Skip to content

At last, a proper review of Fodor and Piatelli-Palmerini

Peter Godfrey-Smith reviews What Darwin Got Wrong in the London Review of Books, and finally the review matches the book I am reading. PGS is usually right on everything, so read this one. It is critical, but doesn’t suppose that FAPP have made grade school level errors, like so many of the others do, nor that there is an orthodox view to be defended.

PGS focuses on the most interesting part of the book; the analogy they draw between Skinnerian behaviourism (or should that be “behaviorism”, given their American source?) and Darwinian evolution, and he points out that the error they make is one of seeing selection as an agent (which they criticise earlier). He gives a typically lucid account of selection, one I will give to my students next semester.

7 Comments

  1. Sam C Sam C

    A “proper review”? You mean by a philosopher? ;-(

    Teasing aside, yes, an interesting review, one that doesn’t tempt me to waste money on FPP, especially as the book seems to reproduce all the garbled and incoherent thinking that Fodor’s earlier LRB piece on flying pigs showcased.

    Is there a Dunning-Kruger effect here? It’s clear that Fodor doesn’t “get” modern evolutionary theory, but it’s also clear that he doesn’t realise that he doesn’t get it and he tries to bat away criticism, as though it’s the critics who don’t understand his arguments rather than him who doesn’t understand the core ideas of evolution by natural selection (which he doesn’t).

    When I was taught college-level evolutionary biology decades ago, we didn’t pay any more attention to Darwin than physics students pay to Newton: each was the originator of ideas that laid much of the foundations for a modern science, but we concentrated on the modern science, not on the history of those ideas or flaws in their original conception. When critics try to diss Darwin, it really highlights just how much of what Darwin did was solid and correct, even when he was working with no knowledge of the mechanisms of inheritance. Does Fodor perhaps resent the status of these intellectual giants?

  2. Chris E Chris E

    Thanks, John—it is useful to read the problem well and respectfully expressed. Other than providing some further background material on intension and counterfactuals, though, this seemed to me to line right up with E Sober’s response on Bloggingheads.tv (which was not, obviously, a “review”), right?

  3. I think Samir Okasha’s review, earlier published in the TLS, is as lucid and well-aimed as Godfrey-Smith’s. The definitive review, I predict, will be Sober’s forthcoming in Philosophy of Science (available on his website).

  4. Neil Neil

    I read the review yesterday. I guess it depends on what kind of errors
    people make on grade school, but I would have thought the error attributed must be pretty close to counting.

  5. I appreciate the tone of this review and I agree with the reviewer’s treatment of FPP’s second major argument – that it doesn’t matter for evolution which co-extensive trait is selected for, as long as one of them is.

    However, the reviewer misses the point in responding to FPP’s 1st major argument. FPP list a number of new theories, and some older ones, that demonstrate that natural selection is not the primary agent of evolution, when in fact it is generally asserted by mainstream biologists that it is, along with genetic drift and perhaps some other lesser factors. Rather, FPP show that a number of endogenous and exogenous agents are probably more important than natural selection in evolution.

    Then, in their 2nd and 3rd arguments they attempt to show that NS is in fact an empty theory. As I just mentioned, I agree that their 2nd argument fails.

    But their 3rd argument is much more powerful and is not even mentioned by this reviewer. This is the argument that NS rests on a number of circular or tautological definitions. John has recently posted a number of thoughtful comments about why the circularity/tautology arguments with respect to NS are wrong. I think he’s wrong on this and I’m disappointed that the reviewer didn’t address these powerful critiques.

    Here’s the short summary: anyway we define natural selection (‘survival of the fittest,’ ‘differential reproduction,’ etc.) these all end of being tautologous statements, and thus empty of any explanatory power. Who survives? The fittest. Who are the fittest? Those who survive. And every other attempt to describe natural selection falls into the same set of problems. It’s a disturbing conclusion, but one that is inescapable, despite John’s worthy efforts to escape it.

    In the last analysis, I agree with FPP’s main conclusion: NS is an empty theory. There is no natural selection, there is only natural history, which is just “one damned thing after another.” it is what it is and it will be what it will be. And this is all we can say on the macro level about evolution. We, can, on a more micro level, use NS as a helpful research agenda or metaphysical assumption, but as a true theory of evolution it is empty.

    • John S. Wilkins John S. Wilkins

      I’m about to travel for a few days so I can’t respond in detail. I am also, when time permits, reading FAPP, and aim to review it, but I agree: the PNS (principle) is a tautology. Moreover, FAPP’s critique fails because I do not think there is a sensible distinction between selection of and selection for. Instead, I think there is just sorting by selection processes, and the end result is whatever got sorted. It’s a post hoc thing. FAPP’s problem is not that they think it is a tautology, because I agree, nor that it has no prior objects subjected to selection, because I think that is also true ? the “objects” of selection are whatever satisfies the Price Equation ? it is that they have a false understanding of what counts as a theory. It’s both very physicistic and also something that they derive from philosophy of mind and language. In short, I think it is question begging.

      But I disagree that the theory of natural selection is empty. The principle may be, formally – it’s just the Price Equation or, depending on how you formulate it, the Breeder’s Equation (see Peter Godfrey-Smith’s recent book). But the theory is an elaborated series of models with applications, a mathematical construct that has interpretations. This is true of every theory – take the formalisms and you have basically a tautology, but take the formalism and the interpretation and you say something about the world.

      So it’s critical to distinguish between the principle and the theory.

      • John, thanks for the quick reply. I’m glad you agree with the first few points. And I think you’ll find, if we continue this discussion, that you may in fact agree with my further points (or not).

        I’ve read your posts on the tautology argument in detail and I look forward to seeing your paper. You describe ENS as a “schematic explanation” in the form (to abbreviate) of: PNS + O + I = ENS.

        If you and I both agree that PNS is tautological, as we do, you should just drop it from ENS. As a tautology, it cannot explain and it cannot predict. It’s simply an empty statement that performs no work. You could drop it from your ENS and simplify your model.

        However, if we now have the following form: O + I = ENS, we don’t really have a theory of evolution by natural selection. Instead, we have a theory of evolution and we must decide BASED ON EACH EMPIRICAL EXAMINATION what led to the observed changes at issue. In other words, natural selection becomes natural history, as F&P-P describe.

        We can certainly make examinations in controlled environments, or even examine “natural experiments” that occur without our intervention, and make some judgements about the utility of certain traits in certain circumstances. We may then make some limited predictions based on these observations. But these constitute the O and I in your schematic explanation, not the PNS.

        So, without the PNS in your model, I don’t think it’s accurate to call it the ENS. Rather, we should I think adopt a more open and modest approach to evolutionary theory, acknowledging that we have no idea, in any a priori or general sense, what leads to change in each circumstance. And this is far more akin to natural history than natural selection.

        I look forward to your thoughts.

Comments are closed.

Optimized by Optimole