rate of change

Larry Trask larryt at cogs.susx.ac.uk
Sat Jan 30 14:48:51 UTC 1999


On Sat, 30 Jan 1999, H. Mark Hubey wrote:

[LT]

> > And language change is a *lot* more complicated than baseball change.

> Probably also true. How about posting how you would go about measuring
> language change. Last time we had a similar problem on evollang list, it
> took months before I could get answers from you. Hopefully this time
> it will be a lot sooner.

I have never suggested that it is possible to measure "the rate of
language change" in a meaningful way.  As it stands, the concept is
simply far too diffuse to be usefully quantified.

Take a simple (but realistic) example.  Languages A, B and C all start
off very closely related.  Language A then undergoes very substantial
change in its phonology, but otherwise remains rather conservative.
Language B undergoes massive vocabulary replacement, but otherwise
remains rather conservative.  Language C undergoes major changes in its
morphology and syntax, but otherwise remains rather conservative.

Now we want to compare the rates at which the three languages have
changed.  How can we do that?  The rate of change is the degree of
change per unit time -- so we first have to measure the degree of
change.  But how?  Does acquiring six new phonemes count as a larger or
a smaller change than acquiring a new set of personal pronouns?  Does
acquiring ergative alignment count as a larger or a smaller change than
losing all final unstressed syllables?

How on earth can such questions be answered, except by assigning
arbitrary weightings?  "Let's see -- becoming ergative counts as four
doobies' worth of change, but losing final syllables counts as five
doobies."

It won't work.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk

[ Moderator's comment:
  We can now consider this discussion closed.  Thank you.
  --rma ]



More information about the Indo-european mailing list