John Wilkins wrote:
> Phil Roberts, Jr.  wrote:
> 
> 
>>John Edser wrote:
>>
>>>PR:-

>>
>>The folks I have cited (Bhaskar, Kuhn, Toulmin, Feyerabend, Manicas
>>and Secord, Harre, Brown, Hanson, Polanyi, etc., have absolutely
>>nothing to do with post modern reconstructivism, a decidedly
>>philosophical bent.  The fact that the criterion of certainty
>>has been abandoned in favor of the more realistic criterion
>>of reasonableness has nothing in common with the thesis that
>>pretty much anything goes (post modernism, as I understand it).
> 
> 
> In analytic philosophy this is most often associated with the so-called
> Gettier counterexamples, where something that was justified (made
> reasonably certain) true and believed was not knowledge. But the attack
> on the idea that knowledge required certainty was best made by
> Wittgenstein in _On Certainty_.

The little brewhahhah over whether or not knowledge is "justified true
belief" (the tripartite criterion, or some such rot) has never played
much of a role in what philosophers of science have had to say about
the nature of science.  Indeed, I would say its the sort of tempest
in a teapot that gives philosophy a bad name.  The best author on
this matter for me has been A. R. White, and his almost unheard of
nifty little book, 'The Nature of Knowledge'.  He debunks the notion
of justified true belief, and argues the issue of justification lies
within the domain of epistemics, not epistemology, a view that Alvin
Goldman has also come to embrace.  For White, knowledge is simply
'right representation'.

Haven't read Witt's 'On Certainty', but I think its widely understood
that the real blow to certainty was the demise of Newtonain mechanics.
Science is rarely influenced by developments in philosophy, I am told.

In philosophy, I suspect Hansen's assault on "the myth of the given"
has probably have more impact on philsopher's of science, if we
are talking about philosophical influences.

> 
> Postmodernism is a kind of "anything goes", but it is easy
to overstate
> this. Caricatures don't help. Some postmodern views have a fairly good
> grounding in things like social and conceptual relativisms.
> 

Yes.  Much better.  Its about relativism, and as such, the foregoing
of the notion of truth, etc.  Everythings pretty much a matter of
opinion.  Perhaps thats a little better way of saying it.

Rather than merely opinion, underlying most modern science is the
thesis of versimiltude, (e.g., Newton-Smith and others).  Scientists
don't regard their work as merely opinion, but rather as
getting closer to the truth, but perhaps with the acceptance that
it is a goal that can never be entirely attained.  Its scientific
realism devoid of certainty, I suppose.

> 
> Popper was most certainly *not* a positivist, and few who know the
> history of 20thC philosophy would take him to be one. His abandonment in
> philosophy of science and epistemology comes from the recognition that
> falsificationism, even when made sophisticated, simply fails to account
> for much of epistemic dynamics.
> 

Authors I have read see an affinity in the fact that falsificationism
was a critque and an attempt to repair verificationism, and cut
from the same cloth in sharing the view that their is a simple
algorithm for the doing of science.  Its part and parcel of the
view that reasoning can be
reduced to logic, rules, procedures.  But Ross Pero nailed it for
me with his 'There ya go.  Now you've got it.  Ya see.  You're thinkin'
outside the box'.

> But to leap from that to aesthetic claims for knowledge is to equally
> misrepresent things. Elegance is a test based on parsimony, which is a
> guiding principle in empirical science. 

It depends on the level of abstraction you are working at.  That,
and other factors, such as the existence of competing hypothesese,
stiff resistance, etc.

The two examples I've cited were of two of the major discoveries of
our century, relativity and DNA.  Both seemed to be heavily driven
by aesthetic considerations.  I know that sounds crazy, but I'm not
making this stuff up:

    Mike Douglas:

       'But how did you know that Linus Pauling's model of DNA
        couldn't possibly be correct with only a brief chance
        to examine it?'

    James Watson (with Crick's nodding approval in the background):

       'Because.  It wasn't beautiful.'


Einstein's explanation of what got him thinking about relativity
can be found in his book 'Ideas and Opinions'.  As I mentioned,
it was to a large extent a result of his disatisfaction with
the assymetry in Maxwell's equations.  That, and a number of
thought experiements that couldn't be actually conducted, not
even in principle.   I do think a minor test was proposed in
the 1905 paper, but it never was of much interest to anyone.
Eddington's experiment was dreamed up by someone else, as I
recall.

> If something is adequate to the
> task and elegant, then it is to be preferred to something that is
> elegant to the task and baroque. 

Yes.  There are many considerations, and they occur in a dazzling
array of hierarchies and machinations.  But that's more or less
my point.  There is no simple royal road to science.  This is
all part and parcel of the romantic notion that you can reduce
reasoning to logic.  We have lots of reasons to believe this
just ain't going to happen, ranging from Godel's proof to all
sorts of paradoxes and contradictions starting to show up in
decision theory (e.g., Newcomb's Problem, prisoner's dillema,
etc.).

e.g., check out Coleman's little spiel:
http://pages.britishlibrary.net/blwww3/education/cooperation.htm

> 
> Sometimes. There is a classical example of siologists of knowledge going
> into the field and ignoring the science in favour of the "power
> relations" or "institutional structures and functions" etc. 

Prof. Kuhn, no doubt.

> 
>>Again, don't confuse me as arguing that testability is of no value.  Its
>>a HIGHLY VALUED epistemic tool, the gold standard, as a matter of fact.
>>Its just that its not the only one available, and rarely of concern when
>>first formulating hypothesese.
> 
> 
> Arguable. One thing that is uppermost in the minds of those I know is
> whether or not they can adduce experimental evidence for a novel idea.

Depends on the context, and the level of abstraction, would be my point.
And, as hard as it is to believe, beauty is often a hallmark of truth
to some of these folks.  For example, subatomic physics is deeply
concerned with symmetry.  If you find a particle of such and such,
you can pretty much bet the farm you are going to find a contrasting
counterpart somewhere in the mix.

> Typically, they devise an experiment. Many do not work out, and the
> ideas are dropped, even when they put a lot of work into it.
> 

What experiment did Dalton propose when he postulated atomic theory
to account for the law of multiple proportions?

What "test" did Rutherford propose to confirm his hypothesis that
the alpha particle angles were caused by atomic nucleii?

> 
> 
> Darwin's tests were manifold. 

On what page of 'Origin', or perhaps just the chapter, will I be able
to find Mr. Darwin proposing these tests?

> Some of them involved finding adaptations
> that served other species, or the lack of a feasible pathway for the
> development of *any* organ or arrangement over evolutionary time. Since
> on of the major explananda was the arrangement of groups within groups
> in taxonomy (explained by common descent), the existence of cases that
> did not meet this would be a test of the universality of common descent
> (and with lateral transfer and studies on hybridisation, Darwin's thesis
> *has* been substantially modified since).
> 

Remember, we are not talking about subsequent "tests", but those proposed
by Darwin himself, given your and John's contention that one can not
possibly propose a theory without also proposing a means of testing it,
that such behavior would be "unscientific".

> 
> Reproductive fitness is not now and hasn't been since Wright began
> publishing in the 1930s, the sole test of evolution. This is a classical
> overselectionist mistake.

I could use a bit of elaboration on this, old bean.   :)  Are we talking
Sewall Wright?

PR
---
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 5/30/03 3:09:15 PM