Wednesday, January 03, 2007

A Confused Philosopher

Darwinism and Its Discontents, by Michael Ruse, Cambridge University Press (2006)

Ruse defines Darwinism as the idea that natural selection is the chief causal process behind all organisms (p.2). He identifies a whole list of people who oppose Darwinism. Some of these are creationists—this book is not about them.

The main "discontents," according to Ruse, are misguided social scientists with their irrational fear of genetic determinism; philosophers who "can't handle the awful truth;" and evolutionary biologists whose objections "cannot be grounded purely in theory or evidence" (p.3). Many of discontent evolutionary biologists are (gasp!) Marxists.

I am one of those scientists who question Darwinism, so this book is all about me.

What does Michael Ruse have to say about us "discontents?"
At the risk of damning myself in the eyes of sound scholarship and of God, let me be categorical. All of the critics of Darwinism are deeply mistaken.
Wrong. It is Michael Ruse who is mistaken and this damn book is full of sloppy scholarship.

Chapters 1-4 cover the basic facts of evolution. Ruse establishes the important contribution of Darwin in discovering natural selection. He points out that natural selection is the "single best idea anyone ever had" (Dennett, 195). I agree.

The "fact" of evolution is explained and the history of life is briefly described. None of this is controversial as far as scientists are concerned but Ruse is setting the stage for the most important part of the book.

Before continuing, it's worth pointing out one of the major failings of the book: the lack of any solid definition of evolution. It seems clear that Ruse is confused about the difference between evolution and one of the main mechanisms of evolution, namely, natural selection. This confusion haunts the last part of the book and makes it very difficult for Ruse to come to grips with the ideas of the "discontents."

Chapter 5 ("The Cause of Evolution") is all about natural selection. Ruse builds the case for natural selection using all the old examples that we are familiar with. Only in Chapter 6 ("Limitations and Restrictions") does he begin to address the objections to classical Darwinism.

First in the dock is adaptationism as a flawed strategy. The adaptationist fallacy is a direct frontal attack on old-fashioned Darwinian thinking. The attack was first launched by Gould and Lewontin in the famous Spandrels of San Marco paper (1979). What does Ruse have to say about this?
Now, what is to be said by the Darwinian in response to this charge? Simply this: whoever doubted the point that Gould and Lewontin are making? It has always been recognized by evolutionists—certainly from the "Origin of Species" on—that however common or ubiquitous adaptation may be, it is only part of the story. (p.135)
Bravo! In two sentences Michael Ruse admits there's more to evolution than natural selection and, therefore, the discontents have a good case. Now let's see if he understands what these other things are and why they are important. (Don't hold your breath.)

Several examples follow. In all of them, Ruse makes the case that adaptation isn't necessarily optimal. Sometimes there just hasn't been enough time for adaptation to succeed, this is why some bird species haven't yet adjusted to being parasitized by cuckoos. Sometimes natural selection has done a good, but not perfect, job; as in the circuitous route followed by mammalian sperm ducts that loop over the ureter. Sometimes natural selection is even maladaptive, as in the large antlers of the extinct Irish Elk. All of these examples are intended to show that Gould and Lewontin were wrong.

What about group selection? That's a major challenge to Darwinism and natural selection. Not a problem. Hamilton solved it by coming up with kin selection. Kin selection has been the greatest gift to adaptationist thinking since natural selection itself.

What about random genetic drift? Now, that's a real issue since there's very little doubt about its importance. (It's by far the main mechanism of evolution, properly defined.) Does Ruse agree? Nope. Ruse notes that random genetic drift was first proposed by Sewall Wright back in 1931 and expanded by Moto Kimura in 1968. But after some initial excitement Ruse concludes,
Wright's theory is not very Darwinian. Natural selection does not play an overwhelming role. Genetic drift is the key player in Wright's world. However, although many of these ideas were taken up by later thinkers, especially by Theodosuis Dobzhansky in the first edition of his influential "Genetics and the Origin of Species," drift soon fell right out of fashion, thanks to discoveries that showed that many features formerly considered just random are in fact under tight control of selection. Today no one would want to say that drift (at the physical level) is a major direct player, although in America particularly, there has always been a lingering fondness for it. (p.150)
There you have it. One of the most decisive and well studied alternatives to natural selection is dismissed as a fad. This is sloppy scholarship. Ruse clearly does not know what he's talking about. He's probably read too much of Richard Dawkins and his fellow philosopher Daniel Dennett, and not enough evolutionary biology textbooks.

