “the hypotheses that the bears were selected for being white and that they were selected for matching their environments” are not hypotheses plural, but a single hypothesis that takes into account, as it properly should, the features of the actual icy environment in which we observe polar bears splashing, gambolling, and viciously eating seals.

So I still don’t understand Fodor’s reason for there being “something wrong with the theory”. Any ideas?

  1. 1  abb1  January 10, 2008, 11:48 am 

    Yeah, it sounds absurd. Maybe it’s all a joke?

  2. 2  Andrew F  January 10, 2008, 11:53 am 

    I’m not a biologist, but I’ve been made proof a thesis written by one. My guess is that the two options are:

    Colour for environmental reasons (better at living)
    Colour for sexual selection reasons (more sexy)

    There are many species who have selected based on non-sensical colours, the blue of birds for example is not for any other reason to look sexy to other birds (in fact often the feamale who doesn’t need to compete sports a camo colour scheme – not very hot at all but much more handy).

    One of the reasons that the female birds think that the colours are attractive is becouse they are hard to produce. Males with good colours are stronger, they can afford to produce colourful coats because they are better at finding food and faster at avoiding predators. The colour is not directly an environmental reason.

    But that said, most animals at the top of the food chain don’t select on weird coat colours, so I think that in orange snow land polar bears would be orange and the white ones would be hungry and lonely.

  3. 3  lamentreat  January 10, 2008, 11:57 am 

    Thanks for explaining all this so clearly. I started reading the original Fodor article but my eyes glazed over after a bit and I went to clean the bathroom.
    Isn’t Fodor the one who periodically has nasty exchanges in the LRB with Daniel Dennett about computers and the philosophy of mind? (Exchanges which, as a soft-brained humanist, I excuse myself from reading with any real attention and so can’t summarize their positions.) Maybe he was just so gung-ho to score points that he made a stupid mistake.

  4. 4  David McGuire  January 10, 2008, 12:23 pm 

    Fodor is talking nonsense. It is the kind of objection you sometimes see raised in discussions of popular science books, where mathematical expressions have been reduced (or however you want to term it!) to the level of language, and those who lack the maths but object to the conclusions decide to tackle perceived problems in the language itself. Their attacks are futile because the real target, the mathematics (or in this case, evolutionary theory), is not there to be hit.

    Perhaps in Fodor we are witnessing the first steps towards the level of misunderstanding and willful blindness necessary to support a belief in ID!

    Although I wonder how long it will take the ID supporters to start linking to this exchange as proof of the ‘the controversy’.

  5. 5  john b  January 10, 2008, 12:26 pm 

    More trivially, it took me a few moments to understand the meaning of this sentence: “Nothing Fodor says bears on that question“.

  6. 6  Steven  January 10, 2008, 12:49 pm 

    Andrew F — I don’t think Fodor is implying the distinction that you point out is available, because he erects his pseudo-distinction between “white” and “matching the environment” in the context of worrying about agency or lack of it in natural selection specifically. (“The theory of natural selection fails to distinguish them.”) In sexual selection, by contrast, there is no mystery about agency: the mates choose.

    Lamentreat — yes, that’s him. They had another to-do during this debate, where Fodor really goes off on one in most entertaining fashion: “Over the years, I’ve been finding it increasingly difficult to figure out which bits of Daniel Dennett’s stuff are supposed to be the arguments and which are just rhetorical posturing. In the present case, I give up.”

    David —

    Their attacks are futile because the real target, the mathematics (or in this case, evolutionary theory), is not there to be hit.

    Yep. I think Blackburn et al’s latest reply is more or less definitive in this respect. The curious thing is that Fodor is forced to concede that there is nothing wrong with the first paragraph I quoted from their letter (beginning “In large numbers of articles and books, published from 1859 to the present, evolutionary biologists use the following style of explanation…”), but then pleads that “they’ve somehow left out the Darwin bit”. Obviously they have not; they’ve just given a very classically Darwinian explanation. (The date of 1859 ought to have been a clue.)

  7. 7  belle le triste  January 10, 2008, 2:28 pm 

    yes i slugged a bit through the letters then left it on one side till xmas cheer and/or torpor had dispersed (ie not yet) — if fodor is wrong, being very much not an idiot on the whole, then it’s an interesting example of the elegant power of darwinism being too elegant for thr brainy (ie fodor is creating a pseudo-problem by seeing the situtation — so why polar bears in the arctic — overcomplicatedly) (as if somewhere someone thought “aha this goes with this”, when actually the unity comes first — ecology-as-creatures-interracting-in-and-wth-a-place — and the distinction (into bears and snow) comes afterwards, with us

  8. 8  Steven  January 10, 2008, 2:46 pm 

    Very nice way to put it, belle.

