Oh dear! I'd hoped I could delay any comment on this almost indefinitely, but I apparently must not. (A new posting today on the Edge.org site "raises the ante"–– so I cannot any further delay comment.) Today's entry can be found at
For those of you who are unfamiliar with the issues involved, it is probably most useful to go way back to Richard Dawkins' The Selfish Gene for an extensive discussion. The idea involved in using the prisoner's dilemma as the prototype anecdote for Darwinian evolutionary process pervades the entire book, but for its specific discussion it's probably simply best just to go to chapter twelve, which is devoted to a succinct summary of its relevance (and very nicely done, in fact).
The underlying assumption of the prisoner's dilemma is that it accurately portrays the advantages and disadvantages of cooperation and conflict in affecting evolutionary outcomes. There are several different versions of the anecdote, but all are based on appraising the advantages and disadvantages of betraying ("snitching" on) a former cooperative partner; that is, of determining how to choose between the likelihood of reward and the risk of punishment, without regard for any concern for the effects of that choice on others.
The "programmed" context of the prisoner dilemma situation thus precisely constrains the options available to the hypothesized participants, but there are also imported into the definition of the situation some unquestioned and highly problematic assumptions. The presumed participants are already identified as criminals who have been caught in a particular offence. Each is not only in consequence presumed to be primarily motivated by self-preservative concerns, but also by lack of concern for the effects on his partner of his decision to snitch. There is also presumed to be no concern about the consequences on society of the criminal activity that got them both arrested in the first place. In sum, then, the whole situation is defined in a way that takes as a given that each prisoner will be primarily motivated by exclusively selfish rather than cooperative concerns. This assumption, of course, is the one that pervades Dawkins' definition of the driving force in evolution itself. (Ironic, then, that the participants in the prisoner's dilemma situation don't evolve. Their criminality is simply not at issue, any more than is their place in any possible larger scheme of things . . .)
Dawkins himself clearly was (and is) ambivalent about a view of evolution that can characterize it as an "arms race." Indeed, the title of the chapter cited above is "Nice guys finish first." It's not yet known, however, how he reacts to the findings about a new solution to the iterated version of the prisoner's dilemma that are summarized in the article––referenced above––that prompted my decision to write this post.
That article is summarized by William Poundstone (in response to a request from Edge.org) as follows: "Robert Axelrod's 1980 tournaments of iterated prisoner's dilemma strategies have been condensed into the slogan, Don't be too clever, don't be unfair. Press and Dyson have shown that cleverness and unfairness triumph after all."
Whew! Perhaps this is a good moment to step back a bit, and take a breather––or at least a moment to come at the underlying issue from a different direction. Such a different direction is at least suggested by another Edge.org offering, which is at
Here's some background . . . The discovery by Lynn Margulis and Carl R.Woese of the importance of horizontal gene transfer and endosymbiotic processes in evolution resulted in a rush to reconcile the new information with traditional views, and thus to find a place for cooperation in evolutionary theory that would leave intact the "selfish gene" hypothesis. Much of that activity resulted in new hypotheses about the importance of altruism in evolution, and to articles, such as the above, that argue the pros and cons of the alternate hypotheses.
Perhaps it is too crude to suggest that the whole situation is one in which various of the participants are attempting to retain the hypothesis of a "mindless" algorithm for evolutionary processes, since the only viable alternative might seem to be a capitulation to the possibility of a "god-like" final cause.
Readers of MSOU will already have discerned another possibility (developed perhaps over-simply in the last chapter of the book). Since the universe has now evolved to the stage of itself manifesting the presence of mind, and since there is evidence of a directionality in its gradual (although in terms of astronomical time, sudden) emergence, it no longer makes sense to act as if that mind itself is fated to be inevitably stuck in Darwinian era competitiveness. Indeed, the implications of Woese's and Margulis's work suggest otherwise. And it is not inconceivable that a continuing refusal to accept the responsibilities that a self-observing mind is capable of recognizing would be amply sufficient to assure its extinction. It would amount to evidence of our own unfitness to survive.
write a comment