Left-right asymmetry and beyond

[Jon Aske]

Dear Lena and other fellow typologists,

I don't think I said that one cannot reach interesting and suggestive
generalizations without going beyond grammatical categories.  What I said is
that without going beyond grammatical categories chances are that we will
not go as far as we can go in refining our generalizations, and that we may
be misled into making spurious generalizations and asking the wrong
questions.  In some areas this is more obvious than in others, but I believe
it is always true to one extent or another.

Generalizations based on grammatical categories alone can yield valid,
pre-theoretical typological generalizations in some contexts (e.g. order of
S, V and O in decontextualized examples), but often little more.  I truly
believe that the notion of basic word order is of this type, for example.
Nobody would deny that this notion, as used by typologists for the last 30
or 40 years, has yielded very interesting generalizations, including those
about correlations of word order with other formal characteristics of
language.  There are plenty of very interesting generalizations that have
been arrived merely by looking at the order of grammatical categories, such
as what Matthew Dryer calls generically the 'head-dependent theory' of
harmonic constructions, and his own refinement of that theory, namely the
'branching-direction theory' (Dryer 1992).  Those are very interesting
generalizations indeed.  Or take the generalizations about negative morpheme
position extracted from many languages by Östen Dahl and by Matthew Dryer.
You don't need fancy theories to come up with those very interesting
generalizations.

However, without going beyond the grammatical categories with an
understanding of the functional categories involved, the meaning/function
pole of the categories and the constructions, our understanding of word
order can only go so far and, furthermore, the study of word order will
inevitably take many more wrong turns than it needs to in order to advance.
So, for instance, the notion of basic order based on grammatical categories
depends crucially on vague notions such as neutral order and unmarked order,
which I believe are makeshift solutions that cover up our incomplete
understanding of what's going on.

Also, as Lena observes, the simplicity of the basic word order model which
includes only the elements S, V, and O, becomes quite a different story as
soon as you add another element into the equation.  That, I believe, is
because if you take a clause with an overt subject and an overt object and
you turn it into a statement (a declarative assertion), the assertion that
first comes to mind, with minimal or no context, will have a certain
information structure, namely one in which the subject is the topic and the
object is the focus (notice that I am not calling it unmarked).  Thus if you
just deal with this kind of sentences, you can sort of ignore information
structure and you won't go very far astray.  But as soon as you introduce a
fourth element, the possibilities become more complex.  In particular, the
third element may compete with the object for the focus role, and it adds to
the possible possitions available for the focus constituent.

I would like to develop this point further and go over Lena's Russian
examples, but for that I would have to lay out my theory of information
structure and, although I can do it in about one page, maybe I should spare
you of that today.

As to Lena's second question (Is typology possible?), I do think it's
possible.  I don't see how my cautionary note about the perils of doing
typology should necessarily be interpreted as meaning that typology is not
possible.  But typology can be performed at different levels of generality,
different degrees of approximation, asking different questions, and so on.
My own personal preference, or bias, is to compare a few languages in
detail, but I would not know where to start if I didn't have all the
wonderful generalizations that typologists have been gathering about word
order from many languaes in the last 30 or 40 years, all the way from
Greenberg to Dryer.

On the other hand, I object, for instance, to some of the ways in which
people have attempted to pigeonhole languages into basic word order types,
just because it was believed that they had to belong to a word order type,
even if the fit was nowhere near as good as it is say between English and
SVO.  The literature is full of such examples and they all stem, in my
opinion, from a lack of understanding about the principles that guide word
order in those languages.  Well, actually I'm sure many linguists did have a
pretty good understanding of what was going on, but they still felt that
they had to choose some 'basic order' or another and so they chose some
criterion to guide them, such as statistical predominance, or some other
such less than fully satisfying criterion.

Finally, I don't think at all, Lena, that your comments are "too trivial to
be worth expressing in the discussion list".  I think they're very important
and that that is exactly what this list is for, especially for those of us
who do not have the luxury of having many opportunities of discussing these
issues with like-minded colleagues.

Till the next one,

Jon

-----------------------------------------------------
Jon Aske
Jon.Aske at salem.mass.edu - aske at earthlink.net
Department of Foreign Languages
Salem State College
Salem, Massachusetts 01970
http://home.earthlink.net/~aske/
-----------------------------------------------------
People demand freedom of speech as a compensation for the freedom of thought
which they seldom use.  --Kierkegaard.