Text 1036, 171 rader
Skriven 2004-12-16 12:30:00 av Tim Tyler (1:278/230)
Ärende: Re: SPITE:  Hamilton's La
=================================


John Edser <edser@tpg.com.au> wrote or quoted:

>  Tim Tyler <tim@tt1lock.org> wrote:-
> 
> > TT:-
> > If it helps, here's the same content as a PDF file:
> > http://blackwell-synergy.com/links/doi/10.1111/j.1420-9101.2004.00
> 775.x/pdf
> 
> JE:_
> Here is what I got:-
> ?Database Error while loading DOI '10.1111/j.1420-9101.2004.00'?

That would be what you would get if you visit the URL:

http://blackwell-synergy.com/links/doi/10.1111/j.1420-9101.2004.00

Long URLs in SBE get spilt somewhere during the moderation process.

Here are the URLs as TinyURLs:

Abstract: http://tinyurl.com/5befh
HTML: http://tinyurl.com/544op
PDF: http://tinyurl.com/457jr

> > JE:-
> > Spite:-
> > ?an action that harms a recipient at no direct
> > benefit to the actor, could evolve if interactants
> > were negatively related?.
> 
> That's the idea exactly, yes.
> If you can't improve the lot of your offspring directly, you
> might still be able to benefit them and their genes indirectly -
> by sabotaging their prospective competitors.
> 
> JE:-
> _____________________________________________
> Spite constitutes an absolute fitness loss
> for _both_ the actor and recipients.
> _____________________________________________
> 
> Does anybody appreciate what this means?
> Within Hamilton?s Rule (which remains 100%
> relative without the total fitness of the
> actor) one side can _appear_ to benefit
> relative to the other only because one side
> is losing out at a lessor rate than the other.

The "spiteful" process is not very pleasant for either party.
However the spiteful party might still be doing very well -
and be part of a thriving population despite their vindictiveness.

> Hamilton?s mutual absolute loss nonsense
> has to compete against the people next door
> (other members of the same population)  who
> carry on mutually exchanging total fitness
> gains.  Guess who is selected FOR (no points
> for guessing correctly).

That would be relevant if there are adjacent neighbours
without comparable domestic problems ready to move
in on the niche.

That's not always the case, and where the species is not
under substantial pressure from close competitors,
traits like spite could evolve.

There are a couple of analogies here:

You might think that "impractical" sexual ornamentation would
also fail in competition with more practical alternatives -
yet it can sometimes get established - despite its drawbacks.

Also, a similar situation arises with segregation distorters.

These get suppressed by other genes - but exist in considerable
numbers nontheless - and sometimes manage to get expressed.

> Hamilton?s theory for the evolution
> of spite illustrates the enormous error
> that remains within his rule: it does
> not contain a constant term to provide
> a missing reference point.

It doesn't model competition with other species -
....but it doesn't pretend to do that in the first
place.

If you want to use multiple-level-of-selection
models along similar lines, Price showed how
to do that - i.e. this:

``W.D. Hamilton (1996) failed to recall when Price first contacted him, 
  but says he Price had read Hamilton's 1964 papers on kin selection, and 
  with no training in population genetics or statistics devised the Price 
  equation, a covariance equation that generated the change in allele 
  frequency of a population. Although the first part of the equation had 
  been previously been derived by C.C. Li , its second component allowed 
  it to be applied to all levels of selection, meiotic drive, traditional 
  natural selection with an extension into inclusive fitness, and group 
  selection.''

 - http://en.wikipedia.org/wiki/George_R._Price

> This allows any mutual loss that is not equal to show up as a ?gain?
> for one side. Would you invest in company A
> only because it only lost 100% of its capital compared
> to company B that lost 105% providing a relative
> gain for company A of 5%? Both companies are
> bankrupt. You DO have the opportunity to invest
> in many other companies that make a profit!
> Choosing either bankrupt company remains
> hopelessly irrational.

A more sensible analogy would involve a company that had not lost all
its capital - but had merely made an investment in sabotage against a 
competitor.

An example would be a company's employees posing as dissatisfied
customers of their competitors - or the "anti-marketing" technique
known as "Fear, Uncertainty and Doubt".

Unfortunately, companies often exist in markets where the "effective
population size" is rather small - and so covert sabotage operations
can be effective - *if* they can be effectively performed anonymously.

> TT:-
> Calling it "relatedness" still seems fairly reasonable, though -
> since it's usually a close match for Wright's idea of relatedness.
> --
> JE;-
> Calling a regression analysis of
> biological relatedness
> ?relatedness? and then allowing a
> theory of relatedness to be replaced by
> this regression model which was derived from
> the theory via the Neo Darwinian process
> of simplification/over simplification,
> such that the model is allowed to contest
> the theory it was simplified from, remains
> rationally absurd.

Hamilton's inequalty is trying to model the conditions
under which genes are likely to spread in a population.

The quantity "r" /is/ what comes out of such models.

If you object to it being called "relatedness" then
I suppose that's OK - though you ought to be able to
understand why others don't object - since:
r(mother) ~= 0.5, r(brother) ~= 0.5, r(husband) ~= 0,
etc.

....and if you are arguing that competition between
species means that the model's "population" needs to
be expanded to include members of other species then
you need to look at Price's equation - which deals
with this case.
-- 
__________
 |im |yler  http://timtyler.org/  tim@tt1lock.org  Remove lock to reply.
---
ū RIMEGate(tm)/RGXPost V1.14 at BBSWORLD * Info@bbsworld.com

---
 * RIMEGate(tm)V10.2á˙* RelayNet(tm) NNTP Gateway * MoonDog BBS
 * RgateImp.MoonDog.BBS at 12/16/04 12:30:34 PM
 * Origin: MoonDog BBS, Brooklyn,NY, 718 692-2498, 1:278/230 (1:278/230)