Date: Sun, Aug 17, 2003
From: "Herb Gintis" 
Subject: Re: [evol-psych] Kin Selection vs. Group Selection?

Phil Roberts, Jr. wrote:

>Herbert Gintis wrote:
>> As I said in a previous posting, I refuse to use the term 
>> "altruism" and its opposite "self-interest" in the way
>> biologists do. This is because there is no altruism in the
>> biological sense, so why waste a good word?
>I don't understand this remark, Herb.
>I agree there SHOULD be no biological altruism.  But that's light
>years away from your claim that there simply IS no biological
>altruism, particularly in the light of Dawkins' discovery of an
>explanatory void (to be filled with memetics) in our naturalistic
>understanding of human nature, and which certainly appears
>to include lots of biological altruism (concern for the suffering
>of animals, 9/11 terrorists and rescue workers, etc.) as I have
>come to understand the concept:

At the level of the gene: a gene (or gene complex) that sacrifices 
fitness on behalf of other genes cannot evolve. In this sense, there
cannot be "altruism" at the level of the gene. So why waste a good 
word (altruism) for something that cannot exist. At the phenotypic 
level, altruism is behavior that reduces payoffs to the altruist, 
while increasing payoffs to others with whom the altruist associates. 
This is the "behavioral" or "phenotypic" use of the word. As you say, 
we know it exists, and the problem is to explain why. There are now 
many interesting models doing just that.

> Unlike [Lorentz and Montagu], I think 'nature red in
> tooth and claw' sums up our modern understanding of
> natural selection admirably. (Dawkins).
>> I and my colleagues use "self regarding" and its opposite,
>> "altruistic," in the behavioral, observable way, applied to
>> phenotypes.
>If I'm taking you too literally here, I apologize, but I'm one of
>those lonely souls who thinks the tendency to equate 'behavioral' 
> with 'observable' is more scientism than science if by that
> you mean to exclude introspection, self reflection, etc. from the
> mix, and has resulted in a grotesque underappreciation of the
> centrality of self-esteem in everything we humans think and do,
> at least to the extent that my own mind is not atypical.

The current usage of the term "behavioral" is not the same as the 
use in psychology in the early to mid-Twentieth century (Watson, Skinner, 
et al.). Rather, it refers to an experimental approach to modeling choice 
and behavior through experimental games. It does not reject introspection, 
but it requires that propositions derived from introspection be validated 
in the laboratory or by other scientific means.

Herbert Gintis
Emeritus Professor of Economics, University of Massachusetts
External Faculty, Santa Fe Institute, Santa Fe, NM
Recent papers are posted on my web site.

Get Game Theory Evolving (Princeton, 2000) at