Re: Hamilton's Rule In The Mirror Evaded


<name_and_address_supplied@xxxxxxxxxxx> wrote in message news:da3sj6$10ct$1@xxxxxxxxxxxxxxxxxxxxxx
>
>
> Malcolm wrote:
> > Hamilton's rule was designed only for when b and c were positive, and r a
> > value between 0 and 1. Maybe it is possible to unify Hamilton's rule with
> > mutualism and spite, also with exploitation (cost and benefit both
> > negative), which you call OFS. As you point out, this last is a special case
> > because it may not be possible to distinguish refraining from exploitation
> > from altruism.
>
> It is true that Hamilton's main interest was in altruism, where both c
> and b are positive, yet Hamilton's rule is, and always has been, more
> general than that.

I completely agree.

> Also, it is wrong to say that relatedness is bounded
> by 0 and 1. In the canonical derivation of the rule, Hamilton (1970)
> made mathematically explicit his reasons for calling the coefficient of
> relatedness a regression coefficient in his 1963 paper, and highlighted
> that the coefficient could take negative values.

I have to quibble. What NAS says here is literally true. But the
paper most often cited is the 1964 paper, and in that paper he identified
"r" with Wright's coefficient or with Malecot's IBD. So in 1964, he
was bounding "r" between 0 and 1. But NAS is right that he very
explicitly removed those bounds in 1970 - returning to the logic he
had sketched in 1963. (Actually, the 1963 paper was written after
the 1964 paper).


.