Misunderstanding Darwin
March 20, 2010 10:23 AM   Subscribe

Misunderstanding Darwin: Natural selection’s secular critics get it wrong. Ned Block and Philip Kitcher review Jerry Fodor's (previously) and Massimo Piattelli-Palmarini's book What Darwin Got Wrong. Fodor and Piattelli-Palmarini respond: “Misunderstanding Darwin”: An Exchange.
posted by homunculus (62 comments total) 20 users marked this as a favorite
 


Jerry Fodor and Elliott Sober discussing the book on Bloggingheads TV.
posted by Dumsnill at 10:32 AM on March 20, 2010 [1 favorite]


See also Papineau on Fodor on Darwin.
posted by ageispolis at 10:34 AM on March 20, 2010


Awesome to see this here. Please please please read/listen to the links before you comment, folks!
posted by anotherpanacea at 10:43 AM on March 20, 2010 [3 favorites]


This is the kind of totally inane argumentation that made me want to slap people in my Phil classes. Why would someone write a philosophical work accusing a scientific theory of engaging in an "intensional fallacy?" Intentionality is so beyond meaningless in the context of any scientific activity it makes me groan out loud. The whole point of the scientific method is to make these distinctions irrelevant! They should do Einstein next, hopefully their heads will explode before the book is finished.

How did I know without looking that these guys were philosophy of mind professors? They clearly wrote this book just to torture students with it. It reminds me of the phil of mind class I took where I spent an entire term being taught that human beings are brainstems and the rest of our body is just interchangeable parts... "human identity without psychology"... hint: it's not useful!
posted by mek at 11:13 AM on March 20, 2010 [4 favorites]


So it sounds like these two groups of authors disagree on what it means for Natural Selection to be a good scientific theory. Fodor and Piattelli-Palmarini say it needs to distinguish between traits that are selected for by fitness and other "free-rider" traits that are correlated with the causal trait. Block and Kitcher think that this requirement is not necessary.

Given that there are experimental ways of determining the difference between the two kinds of traits, at least to the limitations of the general problem of causality, I don't think this argument has anything to do with the actual scientific power of the theory as description of the mechanism of how useful traits tend to occur in animals.
posted by demiurge at 11:34 AM on March 20, 2010


I'm baffled by the interpretation of natural selection done by Fodor and Piattelli-Palmarini. As they say in the main link: "All natural selection can do is respond to correlations between phenotypic traits and fitness. And that doesn’t help because, by assumption, if either of the confounded traits is correlated with fitness, so too is the other." And that first statement is totally true, so what is odd is that they treat it like a bad or unknown thing. That's one of the fascinating things about evolution, and the accidental secondary traits (e.g. having a hard shell makes it hard to be eaten, but it also allows one to leave the water for short periods of time without drying out) are surely part of a great deal of evolutionary history.

Perhaps they believe that biologists see it as their ultimate goal to endow every evolutionary step with a direct, specific cause. Or maybe they just like to use "cause" in a different way that the biologists care to, because anything with all the required nuances takes ages to say and doesn't convey more information. Or still yet, maybe they accidentally invert that a genetically heritable trait that causes differential fitness will be selected for to think that everything selected for comes from a trait that causes differential fitness. I don't know. But what they call "free-riding" is observed, so how they treat it as a flaw in the theory instead of a feature is absolutely beyond me.
posted by Schismatic at 11:57 AM on March 20, 2010


Darwin's theories have always been open to refinement, and the art of Philosophy often has useful contributions to make to the art of interpreting scientific data.

Unfortunately, Fodor and PP's publisher naturally selected a controversial block-buster title for their document in order to rake in book sales, at the expense of slotting them in with the cretinists and flat-earthers. Presume that the authors are aware that this lucrative commercial decision has dismal repercussions. Shame.
posted by ovvl at 12:01 PM on March 20, 2010


I can't believe Fordor and Piattelli-Palmarini even get taken seriously. They state flat out, "Prima facie, free riding is a counterexample to natural selection." A simple example shows that's not the case.

Assume I have the ability to throw a seven at the craps table if I concentrate hard enough. A side effect of this concentration is that my face flushes bright orange. The selection process (whether I win the throw or not) depends on my ability to roll the seven. The fact that it's entangled with flushing orange makes no difference. As long as the free ride trait doesn't interfere with the selection process, (which it might, for example, if the trait linked with extreme concentration was a muscle spasm), then it can be discounted.
posted by CheeseDigestsAll at 12:09 PM on March 20, 2010 [1 favorite]


As Block and Kitcher say, "Fodor and Piattelli-Palmarini are imposing a requirement on natural selection that no biologist or philosopher whom we know of has ever suggested." My reading is that Fodor and PP are suggesting that natural selection must and can only imply pan-adaptionism. But this is a ridiculous statement. Am I reading uncharitably, or is that their claim?
posted by PMdixon at 12:21 PM on March 20, 2010 [2 favorites]


Conversely, if the environment (pit bosses) starts selecting against orange-flushers, it's no longer a free-rider. The term is not an absolute or immutable one.
posted by ChurchHatesTucker at 12:26 PM on March 20, 2010


Hmm, what are the chances that two philosophers, who have at best a layman's knowledge of what happens in nature, can point out new flaws in a theory that has been tested heavily for 150 years now? Tested by biologists, moreover, who have to compete with other biologists for subsidies and grants ans honours, and have therefore everything to gain from pointing out the flaws in the each other's work?

