Underdetermination and Scientific Realism

[N Tasker <pia03nt@sheffield.ac.uk>]

To reply to this message or start a new topic please email: BUPS-DIS@bups.org




Dear Andrew,

A. You said: ?Having said this, I am still interested to know what you mean by
evidence, and confirmatory instance (these notion have been left a little vague
in this discussion). Could an application of Occams razor confirm a theory??

Aesthetic considerations like simplicity have already been discussed:
http://www.bups.org/bups-dis.w3archive/0603/msg00025.html
http://www.bups.org/bups-dis.w3archive/0603/msg00029.html

I think confirmation of a theory can be derived from (i) observations of a
theory?s empirical consequences, when those observations have been derived from
rigorous scientific testing which includes controls for spurious correlations;
(ii) the fact that a theory, t, is embedded in another theory, T, which is well
supported, even though T does not entail t; (iii) the fact that a theory serves
as a bridge between two well supported theories, even though it does not entail
these others.

No doubt there will be other ways in which a theory can be confirmed. 

B. The arguments, a, b, and c which you give in your first contribution are all
based on specific examples. I think you succeed in showing that there will be
certain metaphysical states of affairs which no one will be able to discover,
but this just highlights the different projects which you and I have in mind
when thinking about underdetermination. You see, arguments for
underdetermination in specific examples can never be sufficient to challenge
scientific realism. Scientists do not have to be able to claim omniscience in
order to assert that our scientific theories usually give us knowledge of a
world beyond our observations. In order to threaten this more modest claim,
underdetermination would have to be universal, or at least vastly more
widespread than enumerating a list of examples could ever show. So maybe we can
never know the topology of time, but this doesn?t mean we can?t know which
genes code for eye colour, or whether the planets move in ellipses.

You do subsequently propose a more general argument for underdetermination
derived from Putnam. I?m not convinced this threatens scientific realism. As a
challenge to scientific realism, underdetermination arguments apply only to the
epistemological claims of the doctrine. The reality of unobservable entities is
not in doubt, so the possibility of alternative theories which assign different
referents to the theory?s terms is no threat. Scientific realism will fall if
it can be shown that there are always alternative theoretical descriptions of
the unobserved entities which a theory postulates. But the argument you cite
cannot show this. 

Best wishes,
Nick


Browse or search the BUPS-DIS archives, or unsubscribe from the mailing list at: http://www.bups.org/mailinglist.shtml