  9. 9  richard  January 10, 2008, 4:15 pm 

    we might equally ask how rice came to be white, and therefore to appeal to peoples with a religious whiteness fixation.

    Has anyone tried taking polar bears to warm temperate climes and spraying them orange, to see if this affects their ability to breed in captivity? Perhaps it would cheer them up a bit (I don’t think it could drive them crazier). A side project for sw, perhaps?

  10. 10  Rob  January 10, 2008, 4:16 pm 

    I think the polar bears case may involve a bit of misdirection, since it does seem plausible both that the trait ‘being white’ and the trait ‘being adequately camouflaged’ are one and the same in their case, and that ‘being adequately camouflaged’ is a trait which is obviously going result in greater sexual success. I thought that Fodor’s point, though, was that that may well not be the case in lots of other situations, that we may not know how to characterise the relevant trait or why the trait is one which results in greater sexual success. That would seem to make sense of what he says about the necessity of a notion of ‘selection for’. That isn’t to say that he’s right, obviously.

  11. 11  richard  January 10, 2008, 4:43 pm 

    rob – does that mean he’s claiming the notion of “selection for” should take account of our ignorance? (while Steven seems to be saying Fodor reckons it should take account of any arbitrary scheme of distinctions we can come up with.)

    As a side note, they’re both white and colour neutral (since their whiteness derives from scattering in transparent, hollow hair fibres, not from any pigmentation). Does that mean we should add a further category of selection?

  12. 12  dsquared  January 10, 2008, 6:41 pm 

    Functional explanations always trip people up in this manner. The question of whether the barley, hops and yeast which go into Miller Lite have been selected so as to produce the effect of “Tastes Great” or “Less Filling” has been going on for over twenty years now, seemingly without an inch of progress.

  13. 13  Steven  January 10, 2008, 7:44 pm 

    Cheers, dsquared.

    Rob –

    I thought that Fodor’s point, though, was that that may well not be the case in lots of other situations, that we may not know how to characterise the relevant trait or why the trait is one which results in greater sexual success.

    I don’t think that’s all Fodor can be arguing, since all it really amounts to saying is that the current canon of explanations of the “natural selection” type in evolutionary biology doesn’t yet give a complete explanation for all features of all organisms, and no biologist would be so silly as to claim that it does, because firstly, no one thinks that environmental selection is the only evolutionary mechanism, and secondly, evolution is not a complete and finished science, just as physics isn’t. Fodor seems to be complaining that in principle evolutionary science as he understands it cannot answer some important questions, because there is some kind of incoherence in the very idea of selection as he understands it. But his other example of the supposed impossibility of deciding a selection question in “coextensive” traits (this time, curly tails and tameness in dogs) seems to me to have been adequately answered by Tim Lewens’s letter.

  14. 14  Richard Carter, FCD  January 10, 2008, 10:11 pm 

    I clearly wasn’t the only one perplexed by Fodor’s piece, which I read three times. He’s writing a book on the same subject, apparently. I hope it’s written a bit more clearly than his LRB piece. Not that I’ll be buying it.

  15. 15  Alex Higgins  January 10, 2008, 11:01 pm 

    RE: Brendan O’Neill’s piece:

    “According to Bjorn Lomborg…”

    Isn’t there some kind of logical fallacy inherent in even beginning a sentence like that?

  16. 16  Jay  January 11, 2008, 12:26 am 

    I think that Fodor has not understood that to say a trait “has been selected for” means that the trait has been selected for in the species’ ancestral environment.

    Polar bears are bears that adapted to a polar environment. As a result, their traits tend to be useful in a polar environment. It is unsurprising if their traits are not adaptive in an orange environment, or in the tropics, or in deep space. This causes no problems for evolutionary theory, as ancestral polar bears were not selected for survival in non-polar environments.

  17. 17  Matt McGrattan  January 11, 2008, 12:37 pm 

    Philosophers of biology have been making the distinction between selection for a trait and selection of a trait for decades.

    It’s a fairly common-place distinction — despite what Fodor seems to think — and the apparatus of ‘evolutionary theory’ can comfortably handle it.

    There are real world cases where there’s some confusion about which trait is the selectively important one, when the two traits are coextensive in some way. Cases of pleiotropy being the obvious one. If some gene codes for several phenotypic traits, it might be very useful for biologists to find out which of those phenotypic traits is the one that’s driven that gene to fixation in a given population.

  18. 18  Matt McGrattan  January 11, 2008, 12:39 pm 

    And reading Lewens letter, he brings up Sober’s for/of distinction.

  19. 19  guano  January 11, 2008, 1:24 pm 

    Is it a logical fallacy to start a sentence “According to Bjorn Lomborg ….”

    Well, it does suggest that O’Neill is not trying very hard to win over people to his point of view, like Ollie Kamm’s habit of citing only Nick Cohen and OK’s editor (former editor ?) at the Times. Even if the point of view is a valid one many people will ignore it because of it being associated with someone who has an axe to grind.

