[ExI] Falsifying evolution/was Re: Really? and EP

[Dan]

--- On Wed, 4/22/09, Stefano Vaj <stefano.vaj at gmail.com> wrote:
> Why, it is true that evolution after a fashion is
> "non-falsifiable" in a Popperian sense, since
> every genetically-influenced trait must have evolved, one
> way or another, so that it may true that it is per se more a
> worldview than a scientific theory.
 
A firm distinction should be made between evolution the phenomenon -- or set of phenomena -- and proposed evolutionary mechanisms, such as environmental selection, neutral selection (which might, on a molecular level, make up most change we see), orthogenesis, etc.

In the Popperian sense, few of these mentioned mechanisms are non-falsifiable.  With environmental selection (often termed Darwinism), the method of falsifiability is simple population genetics in the lab.  Outside the lab, and especially with macroevolution, one can form hypothesis about particular evolutionary paths -- mapping these onto supposed selective pressures -- and test these using cladistics methods.*

Regarding evolution itself, I think the oft quoted phrase from Haldane serves well here "fossil rabbits in the Precambrian era [strata]" would, for him, disprove/falsify evolution.
 
> What are falsifiable, on the other hand, are theories
> concerning how a single trait has evolved and why. 

Well, the theories are often built from the micro-level observations or on the basis of micro-level assumptions.  E.g., Darwin's theory is basically based on some micro-level if informal observations of variable fitness along with the presumption that at least some traits that have an impact on this variable fitness are inherited.
 
> Now, some theories may be more far-fetched than other,
> other may seem at a time plausible to be then thoroughly
> confutated by subsequent hypothesis, but all in all this
> provides for a conceptual framework to understand and deal
> with the issues concerned which end up being more
> interesting that "God wills it". :-)

I can't read Damien's mind (yet!), but I think he was getting at how fast and loose some EP hypotheses seem to play -- much as earlier sociobiological ones did -- among EP's fans.**  Yes, there's a general presumption that seems to be correct: that evolution occurs and Darwinian mechanisms probably account for some observed behavior.  But from this, I believe, too much unsubstantiated theory is currently generated.  This might be especially so when dealing with very complex social changes.  (As I mentioned many years ago on this list, there's always been a tendency among amateurs and even experts to go for "just so" stories of evolution -- as opposed to minutely testing the narratives.  Testing here would involve not looking for supporting evidence, but looking for disconfirming evidence.)

Regards,

Dan

*  Not saying cladistics is trouble free or that cladistics methods are the only valid methods to attack these problems.

**  For instance, I agree, having read the fox data long before I heard of Clark and also being familiar with popular works like _Before the Dawn_, that it's possible for traits, behavioral or otherwise, to be fixed in a population in a very short time.  However, that this is so doesn't mean anyone's favorite laundry list of such traits has been so fixed.  (Side note:  I recommend Brooks' and McLennan's _Phylogeny, Ecology, and Behavior: A Research Program in Comparative Biology_ in considering the evolutionary study of behavior.  Of course, they're treating much simpler behaviors of non-human animals -- it's been a few years since I read the book, but I believe they used fish behavior to illustrate their views -- but it shows, IMHO, how to approach this with a mind toward testing hypotheses as opposed to merely falling for one's favorite "just so" story.)