Re: Trivers' Reciprocal Altruism

From: Stevan Harnad (
Date: Sat Nov 08 1997 - 14:12:53 GMT

On Thu, 30 Oct 1997, Liz Lee wrote:

> By altruism, Trivers is referring to any act committed which is of more
> benefit to the recipient than the cost to the benefactor.

Yes, but strictly speaking, any act that is of any cost to the
benefactor (irrespective of the size of the benefit to the beneficiary
[except if they are closely reated]) is a case of altruism. You have to
explain why I would even lift a finger to help you if it was not in my
own interest to do so.

> These are not
> necessarily high profile acts (saving lives) but may also be
> food/tool/knowledge sharing which are of some benefit to another. The
> reciprocal part is that at some stage in the future we expect our kind
> act to be repaid, if not by the individual himself, then by A.N. Other
> (possibly his relatives in the case of his death)....
> altruistic acts had previously been explained in terms of kin selection
> or group selection (Hamilton and Maynard-Smith),

And the gist of both explanations is that altruism is never done "for
the good of the group" or "for the good of the species." One way or
other, it's got to be based on benefits to the selfish genes.

[However, see:
for a possible return of group selection theory! This paper will be
discussed in PY308 -- Current Debates -- so it will be worth your while
to look at it while the selfish gene theory is fresh in your mind.]

> It seems plausible
> that we should have our own best interests at heart (Selfish Gene) and
> that any altruistic act is done for some ulterior motive, not
> consciously, but all the same because we will ultimately receive
> recompense in one form or another.
> The objections expressed to this suggestion are, I believe, based on
> our dislike of the underlying assumption that we are self-interested
> over everything else.

The objections are partly based on that, but partly also on the feeling
that there IS a way to make sense of group selection theory within the
framework of selfish-gene theory. To preview the idea: Dawkins himself
proposed in "The Extended Phenotype" (a book that appeared after the
two books you read) that the organism's phenotype does not stop at its
skin: it includes things it does in the world, such as the spider's
web, the beaver's dam, and even the chimpanzee's tool use. Are those
things part of the "vehicle," or the "survival machine"? If so, then
social interactions, and perhaps groups themselves, can also be part of
the vehicle. If so, then what's good for the group can become a variant
of what's good for the vehicle.

It's controversial. Dip into it only if you're interested and feel sure
you have mastered selfish-gene theory (otherwise you won't know who is
arguing for what). The URL above is for the article. The commentaries
and response are in the journal, Behavioral and Brain Sciences:

      JN: BBS, 1994, VOL.17, NO.4, PP.585-608

This archive was generated by hypermail 2b30 : Tue Feb 13 2001 - 16:23:08 GMT