  20. 20  mark c  January 11, 2008, 6:51 pm 

    Brendan O Niell is one of the spiked online/Living Marxism crew – they rant on about environmentalism (from a super libertarian point of view, basically) almost as much as they do about Yugoslavia.

    Still, the argument “Polars would bite you if they met you, therefore environmentalism is wrong” is comedy gold!

  21. 21  Gavin  January 11, 2008, 10:04 pm 

    I wonder if Brendan likes Fox’s Glacier Mints; perhaps he avoided them for the same reason.

  22. 22  dominic  January 12, 2008, 3:26 am 

    Fodor doesn’t buy the selection of/ selection for distinction for the following reasons:
    Sober imagined a filter that let only round objects pass through it, and the round objects just happened to be blue. Although (he said) it makes sense to say that there was selection of blue objects, it is roundness that was selected for (cos if the filter hadn’t had round holes the blue objects wouldn’t have passed thru it). Fodor thinks this makes sense only because we have stipulated what the device is doing (or perhaps because the physical constraints are obvious). But in nature the physical constraints are not obvious, and we cannot assume that mechanisms of natural selection are designed in a certain way. So he denies that the distinction makes sense. Jerry’s an old teacher of mine and I admire him a lot, but I’ve never really followed the details of the argument.

    But there are other things in the background: one is his belief that all scientific explanations involve laws, and that all laws support counterfactuals. (As an aside for any phil science nerds reading this, JF once told me that this dispute is an argument about Hempel.)

    Jerry did also, at one point, believe that natural selection cannot be the cause of evolution because it is a trial and error process, and he thought that there were proofs from the history of AI that no trial and error process (a “hill climbing algorithm”) could solve a learning problem.

    Chomsky used to say the same thing, and I think in both men the deepest-seated objection is this: evolution is like learning, in that you have to adapt to the environment; hence, seeing evolution as caused by natural selection is tantamount to empiricism, because it puts the cause of adaptation in the environment rather than within the organism. And empiricism is false.

  23. 23  Steven  January 12, 2008, 10:36 am 

    Thanks very much for that, Dominic. I can’t help but still suspect that Fodor’s argument as you summarize it just goes away once one makes the gestalt shift of ceasing to think of “natural selection” as a black box and instead thinking of the descriptions of the theory for which “natural selection” is the shorthand, ie the first sentence of Darwin that I cited, and the first paragraph of Blackburn et al.

    But in nature the physical constraints are not obvious, and we cannot assume that mechanisms of natural selection are designed in a certain way. So he denies that the distinction makes sense.

    It’s one thing, it seems to me, for Fodor to say uncontroversially that sometimes we will have trouble naming the feature of the environment that made a certain feature of an organism adaptive; quite another thing to claim that such difficulties mean the whole theory is incoherent.

    The fact that deciding between some coextensive traits can be difficult does not entail that it is always impossible. Presumably in human ancestors, for example, big brains were “selected for” over time, because, let us say crudely, bigger-brained humanoids were better at feeding themselves well and avoiding getting eaten before having children (and then better at feeding those children until maturity). Along with bigger brains by necessity came bigger craniums to keep them in. But plainly it would not make much sense to think that it was the bigger craniums that were “selected for”, rather than the big brains, because we can’t think of any plausible reason why a big cranium with a small brain rattling around in it would be any more useful than a smaller cranium with the same-sized brain. The fact that I just made this story up and that it is totally untestable doesn’t make it actually incoherent, or so it seems to me. And of course there are examples of “natural selection” that can actually be observed happening within the span of a human lifetime, such as with the peppered moth; or studies of the evolution of microbial resistance to antibiotics etc. It would be interesting to know how Fodor explains those.

    evolution is like learning, in that you have to adapt to the environment; hence, seeing evolution as caused by natural selection is tantamount to empiricism, because it puts the cause of adaptation in the environment rather than within the organism.

    Though (a) it’s not any individual that “learns”; and (b) the cause, or at least the primary cause, of adaptation surely is located by the theory within the organism, in the mutation or other mechanism that gives rise to a certain trait that may or may not provide a reproductive advantage among a certain population in a certain environment?

  24. 24  Gavin  January 12, 2008, 12:49 pm 

    There is also the point that the environment is also ‘evolving’.

    In the case of the Polar Bear this is not as important as other cases; the environment is relatively poor in biodiversity, but it makes no sense to talk of the environment selecting for traits, since the environment itself is under selection pressure.
    Fodor seems to be inventing himself a problem so that he can be the one to ‘solve’ it.

