Re: McGinn's disproof of Hardy-Weinberg

From: Jim McGinn (jimmcginn_at_yahoo.com)
Date: 01/06/05 

Date: Thu, 6 Jan 2005 06:11:38 +0000 (UTC)

Joe Felsenstein wrote:
> To review, in the discussion of Hamilton's kin
> selection results on altruism, I claimed that
> Jim McGinn's objections are really objections
> to all model-based results in theoretical
> population genetics. I therefore suggested we
> discuss something simpler than Hamilton's
> results, such as Hardy-Weinberg proportions.
>
> McGinn agreed that we could discuss this. But
> when I start to list assumptions of a model that
> results in Hardy-Weinberg proportions, he says:
>
> In article <crfs4b$5tk$1@darwin.ediacara.org>,
> Jim McGinn <jimmcginn@yahoo.com> wrote:
> >> >(I can't wait till we get to the part where you try
> >> >to include the assumption that altruistic genes will
> >> >be (can be) rare in a population. In another thread,
> >> >entitled Hamilton's Nonsense, Perplexed in Peoria
> >> >tried to assert this and ended up eating his words.)
> >>
> [me:]
> >> McGinn has not noticed that here we're
> >> deriving Hardy-Weinberg proportions, not
> >> anything about altruism.
> >
> >> (Also, as an aside, I don't know how he knows
> >> that altruistic alleles *can't* be rare, but
> >> that's another discussion so let that pass).
> >
> [McGinn:]
> >Another discussion? (Keep this comment in mind when
> >you get to the part of your explanation where you
> >include the assumption that altruistic alleles would
> >be rare.)
>
> Huh? I guess McGinn didn't "get it" that when
> we discuss Hardy-Weinberg proportions, we aren't
> discussing anything about Hamilton's results.
> We were discussing whether there was any model
> argument in theoretical population genetics that
> he accepts. And the argument about
> Hardy-Weinberg proportions holding in a
> random-mating population is about as elementary
> a result as you get.

Yes, and I agreed with it. So why are you now
whining?

> So I went on with moe assumptions, and McGinn says:
>
> [McGinn:]
> >I can't imagine why you'd want to include these
> >assumptions, especially this last one, but what
> >the hell. I have no problem accepting any of
> >these six assumptions, 6 through 11.
>
> and after even more assumptions:
>
> [McGinn:]
> >Come on, Joe. Enough with the smoke and mirrors here.
> >This is all standard probability theory. There is
> >nothing substantive in any of this that addresses the
> >issue at hand. If your question is whether or not I
> >accept standard probability theory then my answer is,
> >yes. Now get on with it.
> ...
> >Like I stated above, I can't figure out why you'd
> >want to include these assumptions, but I have no
> >problem accepting them.
> ...
> >I accept your assumptions.
>
> [me:]
> >> If he feels that it will all work out then
> >> I am not sure why he agreed to discuss the
> >> derivation of the Hardy-Weinberg proportions.
> >
> [McGinn:]
> >I agreed because you told me It has something to
> >do with showing the validity of Hamilton's rule.
>
> I did? No, I said that McGinn does not accept
> any model in theoretical population genetics,
> so that instead of discussing Hamilton's result,
> let's discuss Hardy-Weinberg proportions, which
> will be easier for people to think about. And if
> he objects to that derivation, then it makes
> clear that he objects to all model-based results
> in population genetics.
>
> That was extremely clear. I was not deriving
> Hamilton's result, I was deriving a very
> elementary result, Hardy and Weinberg's results
> of 1908.

Okay. And I agreed with them, thus showing
that you were wrong to assume I opposed to
all models in evolutionary biology.

> McGinn apparently missed the point. He
> thinks these assumptions were aimed at
> deriving Hamilton's results:
>
> [McGinn:]
> >I accept these assumptions, but Joe, you're
> >just about plum out of assumptions that I
> >would agree to. Pretty soon people are going
> >to start to wonder what all of this has to
> >do with Hamilton's rule. It's taken you seven
> >or eight posts to get up to this point and I
> >don't see how you are even one step closer.
> >Hopefully you are going to soon stop listing
> >assumptions and lay out your theory/hypothesis.
>
> What to conclude? McGinn *seems* to be saying
> that he'd accept the vailidity of a theoretical
> derivation of the Hardy-Weinberg proportions.
> If so, I would be wrong in saying that he would
> reject all models in theoretical population
> genetics.
>
> OK, so what to do? I could set out assumptions
> of a model which would derive Hamilton's rule
> (given those assumptions). But we've been that
> way before.

Yes we have. You failed that endeavor and I'm sure
you realize you'd fail again. Keep and eye on the
thread involving myself and Perplexed in Peoria and
you'll see where you would have failed if you had
made such an attempt.

> As soon as I listed assumptions McGinn demanded
> (post of October 29, 2002) that I show that
> these assumptions have "causal validity". He
> declared that
>
> > I myself, in my own models, am very careful not
> > to include assumptions other than those that
> > are verifiably part of nature/reality.
>
> (which is pretty funny since he has no models and
> no derivations).

Not true. (Your telepathy is misguided.)

> Where are we? McGinn is caught in a contradiction.
> If he accepts the models used to prove
> Hardy-Weinberg, he is accepting models whose
> assumptions (such as having no mutation) are not
> "verifiably part of nature/reality". Then why not
> in the case of models used to derive the
> conditions for kin selection?

Joe, now you're just being argumentative.
Nevertheless you are right on this point.
I retract that statement.

> If he rejects both uses of models, fine,
> but then we can see he rejects the whole
> of theoretical population genetics. Most
> people here would not be impressed by that
> position. But above he seemed to accept
> these simple models when used to derive
> Hardy-Weinberg proportions!
>
> I conclude that McGinn has a double standard
> for models. In this discussion he has
> demonstrated that clearly.

I cut and pasted the following from upthread:

************* begin cut and paste ***************

[Joe]
>> This is the same old objection from McGinn.
>> We go through the same process every time.
>> We come up with a simple model, in which it
>> can be shown that Hamilton's relatedness
>> measure is the one to use to derive the
>> condition under which a rare allele
>> predisposing to an altruistic behavior will
>> be expected to increase in frequency.

[McGinn]
> Your argument is, an has always been, that
> we should take your word that this is what
> it shows. I choose not to take your word
> for it.

[Joe]
Nonsense. I can define a model in which Hamilton's
rules work.

************* end cut and paste ************
So, Joe, would you like to retract this last
statement?

Jim

Relevant Pages

  • Re: c.u.s. FAQ, take two
    ... Stephane CHAZELAS writes: ... If there are no objections, ... > Thanks Joe and everyone. ... There are 10 kinds of engineers, those who understand binary and those ...
    (comp.unix.shell)
  • Re: c.u.s. FAQ, take two
    ... If there are no objections, ... Thanks Joe and everyone. ... last be able to continue scripting whithout someone telling you ... how how harmful it is for one month. ...
    (comp.unix.shell)