Philosophy, once the Queen of science, really has become its nagging neighbour.
posted by ijsbrand at 12:31 PM on March 20, 2010 [5 favorites]


I guess my other question is, what exactly would a theory satisfying F+PP's demands look like?
posted by PMdixon at 12:32 PM on March 20, 2010


From the first linked article:

Apparently unshaken by withering criticism of Fodor’s earlier writings about evolutionary theory,
they write with complete assurance, confident that their limited understanding of biology suffices for their critical purpose.

Whether religious or secular, certainists (some call them fundamentalists), are a basic scourge to intellectual discourse. If there is controversy, there is uncertainty. To sweet all the uncertainty aside, to cherry pick the data points, is, well, fundamentally what bullshit is made of.
posted by Doohickie at 12:44 PM on March 20, 2010 [1 favorite]


To sweep.

Dang it, Matt... when will you ever give us a 30-second window to correct our posts?
posted by Doohickie at 12:45 PM on March 20, 2010


To ljsbrand and other people with the same line of thinking: I'm not really a fan of Fodor, but I don't think his and PP's supposed ignorance of biological facts has anything to do with their ability to criticize natural selection on non-empirical grounds.

It is important to realize that F & PP are not doing biology, they are doing philosophy. Biologists cannot do better than these guys at the task they're trying to do: just as philosophers should not be expected to practice biology, it's not the job of biologists to worry about philosophical issues either.
posted by k. at 12:46 PM on March 20, 2010 [3 favorites]


Also, a great example of a philosophical contribution to biology from a philosopher is The Case of the Female Orgasm: Bias in the Science of Evolution by Elizabeth Lloyd.
posted by k. at 12:49 PM on March 20, 2010


Intentionality is so beyond meaningless in the context of any scientific activity it makes me groan out loud.

But what about intensionalilty?!

This looks like an interesting discussion. It's too bad they decided to go with a "provocative" title; that'll just ugly the whole thing up.
posted by mr_roboto at 12:50 PM on March 20, 2010 [1 favorite]


I have a bachelor's degree in philosophy (a BS, appropriately enough) and, like mek, was driven absolutely up the walls by this sort of thing in undergrad. Philosophy students- and philosophers- tend to get the idea that we're qualified to make all kinds of judgments of other fields using our own tools and little else.
posted by Pope Guilty at 12:56 PM on March 20, 2010 [1 favorite]


ovvl: "slotting them in with the cretinists and flat-earthers"

prejudice against cretins is surely going to marginalize this document
posted by idiopath at 12:56 PM on March 20, 2010 [1 favorite]


[C]ertainists (some call them fundamentalists), are a basic scourge to intellectual discourse.

Well played, sir! Down with certainism!
posted by rdone at 1:05 PM on March 20, 2010 [1 favorite]


Yeah, that's a cute end-run around the absurdity of the phrase "fundamentalist atheists".
posted by Pope Guilty at 1:14 PM on March 20, 2010 [1 favorite]


Reading F&PP's response, I'm struck by this:

You and I can work out what caused what in all sorts of ways: we can use Mill’s method, or we can take the system of causes and effects apart and find out what mechanisms operate inside it, or we can ask the guy who built the thing (if somebody did) how it works, or we can look it up in the scientific journals, or we can ask Google, and so on and on. But natural selection can’t do any of these things. It can’t look inside, and it can’t run experiments, and it can’t contrive theories about internal structure, and it can’t consult the intentions of the builder. Doing any of those requires having a mind, and, by general consensus, natural selection doesn’t have one. All natural selection can do is respond to correlations between phenotypic traits and fitness.

But it can't "respond" either, because "it" is not an active agent. It's merely a phenomenological description of how traits are selected in evolutionary history (they're selected according to relative fitness). Is all of F&PP's argument based on assigning agency to some entity that they call "natural selection"? This seems profoundly misguided, and it certainly does not accurately represent biologists' view of the phenomenon.
posted by mr_roboto at 1:28 PM on March 20, 2010 [2 favorites]


But it can't "respond" either, because "it" is not an active agent. It's merely a phenomenological description of how traits are selected in evolutionary history (they're selected according to relative fitness).

I think they would completely agree with this if you changed the wording to "how some traits are selected in evolutionary history". That fact that natural selection doesn't explain all traits, and that close experimental examination of each instance is required to separate traits that were selected for from those that were merely selected, means that natural selection is not an adequate explanation for all evolution. I think this point is obvious, and I am surprised that it took a whole book to make this simple point.

It seems that a lot of the criticism is based on misinterpreting them as saying "natural selection never occurs". All they are saying is that the theory of natural selection is insufficient as a predictive theory of evolution, and that likely no other single theory could be either.

Maybe the book really does make other claims, but based on everything in the post (and the discussion Dumsnill linked to) that is all they are saying. It doesn't seem that many people are arguing with their actual thesis but instead a straw man version of it.
posted by DanielDManiel at 1:56 PM on March 20, 2010


This looks like an interesting discussion. It's too bad they decided to go with a "provocative" title; that'll just ugly the whole thing up.

Yeah, I completely agree. I think they are making a great critical point about Darwinism, but that the title made me initially think "OMG! Some stupid philosophers are trying to prove evolution wrong!". The title would be fine if the creation science vs evolution thing weren't such a hot button political issue right now. It's hard not to immediately think of shit like Darwin on Trial.
posted by DanielDManiel at 2:16 PM on March 20, 2010


How then could there be a sense in which one of the properties—being-a-melanic-moth—rather than the other—being-a-melanic-moth-and-smaller-than-Manhattan—caused the increased reproductive success?

We suggest that the question deserves a shrug.