Now we turn to punctuated equilibria. If Ruse is an opponent of Gould you would expect to see the standard references to saltation in this part of the book. You won't be disappointed. Although saltation and hopeful monsters have nothing to do with punctuated equilibria—and certainly nothing to do with the challenge to Darwinism—they are obligatory strawmen whenever you want to discredit Stephen Jay Gould. It's another indicator of poor scholarship.

Species selection, the real hierarchical challenge to Darwinism, isn't even mentioned. This omission is all the more remarkable since Ruse recognizes that in order to make a case for evolution at higher levels a non-Darwinian mechanism is needed; one that will decouple macroevolution and microevolution.
[Gould proposes] that at upper levels there are other mechanisms that the microevolutionists miss. Which of course might be so, but until some convincing alternatives are supplied, Darwinians continue to argue that in important respects macroevolution is microevolution writ large. Natural selection working on random mutation is the key to evolutionary change, long term as well as short term. (p.159)
What a remarkably crude way of dismissing all the work done by a large number of paleontologists, not to mention a 1433 page book called The Structure of Evolutionary Theory. Ruse may have good reason for rejecting species selection but we'll never know. Sloppy scholarship, Ruse should be ashamed.

Chapter 6 is the most important chapter since it covers the main objections of the discontented. Ruse fails to meet any of those objections; indeed, he fails to understand most of them. The rest of the book doesn't get any better.

I'll finish this off by quoting from the concluding paragraph of Chapter 6.
What is our end point? It is just plain silly to say that Darwinism is an exhausted paradigm or that selection is a trivial cause of change—or even that it calls for significant revision or augmentation. It is a powerful mechanism and has proven its worth time and time again. It is not all-powerful. Natural selection has its limits—limits that have been recognized since the time of Darwin (he himself noted many of them)—but taken as a whole, it is the key to understanding the organic world. There is no call for theory change yet, nor is there any prospect of such change in the near future. (p.165)
Speaking for the discontents, I beg to differ. Random genetic drift is by far the most common mechanism of evolution and modern evolutionary theory fully acknowledges this fact. Darwinism (natural selection) is important but it ain't the only game in town. Darwin knew nothing about random genetic drift. That's why it's wrong to describe modern evolutionary theory as Darwinism.

Gould and his colleagues have proposed a hierarchical theory of evolution in which natural selection is only one mechanism and it operates at only one level (individuals within a population). Hierarchical theory may not be correct but you'll never know from reading this damn book.

9 comments :

  1. Ruse never impressed me the way Will Provine does with his knowledge and understanding of evolutionary theory.

    ReplyDelete
  2. I've not read Michael Ruse's book (or Stephen J Gould's magnum opus, although I expect I'll get around to it one day). I'm not formally educated in any of the relevant fields beyond high school level biology and study on the intarweb thingy. Perhaps it is not suprising that I am confused.

    I've got at the back of my mind a top level evolution model of heredity/variation/selection as the explanation of the diversity of life. I think of this as Darwin's model without elaborating the different types of variation, and not separating out the different types of selection. As such the debate about neutral drift versus adaptation (whilst technically fascinating) is subsumed within the category of variation. Selection is not an active process and is blind to the source of variation (mostly); it acts as a gateway stopping some variations being passed on by heredity. I have read that neutral drift changes which improve fitness move very slowly to fixation, those that decrease fitness are lost fairly rapidly. Does this mean that a lot of the variation in genotype at any one time is down to neutral drift, but visible changes to phenotype are more likely to come from adaption?

    I am not convinced that species selection, if it exists, requires a different evolution model - I am not aware of a separate coding mechanism for species level characteristics. Having said that, any species of organisms that transmits information to next generation by cultural means ("Don't eat the toxic plants, Junior") could do so without involving DNA changes...

    Perhaps I'm confused because my mental model is too simple - or is everybody (or different schools of thought) using private definitions of key words?

