Paul Gallagher  wrote:

> In  wilkins{at}wehi.edu.au (John Wilkins)
> writes:
> 
> >Sober describes fitness thus in his 1984: 
> 
> >"Evolutionary theory understands fitness as a probabilistic
quantity. An
> >organism's fitness in the above model is its _chance_ of surviving. A
> >genotype's fitness is the average of the relevant probabilities
> >attaching to the organisms who have that genotype. The census
> >information of who lives or dies is evidence used for _estimating_
> >fitness; actual survival rates do not _define_ fitness." p43
> 
> >and goes on to note on p49
> 
> >"There seems to be no single physical property that varies as fitness
> >varies in all cases.
> 
> I don't have a copy of Sober's book with me. The emphasis on survival
> seems wrong: shouldn't the chance of reproducing and leaving offspring
> (and the chance that the offspring will leave offspring... and so on)
> also be taken into account? 

I think that he misspoke. The rest of the book is clearer, but I'd
really love to see him revise it-post Wilson.
> 
> I'm curious if Sober thinks selection alone can reduce fitness 
> or favor the less fit. As I mentioned fitness can be measured in 
> different ways: estimated number of offspring over the short-term 
> or long-term, the variance over time in estimated number of 
> offspring, exactly what defines the same environment (do very 
> rare events make a difference?) Depending on how fitness is 
> measured, selection alone can lead to the survival of the less fit, 
> instead of the survival of the fittest.

He has an example of that in the case where he discusses the Tragedy of
the Commons (did you note that Garrett Hardin died just a little while
ago?) and selection of the relatively fit driving a population to
extinction.
> 
> Now I'm not sure I understand Sober correctly. I think he's saying
> fitness is a real  (but non-physical) property of the organism, 
> even if it can't be precisely measured. The analogy to temperatue 
> is interesting, but isn't temperature precisely measurable, and
> isn't it independent of context? 

It's an analogy of ontology. Temeperature is not relative, to be sure,
but it *is* a *physical* property that applies to differing physical
substrates.

Sober thinks, and I concur, that fitness is a physical property.
Supervenience was proposed initially to deal with consciousness in the
philosophy of mind - if minds were all physically different, how did
they instantiate a physical property of - say - "seeing red" or
"knowing
that it will rain"? Jaegwon Kim proposed supervenience as a solution -
these things can be multiply realised physically, but if any two
physical systems were identical, they *had* to be in that state.

If you know any philosophy of mind, you know this has been extensively
debated, complete with possible worlds interpretations of modal logic,
from which nobody returns sane.
> 
> Clearly there are different properties of organisms that can be
> passed on to offspring and that can affect the properties of future
> populations. Assigning fitness values is a way of predicting the
> effects of these properties. But beyond that I'm confused...
> I think it makes sense to say temperature is the cause of a 
> physical process. But does it make sense to say fitness is a 
> cause of natural selection?
> 
Ultimately, in each particular case (this is critical), no. Neither does
it in each particular case of temperature. What causes, say, paper to
combust at 451°F is the binding of sufficient free oxygen to the carbon
and other reactive molecules of the paper such that they release energy
that causes still more molecules to so bind and release energy. To
entify "temperature" is to run into the same problems as when we
generalise the properties of a particular case of selection - say the
ability of one moth morph to evade capture due to confusing the visual
acuity of the major predator - to all cases. We note a similar dynamic,
and we assign a variable - fitness - into which we pour the specifics of
the physical case one by one.

Selectionist explanations are placeholders for particular explanations.
Theya re periphrases or promissory notes for real explanations. In my
parsimonious and typically sparse Australian ontology, properties that
are supervenient are patterns of dynamics.

-- 
John Wilkins wilkins.id.au
For long you live and high you fly, 
and smiles you'll give and tears you'll cry
and all you touch and all you see is all your life will ever be
---
þ RIMEGate(tm)/RGXPost V1.14 at BBSWORLD * Info{at}bbsworld.com

---
 * RIMEGate(tm)V10.2áÿ* RelayNet(tm) NNTP Gateway * MoonDog BBS
 * RgateImp.MoonDog.BBS at 10/7/03 11:52:20 AM