Oohhhhh SNAP!

I was fortunate enough to take several classes with Prof. Kitcher (and a couple classes with his wife Patricia Kitcher, who is herself a very distinguished philosopher). That was a very elegant response to what amounts to a laughable argument in a *shocking!* jacket.

At least the stupid book led to a very interesting and engaging review. Well, to be fair, I guess it might have been the fact that the book has less words than Finnegan's Wake that yielded the review, and not the book's stupidity. what???

Finally:

There can be coextensive but distinct phenotypic properties, one (but not the other) of which is conducive to fitness, but which natural selection cannot distinguish. In such cases, natural selection cannot, as it were, tell the arches from the spandrels. That being so, adaptationist theories of evolution are unable, as a matter of principle, to do what they purport to do: explain the distribution of phenotypic traits in a population as a function of its history of selection for fitness.

Why are philosophers such terrible writers? I mean jesus f. christ.
posted by Lutoslawski at 2:18 PM on March 20, 2010 [1 favorite]


The funny part is that "natural selection can't explain all evolution" is WRITTEN RIGHT INTO DARWIN'S ORIGINAL BOOK.

Given that I can quickly read and comprehend the critique, but reading and comprehending the response to the critique took two or three attempts per paragraph, I'm going to side with the people who don't purposely obfuscate their points to look more intelligent.
posted by Scattercat at 2:26 PM on March 20, 2010


It seems that a lot of the criticism is based on misinterpreting them as saying "natural selection never occurs". All they are saying is that the theory of natural selection is insufficient as a predictive theory of evolution, and that likely no other single theory could be either.


If that's all that they are saying, then everyone knows it and the idea of them writing a book around it implies they don't have a clue that's the case. Either they are saying something obvious and never bothered to talk to a biologist, and perhaps bridge the enormous language gap they seem to employ intentionally, or are arguing something about the nature of what it means to call something a scientific theory at a level of precision irrelevant to working scientists. Or they are somewhere mistaken about notions of agency and what sort of object "natural selection" is. Considering that F&PP's main response to the critique of Block and Kitcher is that they are misreading their text, I think extracting a clear and simple point from their work is difficult at best.
posted by Schismatic at 3:08 PM on March 20, 2010




Hmm, what are the chances that two philosophers, who have at best a layman's knowledge of what happens in nature, can point out new flaws in a theory that has been tested heavily for 150 years now?

Philosophers don't have to have only a layman's knowledge of what happens in the sciences (or in nature, if you want to put it that way). And it seems to haave escaped your attention that the critics of Fodor and Piattelli-Palmarini are … philosophers!
posted by kenko at 4:33 PM on March 20, 2010 [2 favorites]


I like Jonathan Birch's comment on the review:
Imagine a population of bears. Half have thick white fur. Half have thin black fur. In a particularly harsh winter, the bears with thin black fur all die of the cold.

What caused the bears with thick white fur to be more successful? Was it the thickness of their fur, or the whiteness? Is there a determinate answer?

Yes, there is. It was the thickness. Thickness and whiteness are different physical properties of fur. Thickness protects against cold, whereas whiteness does not.
I agree with mr_roboto - F&PP's main problem seems to be that they think of natural selection as a separate agent within nature. Natural selection "sees" this population of bears, it decides somehow that fur thickness is the real cause of fitness, and it selects for thickness. Fur whiteness comes along for the ride.

But natural selection is not a separate agent that needs to distinguish possible causes. If the bears with thick white fur survived because their fur kept them warm, then fur thickness is what caused them to succeed and have offspring. This is natural selection in action. There was no separate agent that needed to decide anything.

Scientists may not know the real cause of increased fitness; they may have to do experiments to figure it out. But the fact that we don't always know the cause is not a problem for the theory of natural selection. There always was some cause of increased fitness, and that's the trait that was selected for - whether we know it or not.
posted by problemspace at 4:36 PM on March 20, 2010 [1 favorite]


Augh! I knew that this debate would make it to MeFi; why did it have to happen when my only Internet access for the next little while will be on an iPhone.

Kitcher and Block and Papineau and Leiter missed the boat. They did not understand the argument. You can see this in K&B's reply to Fodor's response... they are forced to develop an entirely new attack in light of Fodor's claim that, yes, the peppered moth is brown because that trait brought reproductive success to its ancestors. But Fodor has always accepted this. What's really interesting to me is that I saw Fodor and Kitcher debate live back in November; Kitcher presented Fodor with the peppered moth case, and Fodor made exactly the claim that now force K&B to revise their strategy. I have no idea why Kitcher wrote a review based on the premise that Fodor thought otherwise.

I tried to explain Fodor's argument in the previous thread; I think I can do a better job of it this time. When I have access to a keyboard I'll give it a shot. Some of the first comments in response to the K&B review are pretty good (but the thread degenerates really quickly).
posted by painquale at 5:03 PM on March 20, 2010


Darwinian Climategate: quibbling about details of interpretation, to a stunned audience.
posted by ovvl at 6:40 PM on March 20, 2010


Well, I've read the review, the rebuttal and the rejoinder, and some of the associated comments (including painquale's comment in the previous thread). I'm confused. In particular, I don't understand F&PP's characterisation of the propositions of natural selection (in the rebuttal):
• There is random variation of phenotypic traits.

• There is some ecological variable that is sensitive to the strength of the correlation of such traits with fitness.

