Darwin-L Message Log 5:42 (January 1994)

Academic Discussion on the History and Theory of the Historical Sciences

This is one message from the Archives of Darwin-L (1993–1997), a professional discussion group on the history and theory of the historical sciences.

Note: Additional publications on evolution and the historical sciences by the Darwin-L list owner are available on SSRN.

<5:42>From NEIMANF@YALEVM.CIS.YALE.EDU  Sun Jan  9 07:40:17 1994

Date: Sun, 09 Jan 94 08:39:23 EST
From: Fraser Neiman <NEIMANF@YaleVM.CIS.Yale.edu>
Organization: Yale University C∧IS
Subject: Re: Aquatic apes revisited
To: Multiple recipients of list <darwin-l@ukanaix.cc.ukans.edu>


The old issue of how we evaluate scientific knowledge claims has
surfaced again in the linguistic and aquatic ape threads on this
list.  I don't have anything novel to say on this topic.  But
perhaps it is worth pointing out that the discussion to date has a
rather empiricist flavor.  By that I mean that it fails to
acknowledge that theories about the way the world works and, more
to the point here, hypotheses about history that are constructed on
the basis of those theories, are fundamentally underdetermined by
facts (empirical evidence).  I mention this not to suggest that
there is no telling whether Renfrew or Morgan are right.  I think
there is.  Rather I want to remind folks that the ability of an
hypothesis to account for evidence in a parsimonious fashion is not
the only criterion on which its worth is judged.

The additional set of considerations, which often prove crucial,
involves the fit of the hypothesis or theory with other theories or
hypotheses about the way the world works.   The continental drift
case illustrates this nicely.  The original formulation by Wegener
did not win general acceptance NOT because it did not fit with the
facts.  It nicely fit all kinds of facts, including the
complementary shapes of the continental margins and agreement in
the distribution of fossils and geological formations on opposing
shores.   But it did not fit with other theoretical notions about
the way the geophysical world worked.  It offered no theoretical
mechanism.  The invention of plate tectonics remedied this defect
and acceptance quickly followed.

The "consilience of inductions" runs in two directions, one
empirical and the other theoretical.   I wonder if this goes some
way to explaining the difficulties that Morgan encounters among
card-carrying paleoanthropologists, and the ensuing lack of
resolution between the professionals and the outsiders.  Although
the argument is conducted in terms of "facts" much of the
professional position arises from an unarticulated lack of
agreement between the hypothesis and the rest of their
understanding about Plio-Pleistocene hominoid evolution and
evolution in general.

For example one thing that strikes me about the aquatic ape
hypothesis is that it is a "functional package" explanation.  It
attempts to explain a large suite of traits in terms of positive
feedback linkages among them in the context of a single selective
prime mover.  Now this kind of explanation has an long history in
paleoanthropology, most conspicuously in the hands-tools-reduced-
canines-brain-bipedalism package that goes right back to Darwin.
Scientific progress in paleoanthroplogy over the last 25 year has,
to a large extent, consisted in uncoupling the traits in this
package.  More generally, we have come to understand that
evolutionary histories caused by natural selection are quirky and
historically contingent processes.  Morgan's functional package
runs afoul of this more general understanding.

Clearly historical linguists have a similar kinds of problems with
Renfrew.  But the bi-directional consilience argument cuts both
ways in this case.  European history happened only one way.  The
question is how did it happen?   We are going to have a much better
chance of getting it right, if we follow the lead of Renfrew,
Cavalli-Sforza, Ammerman, et al. and begin seriously to look for
inductive consilience among historical hypotheses constructed from
theory in different disciplines.  The unfortunate tendency is for
practitioners in archaeology and linguistics to send immediately
for the game warden and have the poachers hauled off the estate.
I think a more profitable attitude might be to recognize that the
correct interpretation of the archaeological and linguistic records
is going to have to agree with the correct interpretation of the
genetic record, and vice versa.  This does not mean that Cavalli et
al. are right.  I think they _have_ pointed the way to an important
methodological opportunity.  Although they may not have exploited
it in an entirely satisfactory fashion.


Fraser Neiman

Your Amazon purchases help support this website. Thank you!

© RJO 1995–2016