  25. 25  dominic  January 13, 2008, 2:54 am 

    Well, there are mutations in the organism, but most of the explanation of adaptation lies in the problems posed by the environment. On a Lamarckian story, maybe, it’s the other way round, as the forces within the organism do most of the explaining. I don’t think it’s crazy to think (as several scholars have) that Darwinism and empiricism, if not the same thing, are both instances of a wider way of thinking about interaction between organism and environment, one that puts most of the explanatory structure outside rather than inside the animal.

    I think you’re correct that Fodor is thinking of natural selection as one thing rather than a lot of local, particular causes. (Another of Sober’s analogies was that evolution was a theory of forces, with natural selection as one force among others.) But that is a view of scientific explanation – local mechanisms rather than general laws – that Jerry shies away from.

  26. 26  sam.the.pantisocratist  January 14, 2008, 11:57 pm 

    Faust: Wohin der Weg?

    Mephistopheles: Kein Weg! Ins Unbetretene.

    Indeed. As a personal friend of Dr Darwin I find the fanatical outpourings of Rose & Dawkins disturbing – even they must realise that Darwin was no Darwinist. For the full monty check out Erasmus’ magnus opum – The Loves of the Plants

    Like the shades Steve. V cool. V happening now.

  27. 27  satyarth  January 15, 2008, 3:25 pm 

    I loved your book……reminded me of naom chomsky’s essays….just wanted the gratitude to reach you, and there was no other way.
    so forgive me for cluttering the space here.


  28. 28  hey zeus  January 17, 2008, 5:08 pm 

    does anyone else remember the mum of a brown skinned boy suing nestle when he was turned away from a casting to find the new milky bar kid?
    just saying. coINcidence?

  29. 29  C. Reaves  January 18, 2008, 2:41 pm 

    I suspect Mr. Fodor is a fanboy of Ayn Rand. Unfortunatly he has cast himself in the role of Ellsworth Toohey instead of Howard Roark.

    As you so neatly pointed out – that Natural Selection was a term chosen by Darwin for the sake of brevity – “select for” is also a term of art (jargon) used by population geneticists for the sake of brevity. It is only when it is used out of that context, as normal English, that controversy can be created.

    There are only two possibilities: 1) that Fodor doesn’t know enough about population genetics to realize he is mis-using the term (in which case he has no business writing about the topic), or 2) he does know it and has chosen to deceive his readers to support some sort of self-serving agenda. I choose the later.

  30. 30  C. Reaves  January 18, 2008, 4:35 pm 

    Dominic, your position that Darwnism and empiricism are the same because they both put the explanatory structure of evolution outside the animal seems incorrect to me.

    hence, seeing evolution as caused by natural selection is tantamount to empiricism, because it puts the cause of adaptation in the environment rather than within the organism. And empiricism is false.

    The error you have made is that one cannot look at color or even camouflage per se as the product of natural selection. What is being selected for, from within the organism, is mimickry. Tree bark does not cause the selection of wing color in moths – instead, the variation of traits within the species results in the selection of moth variants that mimic tree bark. The entirety of that transition is contained within the species.

    Through natural selection polar bears mimic ice, moths mimic tree bark, harmless frogs mimic poisonous day-glo frogs, harmless butterflies mimic poisonous Monarchs. Hence natural selection progresses from within each species, unbothered by claims of false empiricism.

  31. 31  Steven  January 19, 2008, 12:29 pm 

    Hi C.,
    Thanks for the illuminating comments.

    2) he does know it and has chosen to deceive his readers to support some sort of self-serving agenda.

    I would hesitate to think this myself. As belle le triste pointed out, Fodor is very much not an idiot, and I have no reason to think he is being deliberately deceptive. Still, the fact that a whole bunch of eminent philosophers and biologists responded to him with a big “Whuh?” should, you might have thought, give him reason to suppose he hadn’t yet explained himself properly, rather than causing him to throw sarcastic tantrums in his own replies.

    By the way, I had assumed that the bloke Fodor is collaborating with for his forthcoming book on the subject would be a biologist, but it turns out apparently that he is a professor of linguistics and cog. sci. Hmmm…

    This seems like another decent reply to Fodor.

  32. 32  Ken  January 21, 2008, 6:53 pm 

    Fodor’s argumentative style seems inspired more by Stephen Potter than anything else these days. I keep expecting him to say “yes, but only in the South”.

  33. 33  Jonathan  January 21, 2008, 9:35 pm 

    I think what Fodor is getting at is the following:

    A “selection for” needs counterfactual support. We need to be able to say something like: if individual X hadn’t had trait Y, she wouldn’t have reproduced so much. Where might the counterfactual support come from?

    1) A force or law. But natural selection cannot be a force or law. It is too context-specific. This is how Sober characterizes it and he is wrong.

    2) Mother Nature. But that’s silly.

    Without (1) or (2) we can’t make sense of the selection of/selection for distinction. A “selection for” is not part of a plan and it is not an instantiation of a law, so it’s nonsense.