• There is some mechanism that alters the relative frequency of the trait in the population so that, all else equal, it varies with the strength of the correlation between the trait and fitness.
But isn't this "mechanism" implicit in the concept of fitness? The word "fitness" as used here just means "ability to pass on traits to later generations". And, over time, differential levels of fitness will necessarily alter the relative frequency of the traits in the population. If you accept the concept of "fitness" I don't see why you'd need a separate mechanism. Is that whole third point tautological?

In the paragraph before this they talk about how "thin" the theory of natural selection is. Seems to me it's much thinner than they think. Maybe you could have meaningfully described natural selection as a theory in Darwin's time, when it was sufficiently different to the other ideas around to be able to disprove them with testable predictions. Now, though, it's less of a theory than an idea which sits on top of a whole range of much more concrete theories (models of gene expression at the molecular level, epidemiology, analysis of virulence in microorganisms etc etc) which don't need to rely on just-so stories or unexplainable mechanisms in order to be useful.

If I'm misunderstanding F&PP here, I think Block and Kitcher are too.

Painquale, in your comment on the previous thread you said:
So, we're left with a choice: either argue for massively context-sensitive laws of evolution, deny that the various special sciences are autonomous from physics and be a strong reductionist, posit some other understanding of what the sciences are doing other than describing laws, or admit that 'selection for' claims are like 'what if?' stories in history.
Now, I'm not an evolutionary biologist, but as far as I can tell the way to go is the massively context-sensitive laws of evolution. Put them all together and you might find that natural selection is useful as a central organising principle, but I doubt it would look anything like the "mechanism" they propose.

P.S. "intensionalism" wtf?
posted by A Thousand Baited Hooks at 7:07 PM on March 20, 2010


I still don't really understand F&PP's argument, though I have been trying since Fodor published his Pig have Wings article. But the title of their book and the fact that they use the creationist coined "Darwinist" to label their opponents leads me to be sceptical about their motivations.

I am sure they are getting one thing wrong though. They seem to think that natural selection works on phenotypes, when in fact the medium of selection is the genotype (more specifically variance in gene frequencies in populations). This is a common mistake, particularly among science writers, (see here for an example).

If phenotype has no connection to the genotype then it has no effect on evolution. If there was some random environmental variable that turned some moths black and some white, it would change the survival rates of individual moths but would have no effect on the populations genotype and therefore the rate of black to white moths would be the same. On the other hand, changes in genotype can affect evolution while having no impact on phenotype. If a gene causes a change in the ability of DNA to copy itself for instance.

If we keep in mind the centrality of the genotype, this paragraph from F&PP looks strange,

It sounds like it ought to work.

But it doesn’t. A way to see that it doesn’t (not by any means the only way) is to consider confounded (linked) phenotypic traits, one but not the other of which is fitness-enhancing. Both traits are then correlated with fitness, so both should count as adaptations according to the formulation of natural selection given above. But only one of them is a cause of selection, so only one of them is an adaptation, and, though both are selected, only one is selected-for. Thus the free-rider problem. Prima facie, free riding is a counterexample to natural selection.


Linked phenotype traits are irrelevant, since what is evolutionarily interesting is the change in the genome. After you have found what gene has changed then you look at what traits it affects and how these might have affected fitness, but to look at phenotype traits before considering genotypic changes would be doing biology backwards.

So in the moth example you wouldn't have to worry about confounded traits because you would see that the gene that changed was linked to coloration and not linked to some chameleon like ability to blend in to the environment.

Maybe Fodor and Piattelli-Palmarini don't dispute this, but if that is true I don't see why they wrote a book called Why Darwin was Wrong.
posted by afu at 9:43 PM on March 20, 2010


This looks aimed at Dawkins' and others' positions (gene-centered competition, very strong adaptationism) rather than the general theory of evolution. Stephen Jay Gould weighed in with similar objections, right down to the idea that this level of reduction ends up kind of begging for a purposeful context, and leads you back to telling stupid just-so stories. So this is not just something stupid philosophers came up with because they're so stupid, stupid. On the other hand, I see no reason why it ought to be written, since the arguments are not new.
posted by mobunited at 10:37 PM on March 20, 2010


I was hastily branded "reductionist" in a philosophy of science class because I had the audacity to believe that my DNA might have something to do with my appearance and condition. That, in a nutshell, is where this all comes from. Since there is no one gene for smart, evolution can't choose for smartness. And it's absolutely true if your population is small, terribly homogenous or you only consider three or four generations.

And then someone does an experiment and shows that those rider genes that don't matter don't matter. Until they do.

I would love to hear Fodor and P-P's take on astronomy. Because I'm pretty confident that, despite not making any stars in the lab, astronomers are scientists and, if you tell them the temperature and mass of a star today, they can tell you what it's going to look like in a couple million years.
posted by Kid Charlemagne at 11:16 PM on March 20, 2010 [1 favorite]


One doesn't need massively context-sensitive laws of evolution. One needs to find out about the nature of the things governed by the laws and the terms in the laws. Moline's "Provided Nothing External Interferes" is good on this.

It also doesn't seem to me that we're in such a bad epistemological spot as painquale's reconstruction of Fodor's argument has us. Say I think that the heart was selected for in dogs (one would think actually in dog-precursors but obviously the heart has been retained, so whatever) in virtue of its blood pumping and not in virtue of its redness. One way to think of this would be to say: that previous dog hearts pumped blood (and that this was heritable) enabled dogs to reproduce and have successful offspring; that previous dog hearts were red did not. Here's an experiment you can do right now: paralyze a dog's heart. It's still red, innit? But it doesn't do so hot. (Admittedly, getting the heart to be another color would be trickier, especially when it comes to the heritability side of things, but supposing one had an artificial dog heart it would obviously not need to be red.) Or take the polar bear example above—does Fodor really want to say that answering the question what would have happened as far as survival in the winter if the bears had, not thick warm fur, but thin fur (with the same insulation/warmth provision per unit thickness as the fur they actually did have), is as hard as answering the question what would have happened had Franz Ferdinand not been shot by Princip? Really? It's not as if we're interested in answers such as: if they had thin fur then concerned animal lovers would have covered them in wool blankets. That might happen in given generation, sure, but it's not the way things happen in the way of things.

