Thursday, February 21, 2008

Flesch-Dawkins Readability Score

A review by Jennifer Schuessler of a new book connecting evolutionary science with literary studies (Comeuppance: Costly Signaling, Altruistic Punishment, and Other Biological Components of Fiction by William Flesch) includes this curiosity:

Our pleasure in narrative, in Flesch’s account, has a double edge. On the one hand, we enjoy seeing bad guys — “defectors” or “cheaters,” in the jargon of evolutionary psychology — punished. But we do this not out of sadism, but out of an ingrained, evolved desire to affirm our shared moral community and to shore up the group. (Flesch cites a lot of classic research from game theory to back up this point, and digs into the decades-long debate over the evolution of altruism; among other things, his book is a challenge to Richard Dawkins’s idea of the selfish gene.)
Having just finished Dawkins' The Selfish Gene, I am at a loss to understand what Schuessler could mean by this. If Flesch cites game theory and discusses the evolution of altruism, he is treading on exactly the same ground as Dawkins, who cites game theory extensively and makes the evolution of altruism one of the principle subjects of his book.

Perhaps Flesch does disagree with Dawkins, but the differences can't be as they're implied to be here. As stated, Schuessler's review seems to indulge in a caricature of Dawkins' argument, namely, that "selfish" genes translate to, determine, or justify "selfish" individuals, and/or that Darwinian evolution cannot account for altruism. It would be difficult to get The Selfish Gene more wrong than that.

Still, the review has done what a review should do: it has piqued my interest in the book, which appears to align with a pre-existing interest.


The Wife said...

Hi Dale,

Thank you for drawing my attention to this book -- which I find instinctively bothersome.

Why? From what I gather, Flesch seems to be on the group selectionist side of the Great Divide separating the American evolutionary scene (the fact that group selection does not seem to be an issue in Europe suggests that we're here dealing with a debate of some cultural specificity which ought to be seen as precisely that).

The group selectionists' spokesman is D. S. Wilson, who has recently teamed up with the godfather of sociobiology, E. O. Wilson, to slay that vicious dragon, the selfish gene, once and for all.

The article resulting from this unexpected double act can be found here:

Having had the pleasure to meet D. S. W. at a conference in the Netherlands last year, I have to say that there is only one dragon that needs slaying: group selection theory.

I simply don't understand why collective action needs to be given a genetic basis -- which is essentially what group selection theory is all about. To me it is perfectly sufficient that "selfish" genes (come on, we all know how Dawkins meant it!) can lead to altruistic actions.

That's the beautiful paradox of evolution! As a former post-structuralist, I love it!

Anyway, Dawkins has responded at length to the above article:,2121,The-Group-Delusion,Richard-Dawkins

I for one don't think that Wilson and Wilson have delivered a coup de grâce to Dawkins's ideas at all (as the latter's adversaries would claim).

I might be wrong about Flesch of course -- I haven't read the book, after all -- but as soon as I have it on my desk I shall confirm (or disconfirm) my hunch.

Great link, thanks so much!

Dale said...

The Wife, thanks for the links and the observations.

I have observed a reflexive anti-Dawkins-ism that seems especially strong here in the USA. Its most common form is a sloppy and tendentious presentation of his ideas. It's easier to argue with the Straw Dawkins, and many seem willing and eager.

william said...

Well, actually I'm somewhat agnostic about group selection, though it's convenient to my argument. I am a big believer in signal-selection, though, and my argument is anti-Trivers more than it's anti-Dawkins. But maybe that's a distinction without much of a difference. At any rate there are two parts: one is the empirical question whether people engage in altruistic punishment or not, and it seems to me (following Fehr and many others) that they do. Then there's the question of why they do. If it's a costly signal, a lot gets explained. And group selection makes it easier to see why it might be a costly signal. However you can get there (as Zahavi does) without group selection. And Zahavi is completely against D.S.W. Also, hello.

Dale said...

William, thanks for the comment and for stopping by. I look forward to reading more about this topic, especially some of the fuller responses to Dawkins and the ESS model that Flesch's (your?) book put forward.

martin said...
This comment has been removed by the author.
william said...

Whoops -- that was from me (another gmail account was left open and I thought that I was logging in, but not). Anyhow, yes, me WF