From: Lee Corbin (lcorbin@tsoft.com)
Date: Tue Aug 06 2002 - 21:59:11 MDT
Chris writes
> Haldane's kin selection idea effectively extends the idea of self. Where
> self doesn't stop with the skin on your body but can be considered to
> include our shared genetic constitution. Very specifically I am unique but
> only relatively so - I share a great number of characteristics with my
> family, with other humans, and to a lesser extent finches, amoeba, trees and
> rocks. To the extent that a family member shares my genetic material they
> could be considered a part of myself.
The fact that people will sacrifice for "the cause", for the sake of
a waitress in a strange restaurant, and so on, would if we aren't
careful completely obliterate the notion of *self*.
Without doubt, evolutionary explanations exist. People don't (usually)
sacrifice themselves for rocks or trees, and I think that we are trying
to account for the existence of altruism, in the sense that we usually
take the term.
I like very much Charles Hixson's definition: "An altruist is someone
who does good things for someone else without expecting a reward."
In the paper Christ refers to below, this more abstract definition
is provided: "Altruism means that a system performs actions that increase
the fitness of another system using the same resources. Selfishness means
that the system will only perform actions that increase its own fitness.
> Of course the case of twins and clones makes it clear that even when genetic
> material is shared perfectly (or nearly so) it is not the case that the self
> is entirely the same - experience makes its mark on the individual too. I
> spent a good deal of time thinking about this side of kin selection - a kin
> selection not based on genetically shared traits but one based on
> memetically shared traits. I searched journals and the internet
> intermittently for years before I finally found an article that touches on
> the subject specifically.
There are plenty of good explanations for the "true/pure altruism"
that we are discussing that don't have anything to do with kin
selection (which was hot about 25 years ago).
> Here is the link (finally)-
> http://pespmc1.vub.ac.be/papers/PapersFH.html#RTFToC22
> I think the idea is a powerful one and it may give the conversation on
> altruism new clarity. 'Altruistic' actions can be viewed as the individual
> seeking to ensure positive conditions for the furtherance of their own
> memetic and genetic traits.
Here is what that paper had to say about group selection:
"The most obvious, but also the least accepted, explanation for the
development of altruism is selection at the level of the group. The
argument is very simple: compare two groups of individuals, e.g. two packs
of wolves. Suppose that one group is more cooperativ e, while the other
consists of more selfish individuals. Now because of the principle of
synergy, the cooperative group will be able to gather more resources, it
will be more fit, and hence will be selected, whereas the non-cooperative
group will be eliminated. Thus natural selection would promote cooperation.
"The error with this reasoning is more subtle. Though it is true that
individuals in an altruistic group will have better chances of survival,
this applies to all members of the groups, including those who are not or
less cooperative (because of blind variation there will always be slight
differences in "cooperativity" among the group members). The more selfish
ones will still have the advantages of the better cooperative organization,
but will have less disadvantages since they spend less resources or take
less risks in helping the other ones. The result is that they will be
fitter than the altruists, and their genes will eventually replace the
altruist genes in the gene pool of the group. In other words, cooperation
in groups on a genetic basis tends to be self-destructive."
Well, this dodges the more recent theories, IMO. Take Elliot Sober's
"Unto Others", for example. He explains group selection using Simpson's
Paradox. Group selection isn't dead.
Lee
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 09:15:55 MST