(Or does he mean that we have no idea what else would have led up to their having that thin fur—perhaps they would have had entirely different optimal body temperatures! but that would be to grant the point that in the actual bears the fur is there for temperature-related reasons.)
posted by kenko at 2:00 AM on March 21, 2010 [2 favorites]


I've read through Fodor's paper, Against Darwinism*, linked to in the old thread, and I think I understand his argument now and why I disagree with him. Basically it come down to a disagreement on how to determine what qualifies as an acceptable part of an intensional statement, and more fundementally what are the "laws" of evolution are.

I'll steal this from painquale because he does such a good job of stating it,

But Fodor's argument is not epistemological, as he is quick to point out. It's not just a problem of our lacking sufficient information. It has to do with the sort of information that we could possibly appeal to. Suppose that we knew all sorts of stuff about the state of the world at the beginning of the last century. To go about answering our counterfactual question about Franz Ferdinand's survival, we might appeal to the psychology of the individuals involved and psychological laws; we might appeal to physics and physical laws; maybe even economic laws could help us out, if such things exist. Fodor denies that there is an autonomous domain of historical laws that would answer our question for us. Likewise, if we wanted to know whether wings would have evolved had air density been different, we wouldn't appeal to laws of evolution, we'd appeal to laws of physics. Are there laws of evolution? They'd have to be massively context sensitive (as Fodor writes, it's obvious that there are no context-independent laws like "being big increases fitness".) And once you introduce enough context-sensitivity, and laws move from the general to the extremely specific, it begins to look like you don't have a law at all.

But we do have specifically biological knowledge to appeal to, specifically we appeal to changes in gene frequencies and the gene expression. No need to go down to quantum physics as Fodor suggests. We can determine the correct outcome of a "Selection for" argument, because we can look at the gene expression, so there is no "intensional" problem. Yes this is very difficult, but it is an epistemological problem, not a metaphysical one.

Furthermore, Fodor has a bizarre idea of what would constitute evolutionary laws,

"So, then, suppose there is a law from which it follows that t1s are selected in
competitions with t2s (where being t1 and being t2 are traits that have intentional properties)."


I hate to keep saying this but, evolutionary laws are not about traits, they are about changes in gene frequences in populations. And contary fodor,

"For example, it’s obvious that no trait could be adaptive across the board."



It isn't hard to come up with a universal law that deals genetics e.g.: All other things being equal, any trait that reduces the reliability of the replication of genetic material will be selected against. Conversely, any trait that increases the reliability of genetic replication will be selected for.

There are lots of other laws that are much more complicated. Population genetics wouldn't work if this wasn't true.

This is the same mistake he makes when taking about Darwin in this article,

"The main thing Darwin had in mind with natural selection was to come up with a theory that answers the question, "Why are certain traits there?" Why do people have hair on their heads? Why do both eyes have the same color? Why does dark hair go with dark eyes?"

No, the main thing Darwin was concerned with was the orgin of species, (thus the name of his book), the concept of species has become much more nebulous since Darwin's time, but the closest modern concept we have for "species" that works pretty much every where is "isolated genetic population," and Darwin was smart to signal out species at that time, as what happens on the population level is really what is important in evolution and species is a pretty good marker of an "isolated genetic population" for the organisms Darwin was studying.

I used to respect Fodor but the more I read of him on evolution, the more I think he has become convinced of his own smartness at the expense of actually having to understand what he is criticizing.






*(Read the footnotes in that piece too, it shows what a cherry picking ass fodor is and that his reading in evolution exists completely of what is offered up by Barnes and Noble, the fact that he never mentions Woese is rather shocking).
posted by afu at 5:30 AM on March 21, 2010


Thomas Nagel in his book The View from Nowhere raises the same criticisms against natural selection that Fodor and Piattelli-Palmarini, as far as I can tell from the articles, are making in their book. Nagel's argument spanned only a couple pages, and my impression was that its writing was prompted by his frustration at seeing evolutionary arguments run amok in in every corner of academic discourse. The example of correlated traits--humans being able to use tools, and humans being able to contemplate the nature of their own existence--is, I think, a much more productive example to consider over moths of different colors.

My impression is that Fodor and Piattelli-Palmarini are motivated by a different frustration, which is that natural selection has ironically been raised up to the be considered by many to be scientific gospel, which would lead people to consider the theory unfalsifiable, which then, according to Popper, would make it unscientific.
posted by seliopou at 6:55 AM on March 21, 2010


You can't blame Fodor: ape males display their strength to attract females, and attacking the prevailing orthodoxy is a good display. Whether he's correct is irrelevant. [/sociobiologist]
posted by alasdair at 3:37 PM on March 21, 2010 [2 favorites]


So, judging from the rest of this conversation, the answer to my question above is "yes"?
posted by PMdixon at 4:34 PM on March 21, 2010


It's more like... Fodor's conception of metaphysics imposes a requirement on natural selection which science does not, which requires a sophisticated understanding of metaphysical philosophy to argue against in any way other than a shrug.