    To take a better example than the one Fodor gives, imagine a population of polar bears. Half have thick-white fur, half have thin-black fur. There is a cold night and all the thin-black fur bears die. Fodor’s point is that, while we can give a post hoc story about what happened, we can’t say it was the instantiation of a law or the will of Mother Nature.

    I think Fodor neglects a third, very intuitive and satisfactory option. Thick fur, not white fur, *caused* the thick-white fur bears to survive in the cold, therefore thick fur, not white fur, was selected for. In the absence of a law or a Mother Nature, we can still invoke *singular causation* to preserve the of/for distinction.

    Still, I have to say I’m impressed he had the balls to take on the established orthodoxy like this.

  34. 34  Steven  January 22, 2008, 1:04 am 

    I like your polar-bear example better than Fodor’s.

    Thick fur, not white fur, *caused* the thick-white fur bears to survive in the cold, therefore thick fur, not white fur, was selected for.

    Doesn’t it make more sense to think of the causation as operating on the other bears: ie, thin fur, not black fur, caused the thin-black-fur bears to die in the icy night before any of them had a chance to reproduce, and all the ones that were left were those with thick-white fur, so naturally their ancestors had thick-white fur too etc. That’s what “selection for thick fur” would really be shorthand for in this case (at least AIUI). So again, I think Fodor’s problem only applies to the shorthand, not to the full description.

    Still, I have to say I’m impressed he had the balls to take on the established orthodoxy like this.

    I think the burden of the criticism at evolutionblog that I linked to at #31, as well as of other responses, is that he doesn’t understand the “established orthodoxy” to begin with.

  35. 35  john c. halasz  January 22, 2008, 6:12 am 

    I read that Fodor piece quite some time ago, and it puzzled me. Fodor does seem to be hampered, like many Analytics, by Quine’s “austere physicalism”, which I think fetishizes physics as the unique model of natural science, and, as well, by residues of the covering-law account of natural scientific explanation, which I don’t think apt. There is also a tendency among some Analytics to assume, due to the logical principle that from a cause an effect necessarily results, a deterministic conception of causality, when stochastic processes have increasingly come to the fore in modern natural science. And the assumption that nature is mindless fails to consider, as far as I can tell, the problematic status of information, which is neither mind, nor matter. That’s, as best as I can make out, some of the a priori sources of his resistence to natural selection as *a* principle mode of biological explanation. Also, he seeks to enlist Gould’s spandrels and evo-devo as countering Darwinism, when both are clearly working within the broad paradigm, even if not in its “hard” reductionist, strict adaptionist mode.

    But I thought what he was trying to get at, his ulterior aim, is that there is not necessarily an account of the philosophy of “mind” to be derived from Darwinian natural selection and the concept of adaption, even if he was implicitly staking such a claim in a wrong-headed and obtuse way. And, indeed, I broadly agree on that point, and think that there is a lot of schlock about nowadays trying to advance such claims. To me, it would suffice, simply to point out that there is no mathematics in nature, and yet human brains evince the capacity to do higher mathematics; indeed, some such brains more than others, and there may be a genetic or biological basis for such different abilities, but not an “adaptive” one. And it’s more than faintly absurd to account for the emergence of so vastly complex and intricately multifunctional an assemblage as the human brain, as if it were a simple organ with an adaptive function like a claw; and as if the brain and the mental capacities it engenders were simply an ad hoc collection of such simple adaptive functions. On the other hand, it does seem to me that brains, as the causal basis of mental functioning, are to be studied biologically, and that means in the broad perspective of evolution, from the first worms, onto fishes and reptile, to mammals and finally that glorious Son of Man, born of a virgin. But an account of the emergence of gray matter and its variable capacities is not likely to yield itself to the terms of strict adaptionism; to the contrary, it’s likely to be an account of increasingly complex integrations and increasingly robust plasticity. And, at any rate, to provided an account of the causal basis of mental *capacities*, on simple logical grounds, is not to provide an account of the causal determinism of such capacities, even if it brings to light constraints upon them.

  36. 36  Steven  January 22, 2008, 11:45 am 

    Hi John:
    Yes, it seems Fodor’s motivation is his long-running bugbear against evolutionary psychology, of whose explanations I often find myself sceptical too (as in this review). On the other hand, I don’t think he succeeds in showing that the possible conceptual or logical abuses in some stories of evolutionary psychology necessitate a wholesale dismantling of evolutionary biology as currently understood.

    In particular, the possible existence of an assumption in some quarters of e.p. that all the cognitive faculties or modules that we observe we currently have can be reliably explained through some story about their enhancing fitness does not mean that Fodor’s account of “adaptationism” in general is not a straw man.

