Re: Trivers' Reciprocal Altruism

From: alexandra bilak (
Date: Thu Nov 20 1997 - 14:34:44 GMT

Trivers' model explaining reciprocal altruism is an interesting one I
think, because it gave me the opportunity to criticise it on Tuesday
(that was probably because I didn't understand it properly, but

Trivers defines altruism as being a behaviour that benefits an organism
whilst being detrimental to the one performing it. Benefit and
detriment here are seen as related to contributions to inclusive
fitness. The first thing to understand here is that when Trivers talks
about reciprocal altruim, he is not referring to simple kin selection,
ie helping someone that is directly related to you, since this can be
explained in terms of gene survival (cf the Selfish Gene), but rather,
he is referring to acts of altruism towards an organism that has
"nothing to do with you", like between members of different species for

The model for altruism that is reciprocal involves variables such as a
cost-benefit ratio, which means that it involves actually weighing up
the "pros and cons" of your altruistic act. More specifically, he says
that an altruistic act will have to benefit the organism you are
helping more than it costs you. For example, the act of saving a
drowning man (ie RISKING your life for someone else) can be seen as a
kind of trade between two individuals. Here, what is being traded is a
50% chance of the individual in the water of dying, for a, say 10%
chance of you dying by rescuing him (because, for example, you are a
better swimmer). Here, then, the cost to you is smaller than the
benefit of the individual in the water.

Trivers points out that another important variable to consider is the
probability of your act being reciprocated in the future by that
individual that you are helping now. Thus, reciprocal altruism is an
exchange between a benefactor and a performer. Two individuals who risk
their lives for eachother will be selected, as opposed to two others
who will let die and thus have more chances of dying in a similar
situation because they will not receive the help of another. Genes for
reciprocal altruism will be passed on as succesful ones, since they
will favour the survival of two individuals acting for the benefit of

However, it is important to consider the possibility of someone taking
advantage of this behaviour, ie of cheating. Indeed, why should the
recipient ever bother to reciprocate the altruistic act in the future ?
Why not just take the help you got for granted and move on, without
ever bothering to help others ? It could be said though, that a cheater
will be selected against if the cheating has later adverse effects on
his life, which outweigh the benefit of not reciprocating. More
specifically, altruistic alleles will be favoured as long as the net
benefit to the altruist exceeds benefit to the non altruist. In that
case, would it be wise to restrict your altruistic acts to other
altruistic alleles ? The question here of course is how can you tell
who the cheat is going to be and thus who to help and who not to ? I
suppose that within a population of altruists cheats would be
successful for a time (like a dove in a population of hawks), but would
soon be discovered since eventually they would all die due to lack of

Trivers believes that there are indeed genuine acts of altruism in
animals, like for example little fish who clean bigger ones. The
altruistic act here is performed by the big fish, who lets the little
one approach him and clean him, doesn't eat him (how nice!) and even
warns him when he is about to move ! This is seen almost as a complete
ritual, where there are set times and places for being cleaned, and
even set cleaning "partners". This behaviour is seen as very altruistic
precisely because of these additional behaviours (eg, warning the
little fish when about to move) which can't really be seen as a direct
benefit to the big fish. The behaviour is selected by natural selection
and is beneficial to the host because the cleaner is worth more to him
alive than dead. Indeed, if it kills it, it loses valuable energy and
is exposed to unnecessary predation in the search for a new cleaner.

This is where I'm getting confused. Trivers is talking here of genuine
altruism. Is it really? On Tuesday, we talked about completely
disinterested acts of altruism, and whether they were possible. For me,
this isn't altruism, since there is always this cost-benefit thing. It
always goes back to the Selfish Gene explanation, because to me, this
particular behaviour seems more selfish than anything else, since it is
completely interested. Trivers says that the big fish is "abundantly
repaid for the cost of its altruism". REPAID ????!!!! Surely wanting
something in exchange for an altruistic act just goes to show that the
act isn't altruistic in the first place. The reciprocal bit makes me
think more of some kind of commercial trade (like Trivers actually
pointed out in the actual model)... but would you say that shop
assistants are being altruistic towards you when they SELL you
something ? I know this is a far-reaching example but this is what this
sounds like. And I thing this is exactly what Dawkins explains. Acts of
altruism don't really exist as such. They are the by-product of a
purely selfish mechanism, in which an individual is primarily thinking
about his own welfare above the welfare of anyone else. In this sense,
then, the notion of "reciprocal altruism" in itself is a complete

I'm really quite confused about this. Does altruism really exist ?Are
there really completely disinterested acts that are possible? I'd like
to go over this quickly one day. Sorry this isn't really a helpful
summary for anyone (and specially not for me) ...I'll have to read
everybody else's before I can get a somewhat coherent idea of all

This archive was generated by hypermail 2b30 : Tue Feb 13 2001 - 16:23:09 GMT