Cladistic language concepts

[John Hewson]

----------------------------Original message----------------------------
On Wed, 12 Aug 1998, Ghiselin, Michael wrote:
 
> ----------------------------Original message----------------------------
>           Dear Dr. DeLancey:
>                Thank you very much for casting your vote.  It would be
>           nice if more linguists would do the same because the sample
>           as it exists is small and perhaps not representative.
 
I suspect the sample is small because Larry Trask's answer was so
comprehensive, and so well expressed, that nothing more remained to be
said, except to emphasize particular points, as Roger Wright has done in
stating that "Spanish *is* Latin, only later", a view that I often
presented in class with the words "French is a 20th Century form of
Latin".
 
It is interesting that there is a related debate going on about Universals
and change. Professor Farlund has claimed that any change in a language is
a change "in the system". This is correct, but grossly oversimplified: a
language is a system of systems; one can have change in the phonology that
in no way affects the grammar. There is a system of parts of speech, and
subsystems such as the nominal system, verbal system, etc. Within the
verbal system one can have subsystems of tense, aspect, voice, and mood.
Within aspect systems one has to treat synthetic aspects as different from
analytic aspects, as the two interact rather than contrast.
 
The only universals in systems like this are operational: binary contrast,
secondary derivation, and so forth. Out of these systems syntax is
constructed, and again the only universals are operational: predication to
two levels, as in Jespersen's "very hot weather" (with "hot" predicated of
"weather" and "very" predicated of "hot") and recycling whereby
these three words can be treated as a single element, a NP which can
then become the support of a verb, which can be expanded into a VP, etc.
We build sentences the same way a child builds models with Lego blocks,
and we don't need rules to do it, any more than the child does. All such
construction depends on cognition: realizing what fits together and what
does not: adverbs cannot be predicated of nominals for example: *very
weather. Rules are created by linguists to describe regularities; they
are part of the description, not of what is described.
 
All this, alas, to make the point that linguistic change *is* a language
universal, and that change is also systemic (although it can be, and too
often is treated atomistically). I hope that it is also clear that what is
meant by "universal" and what is meant by "system" can vary greatly from one
linguist to another, and consequently there are enormous confusions
concerning the very fundamentals of our discipline.
 
The nature of systemic change was realized quite early in such statements
as Grimm's Law, the lowering of all short Latin vowels by Late Latin, and
the loss of length distinctions by early Romance, and so on. The
regularity of sound change is also the direct result of change in the
system: change the _s_ on a typewriter to _$_ (i.e. change the system),
and the re$ult i$ a$ follow$...
 
The systems of a language are mental realities that are not amenable to
direct observation but, like gravity, are amenable to indirect
observation. It is the phoneme, a systemic entity, that triggers the
speech apparatus to produce a sound that will be modified in all kinds of
ways in the processes of speech, producing an endless range of
allophones, the directly observable data. If phonemes were directly
observable, we would not need to train linguists. Such systems are to be
treated as scientific substructures, not as abstractions, which may fail
to correspond to any existential reality whatever. Abstractions, where
anyone's "assumptions" are just as good as anyone else's, are very often,
in methodological terms, a waste of time.
 
 
 
 
John Hewson, FRSC                               tel: (709)737-8131
University Research Professor                   fax: (709)737-4000
Memorial University of Newfoundland
St. John's NF, CANADA A1B 3X9