Postal quote/directionality

[Martin Haspelmath]

Postal's view of language change as fashion-like and entirely
non-functional overlooks one important property of language change: its
overwhelming DIRECTIONALITY. In phonological change, we get

p > f
s > h
ke > ce
u > y
ii > ai
tata > tada > taa, etc.

but the reverse changes hardly ever occur. Similarly, in syntactic
change, we get

N > P
V > Aux
V > C
S+S > S
P > case-marker, etc.

but the reverse changes never occur. This overwhelming directionality
has not been sufficiently emphasized by historical linguists, I feel,
and its true extent can only be seen from a broad cross-linguistic
perspective (which Labov doesn't have, which may explain his skepticism
of functional explanations).

I am not aware of a similar directionality in fashion changes. For a
Chomskyan who thinks of the language system as largely determined by
arbitrary innate principles, with functional principles playing hardly a
role, they must be quite puzzling, and indeed I have not seen an attempt
to reconcile these observations with the traditional generative
ideology. (Optimality Theory is a different matter.)

So functional principles do play an important role in language change,
but they are not sufficient to explain it, because after all languages
do not improve globally by changing. Thus, I agree with Bill Croft that
both the functional perspective and the social perspective are equally
important, and neglecting either one can lead to serious errors.

Martin

--
Martin Haspelmath (haspelmath at eva.mpg.de)
Max-Planck-Institut fuer evolutionaere Anthropologie, Inselstr. 22
D-04103 Leipzig (Tel. (MPI) +49-341-9952 307, (priv.) +49-341-980 1616)