Dennett's response is worth a read, as he says everything I would like to, better. Via painquale in the old thread.
posted by mek at 5:43 PM on March 21, 2010 [2 favorites]


Oh my. That Dennett paper is something else.
posted by PMdixon at 5:57 PM on March 21, 2010


Sorry guys, I've been trying to find time today to respond to this, but I've been under the gun making up for a squandered spring break. In the meantime, let me just link to a blog post by a friend of mine. Fodor has a few ways of presenting his argument(s), and one of them involves an analogy to Chomsky's demolition of Skinnerian behaviorism. The blog post explains how the analogy is supposed to work. It's the clearest exposition of Fodor's argument I've seen. It's certainly clearer than anything by Fodor. I know a lot of philosophers who have been puzzled by Fodor's argument, but after reading this explanation, have been all, "Aha! Now I get it!"
posted by painquale at 6:23 PM on March 21, 2010


Well, I want to quickly respond to a few other comments:

PMDixon: "My reading is that Fodor and PP are suggesting that natural selection must and can only imply pan-adaptionism. "

mobunited: "This looks aimed at Dawkins' and others' positions (gene-centered competition, very strong adaptationism) rather than the general theory of evolution. Stephen Jay Gould weighed in with similar objections, right down to the idea that this level of reduction ends up kind of begging for a purposeful context, and leads you back to telling stupid just-so stories."

Fodor cites Gould approvingly, but his argument against Darwin does not boil down to just the same old Gould v. Dawkins debate. His arguments apply just as much to Gould as they do to Dawkins. Even if you think that exactly one thing on earth has evolved by natural selection and everything else is a spandrel or a product of genetic drift, Fodor will still take issue with you, because you would not have any ability to ground that one instance of selection-for.
posted by painquale at 6:40 PM on March 21, 2010 [1 favorite]


Fodor's argument clearly hinges on the "ubiquity of local correlation," ie. there are an infinite number of correlations available in what the rat experiences, or what nature selects for, which makes it impossible for a theory to predict anything. This is clearly false - while there are an infinite number of properties available, what the rat actually gives a shit about, or what natural selection finds relevant, are finite!

Isn't this built into the whole idea of natural selection? While there are an infinite number of possible variations, only the ones which actually occur and that increase fitness will be selected for?

The "brown snow" example really cinches it for me. This is just mindblowingly stupid:

Now we change the color of the snow to brown. Suppose the brownness phenotype now goes to fixation. It looks like we’ve falsified adaptationism, for it is not true after all that whiteness increases fitness.

We could put all the bears into space and they would all die so clearly none of them have any fitness!!
posted by mek at 6:50 PM on March 21, 2010 [2 favorites]


Yeah, that's a cute end-run around the absurdity of the phrase "fundamentalist atheists".

I have this hope that if I use the term Certainist on enough online forums, the term will grow its own legs and become the term that generally describes the unreasonable folks at the opposite ends of the spectrum. I kind of like the idea of marginalizing them by associating them with those they seek to marginalize.
posted by Doohickie at 6:37 AM on March 22, 2010


I discussed part of this debate today in a class on Nancy Cartwright's essay: "The Truth doesn't explain much." (It's available in her book How the Laws of Physics Lie.)

Basically, the claim is that natural selection only qualifies as an approach or a method of analysis, not a "law." As a model for understanding theories, it only ever supplied ceteris paribus accounts: "all things being equal" isn't a law, just an explanation. A claim about predator-prey populations doesn't have predictive power outside of its specific context.

I'm just not sure why would care. Fodor suggests in the bloggingheads that biologists know a lot of stuff and attribute that knowledge to something they call 'natural selection,' when in fact they're wrong: they know a lot of stuff because of the research and analysis that they and their compatriots do. But then he calls it 'gossip' which is a pretty strongly negative evaluation. It seems we ought to be satisified with the models and ceteris paribus explanations, and indeed it's not clear to me that Darwin would have been dissatisfied with such models and explanations.
posted by anotherpanacea at 8:55 AM on March 22, 2010


Good comment, anotherpanacea! I think you're right when you say that natural selection is a method of analysis, not a law, and I think that A Thousand Baited Hooks is right when he says that it's more of an idea than a theory. But the question then is: what is an approach or a method or an idea in this context? I asked Fodor once if he thought that natural selection might have an heuristic role to play in helping scientists come up with correct theories, in a context of discovery rather than a context of justification, and he said, sure but getting a good night's sleep or eating cream cheese for breakfast or something like that might be valuable from the POV of the context of discovery. That's obviously too thin a role for natural selection. So there's a neat question about the role that natural selection is playing in developing accounts of natural history, and what it takes to be a "method" or "approach" of science. (My personal theory is that statements making claims about natural selection have the same status as statements making mathematical claims. Explaining why mathematics applies to empirical theories is also not easy.)

It's true that Fodor's claim doesn't really affect scientists too much. But if there is no theory of selection-for, then that constrains other theories that have been offered, at least in philosophy. For example, if he's right, then you're arguably not going to be able to develop a teleosemantic account of representation or mental content. (That's what Fodor really cares about, I think. It's how he got on this whole topic in the first place.)

(Incidentally, ap, Fodor does think that laws can have ceteris paribus clauses. All psychological laws have them. Laws of evolution are purportedly so context-sensitive, however, that they can't be handled merely with ceteris paribus clauses.)
posted by painquale at 10:38 AM on March 22, 2010


After reading a bit, I think I'm getting a better sense of Fodor's argument. Very briefly, he seems to be saying that since every trait has an infinite number of correlated traits, natural selection cannot be used as a theoretical construct to distinguish which trait will be selected under a given set of evolutionary pressures. Such a distinction can only be made on a post hoc basis by biologists analyzing evolutionary history, and their conclusions are always trivially consistent with natural selection.

