Rate of language change

[Leo A. Connolly]

Having read several learned but indigestible treatises on whether the
rate of language change is constant, I would urge all to remember the
pseudo-science known as glottochronology.  It is based precisely on the
assumption that at least vocabulary is lost at a constant rate.  Trouble
is, it doesn't work.  For instance, Pennsylvania Dutch would have had to
have separated from German before the discovery of America, to cite one
notorious example.  Neither can we posit a constant rate of grammatical
or syntactic change.  Consider North Germanic languages.  Old Icelandic
and Old Norwegian were virtually indistinguishable; the principal
difference was whether unstressed syllables contained <e o> or <i u>.
The modern languages hardly resemble each other at all, since Norwegian
has changed drasticaly, while Modern Icelandic retains most of the
grammatical features of Old Icelandic and has developed phonetically in
so consistent a way that the spelling has needed only slight
modification.  Now surely part of this is because Iceland is isolated in
the North Atlantic, while Norwegian gradually became rather like Danish.

But if isolation is an explanation, we can only conclude that in
prehistoric times, language change is even less constant than what we observe.

Peace and joy nevertheless.
Leo Connolly