    ReplyDelete
  3. discoveredjoys asks,

    Does this mean that a lot of the variation in genotype at any one time is down to neutral drift, but visible changes to phenotype are more likely to come from adaption?

    First, it's best not to use the term "neutral drift" to describe random genetic drift. Drift affects all alleles whether they are neutral, beneficial, or deleterious.

    Your question has two parts. The first asks whether variation within the population, at the genome level, is due to random genetic drift. The answer is yes, but it needs qualification. All variation within a population is due to ongoing evolution of one sort or another. All alleles are in the process of becoming fixed or extinct and this applies to alleles that are being selected as well as those that aren't.

    Since most alleles are neutral or nearly neutral, it follows that only a minority are being subjected to natural selection.

    The second part of your question is much more controversial. Most adaptationists think that any allele with a visible phenotype must be adaptive. What this means is that alleles for tongue-rolling, the length of your second toe, and eye color in humans, for example, must be subject to natural selection.

    Pluralists, on the other hand, reject this idea and claim there are many visible mutations that are effectively neutral. The issue hasn't been resolved but I think it's fair to say that the extreme adaptationist position (e.g., Richard Dawkins) is no longer tenable.

    ReplyDelete
  4. discoveredjoy says,

    I am not convinced that species selection, if it exists, requires a different evolution model - I am not aware of a separate coding mechanism for species level characteristics.

    True species selection cannot be explained by classic Darwinian natural selection since natural selection is a process that occurs within populations. There's an abundant literature on this topic, including a massive book by Stephen Jay Gould.

    The idea is that there are emergent species level characteristics that don't affect individuals within the species.

    Personally, I'm not convinced that true species selection (or even species sorting) actually exists.

    ReplyDelete
  5. I quit reading Ruse about 10 years ago. I doubt I have mmissed much...

    ReplyDelete
  6. Larry: I agree completely that N.S. almost by definition can't operate anywhere but among populations.

    But more generally, selection (as a process) can happen at any level. If you are a star one of your main "traits" is where you are. The stars near black holes are being selected against. If there was some variation in stars that had to do with where they are in relation to black holes, the distribution of that variable trait would over the long term be affected by the distribution of black holes.

    Species could be "selected" in a similar way. If high latitude species happen to be more subject to extinction than low latitude ones, we would see a pattern in numbers of different species.

    The problem with these models, of course, is heritibility. But still, selection as a process (generalized) is a useful concept for understanding many different kinds of patterns.

    ReplyDelete
  7. Larry Moran says:

    Most adaptationists think that any allele with a visible phenotype must be adaptive. What this means is that alleles for tongue-rolling, the length of your second toe, and eye color in humans, for example, must be subject to natural selection.

    Pluralists, on the other hand, reject this idea and claim there are many visible mutations that are effectively neutral. The issue hasn't been resolved but I think it's fair to say that the extreme adaptationist position (e.g., Richard Dawkins) is no longer tenable.


    Larry, is it really true that Dawkins argues that "any allele with a visible phenotype must be adaptive"? I always read him as making the much weaker claim that all adaptive phenotypic change is due to nautural selection? If anything, Dawkins weakness in his early writings was ignoring evolutionary pathways in general to concentrate on the only thing he was interested in at the time, namely explaining the existence of adaptation in nature.

    ReplyDelete
  8. The biochemical controversy over neutralism is concerned with the interesting and important question of whether all gene substitutions have phenotypic effects. The adaptationism controversy is quite different. It is concerned with whether, given that we are dealing with a phenotypic effect big enough to see and ask questions about, we should assume that it is the product of natural selection. The biochemist's 'neutral mutations' are more than neutral. As far as those of us who look at gross morphology, physiology and behaviour are concerned, they are not mutations at all. It was in this spirit that Maynard Smith (1976b) wrote: "I interpret 'rate of evolution' as a rate of adaptive change. In this sense, the substitution of a neutral allele would not constitute evolution ..." If a whole-organism biologist sees a genetically determined difference among phenotypes, he already knows he cannot be dealing with neutrality in the sense of the modern controversy among biochemical geneticists.

    Richard Dawkins, The Extended Phenotype p. 32

    ReplyDelete