It seems that this argument can be defeated using a form of gene selection. In other words, selection is not between infinite sets of correlated traits, but between individual genes or finite sets of genes. Each gene, then, may have an associated infinite set of traits, but since selection occurs on a per-gene basis it's possible to make a theoretical distinction between the presence or absence of each specific gene based on our knowledge of how genotype expresses as phenotype.

Looking at the moth example, while trait-based selection may be unable to distinguish between the traits "dark" and "dark and smaller than Manhattan", this distinction disappears in the genome, since there is no genetic option for "dark and larger than Manhattan". The traits are trivially coupled in the genome, and selection can only occur on the gene that expresses the dark pigment.
posted by mr_roboto at 10:46 AM on March 22, 2010 [2 favorites]


it only ever supplied ceteris paribus accounts:

Yes, it's amazing to me that some folks miss the underlying requirement of causal statements that assume "all else being equal". That is a colloquial way of stating the counterfactual definition of causality. Unless they state an alternative, this is presumably what is meant when they speak of "laws".
posted by Mental Wimp at 11:26 AM on March 22, 2010


Put in another way, we don't expect to have to state ceteris paribus when we state Newton's laws of motion, but we certainly do not expect them to apply as other variables are changed willy nilly. To state "A body in motion will tend to stay in motion if no other force is applied to change its motion in the interim" just seems silly. Also note that theoretically everything is driven by the laws of physics, but in sufficiently complex systems, applying the laws of physics becomes increasingly intractable with present technology. Within ecosystems, such complexity is the rule. Why we would expect the laws of natural selection to be any less challenging is beyond me.
posted by Mental Wimp at 11:34 AM on March 22, 2010


Well, in a single example, natural selection is indeed that challenging. Given one creature that successfully reproduces (or dies beforehand) there are millions of genotypes to consider, as well as thousands of causally relevant extraneous variables in the environment. It's impossibly complex and any attempt at explanation would be a just-so story. Hell, sometimes survival is a matter of luck rather than fitness: in fact it's always some combination of the two.

But none of that matters because evolution operates on a different scale. The sample sizes tend to be on the large side... it's the trend that matters, not the individual case.
posted by mek at 12:19 PM on March 22, 2010


Laws of evolution are purportedly so context-sensitive, however, that they can't be handled merely with ceteris paribus clauses.

Since you're my local expert, painquale, could you say a bit more about this? It seems like Fodor is kind of Cartwright's "Scarcity of Laws" + intensionality problem in selection-for = Darwin Wrong! Specifically, is it the case that Fodor is emphasizing trait-selection versus gene-selection?

Leiter yesterday had a post from Sober mentioning that Fodor was outright contradicting himself on selection-for, so now his honesty is up in the air, as well.
posted by anotherpanacea at 12:25 PM on March 22, 2010


Sophist nonsense. Darwin is essentially good sense, written down. Some folks think they look smarter by disagreeing with such things. It's nothing but intellectual masturbation. Some folks get the real thing, others only fantasize.
posted by Goofyy at 5:44 AM on March 23, 2010 [1 favorite]


Since you're my local expert, painquale, could you say a bit more about this? It seems like Fodor is kind of Cartwright's "Scarcity of Laws" + intensionality problem in selection-for = Darwin Wrong! Specifically, is it the case that Fodor is emphasizing trait-selection versus gene-selection?

I definitely wish I could say more about ceteris paribus laws or provisos, but I can't... I don't get them at all. I recognize why they seem to be necessary, but I can't tell why they don't just say "If A happens then B happens, except when it doesn't." Fodor had a debate with Schiffer about them back in the nineties, I think... I remember not understanding it. I might recommend to my phil. sci. reading group that we do some papers on CP laws.

Put in another way, we don't expect to have to state ceteris paribus when we state Newton's laws of motion, but we certainly do not expect them to apply as other variables are changed willy nilly. To state "A body in motion will tend to stay in motion if no other force is applied to change its motion in the interim" just seems silly.

That doesn't seem silly at all... it seems totally necessary. You'll note that Newton's laws updated for relativistic physics incorporates the clause that you find unnecessary: "An object continues upon an inertial path unless acted upon by a force." Anyway, the "if no other force is applied to change its motion in the interim" clause isn't a ceteris paribus clause, because we can write it down. What makes ceteris paribus clauses challenging is that they are supposed to capture all the things that could go wrong that we can't write down, just because there are an indefinite number of ways for a law to go wrong. If we can write them down, then we should.

Specifically, is it the case that Fodor is emphasizing trait-selection versus gene-selection?


Fodor is definitely emphasizing trait selection. Note that if genes are individuated by their phenotypic expression (which they usually are, I think?), then he's against gene selection too. But it definitely seems like an area to push him on, like anotherpanacea and mr_roboto are doing) Still, if you want to claim that the units of selection are particular strings of base pairs (this is unorthodox, I think?), then there probably still won't be laws of evolution, because whether any particular sequence of nucleotides is fitness-inducing or not depends on all sorts of other stuff going on in the genome, making it too context-sensitive to be lawful. Laws of the form "GATTCAGCTTA in such-and-such a position increases reproductive success, ceteris paribus" look pretty implausible... just as implausible as "being big increases reproductive success, ceteris paribus."