  37. 37  abb1  January 22, 2008, 11:58 am 

    To me, it would suffice, simply to point out that there is no mathematics in nature, and yet human brains evince the capacity to do higher mathematics…

    Bird migration? I don’t really see how a human brain is so different from a claw, or, say, an eye.

  38. 38  john c. halasz  January 22, 2008, 9:08 pm 


    Maybe I was a tad ambiguous in my phrasing. I meant that there is no mathematical ability in nature. Birds don’t navigate by means of calculations or computations. Apparently, they do so by sensing magnetic fields, which IIRC occurs at the molecular level of certain magnetically sensitive organic compounds in the relevant area of their brains. Presumably, paths and places are a matter of imprinting.

    Brains/central nervous systems are different from single organs, such as an eye or, for that matter, a liver, since presumably they emerge, initially for purely physical and physiological reasons, as a multi-functional integration of organic life processes, interacting variously with most of the other functional physiological systems. Presumably, the “force” behind the evolutionary emergence is the need to mediate between physiological and environmental events and process organic decisions for behavioral responses. Any mental properties or capacities emerge as “effects” of such organic behavioral processing, and there are no clear-cut boundaries between the physiological, behavioral, and mental, but rather much of what is attributed as “mind” is generated through the interactions between the three. But brains are to be distinguished from simple organs in that they do seem to evince generalized capacities involved in the integration of multiple “functions” rather than being simply reducible to a tool-kit. (In general, I tend to conceive of mental properties as involving the capacity to respond to contigent, unforeseeable events, as mediated and matched with the stabilizations of various sorts of memory). Your not seeing the difference is perhaps due to your stepping back into the theoretical abstraction of a general materialist skepticism that obscures, if not obliterates, the relevant distinctions or differences.

    I also want to clarify the last sentence of my previous comment, which was too compressed. It is a pet peeve of mine about the way the problem tends to get set up in “philosophy of mind”, whereby a requirement is projected for a total and complete causal explanation that we nonetheless do not have at hand, which then is somehow to be directly identified with fully mental (descriptions of) phenomena. (These often seems to be a lot of residual Cartesianism in the set up, involving a sheerly inert and mechanical external world opposed to an innate “mind”). It is also a quasi-Kantian point about the limits of explanation. To explain a phenomenon, it must first be identified and delimited, (and the identity of the phenomenon must be conserved in the explanation, else it’s not clear what the explanation is an explanation of, a reductionist pitfall), so, indeed, the phenomenon must be a limited “thing” and, in the context of the explanation, structurally constrained. But mental pheonomena are generalized capacities, which are relatively open-ended “things”. To causally explain the basis or substrate of a capacity is, logically, precisely not to give a total causally deterministic account of the capacity, which would no longer be a phenomenon, but an epiphenomenon, which would be to assume that mental phenomena are not reals and have no function or efficacy. So we might well arrive some day at a precise, empirically adequate explanatory account of the neural processes that generate our thoughts, but that would not be a causally deterministic account of particular thoughts, which would be forever unattainable, partly, I would conjecture, because those actual neural processes work in highly stochastic ways, involving literally billions of interacting events and equipotential and multivalent pathways. (And, at any rate, any full-fledged account of human thought would have to involve language/meaning/symbolic thinking at which point you’re no longer wholely within the realm of the causal). As Wittgenstein constantly reminded us, if one fails to actually scrutinize phenomena, to “look” at them, one is liable to set a priori the parameters of one’s problem wrongly, so as to merely generate false “problems” without solutions, philosophical fantasies. And explaining “things” is something that we do, so it is not an extrinsic or irrelevant question to ask what explanations are for, how are they applicable.


    Yes, we’re agreed, at least, that Fodor’s was a rather bizarre strawman account of evolutionary adaption, and actually superfluous to his purpose. But I actually wandered over here last night to try and see what, if anything, might have occasioned the following comment at Ophelia Benson’s redoutable blog:

    “Zizek will never be intellectually disgraced, because he has already proved you can say the most *outrageous* bollocks, and if you are a freaky, kooky East European with a scarily intense knowledge of postmodern gibberish, Americans will fawn over you ’til Kingdom come…”

    “Americans?! It’s not just Americans, I’ll have you know! Steven Poole is no American.
    | OB | “

    Whatever cruel or means thing did you ever do to the poor lass?

  39. 39  Steven  January 22, 2008, 9:52 pm 

    I like your third paragraph very much, will have to think about it. It reminds me of some problem I’ve long had with the idea of philosophical “zombies”.

    As for “Ophelia Benson”: well, in an idle moment last year, I felt moved to point out that someone’s attack on Zizek was idiotic. Subsequently, “Ophelia Benson” dully attempted multiple times to criticise my response to the idiotic attack at her “website”, so I felt obliged to point out in turn that Benson herself was making false claims not just about what I and other people wrote but about her own book. I’m flattered to hear she still thinks of me from time to time.

