Re: Disparaging Popper

["Scott Hartman" <scott_hartman@hotmail.com>]

> > Popper believed that it was possible to know truth and that
science could find it, as far as I know.<<

    First let me say, that I agree that most working scientists do adopt 
Popper's concept of falsifiability into their work.  But to my knowledge, 
Popper did not believe in it was possible to know "truth."  "Truth" is 
really an ambiguous term anyway, I prefer "reality." But either way, science 
approximates reality; It may end up revealing reality at the level of 
resolution of the question we researched, but we can never know for certain 
which aspects of science have arrived at reality (or the Truth) and which 
parts still need more refinement.  So while it may be true that light always 
moves at just under 3x10^8 m/s (and I certainly have no reason to suspect it 
isn't true), we will never know for certain whether there aren't exceptions. 
(actually, of course, but theyinvole cheats like running through diffuse 
mass, which kinda compromises that "in a vacuum" bit, but there may be more 
exception that are as yet unkown).  And this is true of all science.

> > But whoever posted this is correct, it doesn't do a very good job of 
> > explaining truth criteria in sciences which do not proceed by experiment 
> > (such as dino science<<

    It is a source of continual frustration to me that even some well 
respected philosophy of science profs seem to think there are "two kinds of 
science."  There is no fundamental difference of technique between so called 
"experimental" sciences and the "historical" ones.
    When a subatomic particle is accelerated to near-light speeds and we 
measure the resulting collision, no one actually "sees" it.  We indirectly 
read results from sensores that detect where particles have hit them, and we 
match that agains competing models of atomic physics.  Likewise, when we mix 
two chemicals together in a beaker, and watch to see what precipitates, no 
one "sees" the chemical bonds created and/or broken.  We infer them from 
models of chemical interaction.
    This is _exactly_ what paleontologists do.  The create a model, either 
rigourously (such as, a cladistic hypothesis) or more informally, (like the 
insertation point of the m. sulcatus), and these models make implicit 
predictions, even if they aren't specified in the paper.  We then compare 
them to new data, e.g. new specimens, or unanalyzed speciemns in 
collections, or data from other fields.
    It's true that with supercolliders, you can order up new tests easier 
than a paleontologist can find a key find that could falsify a key theory, 
but this is a difference in degree, not kind.
    Sometimes a frustarating lack of data allows for a sloppy aceptance of 
a proposal prior to sufficient testing, but like all science this is self 
correcting with new data.  And I would like to see more theoretical papers 
(and that includes phylogeny) include explicit predictions for the fossil 
record (and yes, my research on bird flight origins will have explicit 
predictions).
    A historical science is the exact same methodology, it just has less 
control over data collection.

Scott
_________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.

Share information about yourself, create your own public profile at 
http://profiles.msn.com.