Leiter yesterday had a post from Sober mentioning that Fodor was outright contradicting himself on selection-for, so now his honesty is up in the air, as well.

I don't see that Fodor contradicts himself anywhere in that exchange. I think you mean that Sober's citation of Fodor's sentence "natural selection cannot distinguish between coextensive phenotypic traits" in his book. But he still agrees with this. Particular theories in biology can support counterfactuals, but these will not be general theories of natural selection. No theory of natural selection can distinguish coextensive traits.

It's nice that people (Sober, Kitcher, etc.) are finally starting to get Fodor's argument and are engaging with it. I would honestly like to know how to respond to it, because while I know that there needs to be something wrong, I cannot say what is, and I haven't seen any really convincing refutations. I've gotta say that I'm finding the blogosphere reaction to this book really frustrating... not because I agree with Fodor, but because the hostility and smug certitude from evolutionists who clearly don't understand the argument is making me feel like the pro-evolution camp I identify with is entirely anti-intellectual.
posted by painquale at 11:03 AM on March 23, 2010


I'm going to be chewing on this for a while, especially because it seems like it might be trivially true that Darwin was wrong to emphasize traits because we now understand that selection-for targets genes which drastically reduces the scope of the intension.

On the self-contradiction matter: his appearance on bloggingheads was much more amiable because when Sober pushed him, he gave up on the claim that natural selection can't distinguish free-riders from fitness-enhancers.

On CP laws: there's a similar problem that I run into in economic forecasting where a hedge fund manager will engage in backdating and make it seem like their stategy is a guaranteed win.... Out-of-sample testing basically proposes that a method ought not to be generated purely inductively from past data, but ought to be equally effective on eyes-unseen markets if it is to qualify as identifying predictions. This isn't a universal view in finance, but it's close. I wonder if maybe Fodor is charging biologists with 'data-snooping.'
posted by anotherpanacea at 3:05 PM on March 23, 2010


Still, if you want to claim that the units of selection are particular strings of base pairs (this is unorthodox, I think?)

It's essentially Dawkins' view circa 1974. A more accurate picture of the contemporary view would include networks of genes, regulatory elements, and epigenetic considerations. But it's all conceptually equivalent for the purposes of this argument, I think.

Laws of the form "GATTCAGCTTA in such-and-such a position increases reproductive success, ceteris paribus" look pretty implausible... just as implausible as "being big increases reproductive success, ceteris paribus."

I don't think this is the form of law that adaptionists are trying to...uh...adopt, however. I believe a more accurate statement of a law of natural selection would be along the lines of "if a heritable element increases reproductive success, it will become fixed in a population." You can take or leave the ceteris paribus, I suppose.

This entire argument seems focused on that initial "if"--specifically confusion resulting from the mechanism by which a trait leads to increased reproductive success in the context of infinite coextensive traits. This mechanism isn't what natural selection is about: the neodarwinist theory of natural selection is that genes become fixed in a population when those genes confer a relative fitness advantage. It doesn't matter if the gene is interpreted as leading to "white" or "color of snow" or "color of paper" or "existing in the geographical location in which all white bears and only white bears exist". This is all essentially linguistic gaming that the gene doesn't care about.
posted by mr_roboto at 12:18 AM on March 24, 2010 [1 favorite]


(Incidentally, ap, Fodor does think that laws can have ceteris paribus clauses. All psychological laws have them. Laws of evolution are purportedly so context-sensitive, however, that they can't be handled merely with ceteris paribus clauses.)

What psychological laws are ceteris paribus?

I've gotta say that I'm finding the blogosphere reaction to this book really frustrating... not because I agree with Fodor, but because the hostility and smug certitude from evolutionists who clearly don't understand the argument is making me feel like the pro-evolution camp I identify with is entirely anti-intellectual.

This probably wouldn't have happened if he hadn't chosen an intentionally chosen an inflammatory title for his book. Fodor was looking to pick a fight with biologists and if he wanted them to understand his point he should have done a better job explaining it.
posted by afu at 12:26 AM on March 24, 2010 [1 favorite]


Still, if you want to claim that the units of selection are particular strings of base pairs (this is unorthodox, I think?), then there probably still won't be laws of evolution, because whether any particular sequence of nucleotides is fitness-inducing or not depends on all sorts of other stuff going on in the genome, making it too context-sensitive to be lawful.

The definition a gene in the functional (and evolutionary) sense is one of the biggest problems in Biology at the moment. Since different areas of genes interact in non-understood ways (e.g non-coding interacting with coding sections), a particular string of base pairs is not always going to do the exact same thing. However this appears to me to be an empirical one, and once your have a decent definition of gene, then you know what trait it affects and Fodor's selection for argument seems to crumble.
posted by afu at 12:51 AM on March 24, 2010


The definition a gene in the functional (and evolutionary) sense is one of the biggest problems in Biology at the moment. Since different areas of genes interact in non-understood ways (e.g non-coding interacting with coding sections), a particular string of base pairs is not always going to do the exact same thing.

I took the original comment to refer to the entire genome and any germ-line epigenomic coding. That would take care of gene x gene interactions and epigenetic effects. However, your criticism is correct if one focuses more locally on a subset of the genome. Adaptation of phenotype to the environment rests on the capabilities of this genome (+epigenetics), so this is the object and subject of evolutionary pressures.
posted by Mental Wimp at 6:51 AM on March 24, 2010


« Older We Are On the Verge of a Shift to Biosphere...   |   ♫ I can't do this all on my own... ♫ Newer »


This thread has been archived and is closed to new comments