  40. 40  abb1  January 22, 2008, 10:34 pm 

    Well, yes, exactly: at some level of abstraction there’s no difference between the brain and the claw. Moreover, once you’ve developed a claw you’ll definitely need a decent brain to operate it, and then it’s just a short leap to lambda calculus and Wittgenstein.

  41. 41  john c. halasz  January 23, 2008, 4:42 am 


    Maybe it’s an old residue of Hegelian thinking, but I think that one goes from the abstract to the increasingly concrete and differentiated, rather than the other way around.

  42. 42  C. Reaves  January 29, 2008, 11:29 am 

    To me, it would suffice, simply to point out that there is no mathematics in nature, and yet human brains evince the capacity to do higher mathematics…

    I think what you really mean is that nature has no symbolic mathematics (but even that could be argued – more later). Nature is rife with non-symbolic mathematics. When you toss a ball and your dog catches it both your brains are performing some very complex physics calculations. Every warp of a bird’s wing to adjust the flight is a mathematical calculation. These calculations are not pseudo math, but real math done by very complex organic computers. It only seems like it might not be math because no symbolic equations are written – the brains simply calculate the answers with no intermediate symbolic representations required.

    As for nature lacking even symbolic mathematics: that would require humans to be put outside of nature – something I’m not willing to do. One would never say that because only one whale species can dive to 5,000 feet that there is no deep diving in nature. Similarly, one should not say that because only one mammal has developed symbolic math that there is no symbolic math in nature.

  43. 43  Steven  January 29, 2008, 12:04 pm 

    When you toss a ball and your dog catches it both your brains are performing some very complex physics calculations. Every warp of a bird’s wing to adjust the flight is a mathematical calculation. These calculations are not pseudo math, but real math done by very complex organic computers.

    Hmmm, I thought the current view leant more towards saying that a dog’s (or a person’s) successfully catching a ball was the result of the iterative application of a simple heuristic, not that calculus was actually being done in the wetware.

  44. 44  C. Reaves  January 29, 2008, 1:08 pm 

    I would hesitate to think this myself. As belle le triste pointed out, Fodor is very much not an idiot, and I have no reason to think he is being deliberately deceptive.

    You are probably right, Steven. I find that as I get older I tend to see more manipulation in “mistakes” than I used to. For example, in a similar circumstance in the 80’s Robert Shockley (very smart – Nobel Laureate, inventor of the transistor) decided late in his career that he was an expert in genetics as well as physics. But like Fodor he only had a surface knowledge of his new specialty – anyone who had had any in-depth understanding of genetics knew he was making some very basic mistakes. But everyone knew he was Very Smart so his ideas got a lot more attention than they deserved. I did not accuse Shockley of having ulterior motives at the time – I simply decided that he was not as expert as he thought he was in his new field. I should give Fodor the same benefit of the doubt.

    I have no doubt that Fodor is very much not an idiot, but unfortunately that characteristic didn’t prevent Shockley from looking foolish to real experts. In the end Shockley severely diminished his reputation.

  45. 45  john c. halasz  January 30, 2008, 2:16 am 

    C. Reaves:

    No, when we explain the trajectory of a ball flying through the air, we apply mathematical equations to selected parameters. It’s bootless to claim that the ball, the air and the earth are themselves jointly performing calculations. That would be a version of the “fallacy of misplaced concreteness”. And the point is of some small importance because there are instances where “computations” are part of natural reals: the distribution of breathing pores on the backs of plant leaves, for example. To say that natural things are shaped by natural forces is virtually a tautology, but it’s not the same as saying that they are shaped by our computations of those forces.

    Brains are primarily analog pattern-matching devices, not digital computational ones. When one or one’s dog catches a ball, a mathematical computation is not occurring, which then serves as an instruction for behavioral responses. Described on a neural level, various partial adjustments, “computations” only in a loose sense, are feeding back through re-entrant connections onto one another, leading to an overall result without a central and prior processor or instruction. To speak of a reiterated heuristic is a description on a psychological level, referring to the fact that catching a ball is a practiced skill, integrating recurrent perceptual experience with motility.

