Darwin-L Message Log 4:16 (December 1993)

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.

<4:16>From GA3704@SIUCVMB.SIU.EDU  Tue Dec  7 07:49:56 1993

Date: Tue, 7 Dec 93 07:49:39 CST
From: "Margaret E. Winters" <GA3704@SIUCVMB.SIU.EDU>
To: darwin-l@ukanaix.cc.ukans.edu
Subject: drift

I like Bob O'Hara's characterization of drift (linguistically
speaking) as being carried along by a current.  What is interesting
from the point of view of the history of linguistics (even recent
history) is that Sapir (1921) spoke of the drift of individual
languages (the loss of `whom' as part of the loss of inflectional
endings in general, for example), while others talk of drift
within a family (Robin Lakoff in her dissertation) or Sally
Thomason in her posting here about splits.  To carry it a step
further, Theodora Bynon, in her book in the red Cambridge series
(yes, that is how linguists often identify this series of
books on relatively basic topics), puts drift in her chapter on
non-genetic change and implies that the term can be used to
talk about change in geographically proximate languages which
don't come from a common source.

As a rule, historical linguistics doesn't like to think about
random change and, in fact, pushed by work in sociolinguistics
about the non-randomness of variation as  long as we can find
enough factors, would probably deny pure randomness.  The
closest we would come (help! Tom, Sally....) would be a class
of sound changes which are unconditioned; that is, there are
no circumstances that can be identified as motivating the
change.  One example might be Latin /u/ > French /y/ where
the /y/ is the sound in words like `rue', street or like
the German `u"' - with an umlaut.  This change happened
in every phonetic environment in French.  However even here
we can talk about structural conditioning since Latin /o:/
(long /o/) became /u/ and may have pushed the original Latin
/u/ forward to /y/.  As I said, historical linguistics doesn't
really look at a class of random changes - even meaning change
is being studied more and more in ways that remove the feeling
that meanings just shift, for no reason, all over the place.

             Long-windedly,
             Margaret Winters
             <ga3704@siucvmb.siu.edu>