    Mathematics consists in formal operations. A monkey can recognize differences in small groups of objects in terms of more or less, but it does so through analog pattern recognition, not by counting. Though doing math is much different than discursive reasoning in natual language, (and probably would show up much differently on a brain scan), I don’t think it’s possible without working on an underlying substrate of discrete symbols. And my point was that there is no natural evolutionary selective-adaptive advantage for our ability to do math, but rather it’s a contingent and superflous by-product of the natural evolution of our peculiar brains, and a relatively late-developing cultural recognition. Indeed, natural language itself, which presumably emerged out of a prior analog system of animal communication, as a partial digitalization of it, is itself largely a contingent outcome of prior evolution. It is bootless to try and claim that all meanings are naturalistic and therefore part of nature. “House” refers to a socio-cultural artifact, not to a natural reality, and “Platonic” universals such as “justice” have no natural basis. Once natural language comes into existence and use, it develops a non-causal dynamic of its own, and much of what we are wont to identify as “mind” is, in fact, an internalization of communicative processes across the world, which brains must have the processing capacity to carry, but don’t causally, internally, and innately originate. The symbolic institutes a socio-cultural world, in which brute facts, like the impacts of physical masses, mix with conventional facts, like the existence of promises, such that explanation in terms of purely naturalistic “meanings” and strictly causal processes begins to lose its force. The project of stuffing the socio-cultural world artificially back into the naturalistic and biological is precisely the sort of thing that lends, at its deepest level, the “fallacy of misplaced concreteness” its proleptic force.

    Fodor was not being idiotic, nor ignorant, to the eternal perdition of whatever reputation he maintains.He’s agnostic, though for highly tendentious reasons. His specific claim, which his rejoinder to Blackburn et alia, if nothing else, makes clear is that the theory of natural selection lacks adequate conceptual means for specifying empirically just what is “selected”. In many empirical instances, given the difficulties involved, I think that’s true, though I don’t think it’s true in principle. That emergent structural constraints of organic processes determine the range of possibilities as to what can be “selected” is something that I don’t think any Darwinian biologists would deny. Fodor seems to be objecting to treating an organism as a tabula rasa randomly accumulating traits from its environment, in which case natural selection amounts to an environmental determinism, (when “adaption” is rather more a matching of the regularities of organic and environmental processes). And natural selection is not an exclusive, self-sufficient causal account of biology and evolution. In addition to sexual selection, which often runs counter to adaptive fitness, (as with the cumbersome tails of peacocks), symbiosis, the co-evolutionary dynamics of ecosystems, and the self-organized properties of matter, which become chained into and physically support and structure organic processes, (such as the folding of limpids when dropped in water, or the “tensegrity” determining the structure of tissue masses), are also significant causal mechanisms in the mix. So Fodor is half-right that natural selection alone is insufficient to account for the significant features of organisms and specify how they are “selected” for, but I don’t think that Darwinian biologists, which is to say, virtually all real biologists, would think to disagree with that. Ironically, I think that it is Fodor’s own combination of “austere” physicalism and innatism/internalism, in opposition to “Darwinian” accounts of “mind”, that leads him astray. Still, it would be interesting to see what he comes up with in his forthcoming book, which presumably will not be ignorant of the writings of actual biologists and fail to address their claims. A dose of agnosticism toward evolutionary theory is not necessarily unhealthy and might serve to prevent premature foreclosures, as with the reformed claidistics controversy two decades ago.

    By the way, I think you meant William Shockley.

  46. 46  richard  January 30, 2008, 6:47 pm 

    john c. halasz:

    I’ve been greatly enjoying your comments, here: you’ve provided a lot to chew on, and reminded me that I really must go and investigate the literature on philosophy of mind. I’m (tendentiously) puzzled by your use of ‘natural,’ though:
    It is bootless to try and claim that all meanings are naturalistic and therefore part of nature. “House” refers to a socio-cultural artifact, not to a natural reality, and “Platonic” universals such as “justice” have no natural basis.
    But I rather like those boots, and I’m not sure why I should do without them. I understand that nature and culture are often employed as opposites, and I understand that your point is that biological explanations won’t be sufficient when seeking explanations for cultural phenomena. Does it follow, however, that culture can be placed outside nature, implying that humans are outside nature? If “house” is beyond the pale of this discussion, what about “shelter” or “comfort?” Also, are “Platonic” universals such as “justice” different in quality or status from ideas like “the stock market,” “casserole” or “politeness,” in your argument? I suspect not, but your mention of Kant earlier has set me wondering about the rest of your Weltanschauung.

  47. 47  john c. halasz  February 1, 2008, 2:45 am 


    I did work out a response to your queries, but it was long, garrulous, and boring. Rather than poaching on byte space here and further embarassing myself, I decided to let it go. But if you’re somehow still interested, I can be reached at

  48. 48  Steven  February 2, 2008, 2:34 am 

    But I’m interested too! I must say that I’ve found it intensely relaxing to follow the superintelligent conversation going on here while I’ve been too busy to post.

  49. 49  abb1  February 4, 2008, 11:55 pm 

    Suddenly it occurred to me that there might be something Ayn-Rand-objectivism-ish about contrasting ‘normal’ nature, animals – with the human mind and its mathematics and ‘rationality’ in general. Not that anything’s wrong with that, necessarily; just sayin’.

    Or maybe I’m just talking nonsense.

hit parade

    guardian articles

    older posts