various topics

[Diego Quesada]

On Sun, 27 Feb 2000, John Myhill wrote:

> In discussions between formal linguists, it is routine for
> the participants to refer to 'facts' about 'exotic languages' which no one
> present knows and which no one present knows of a responsible work on, with
> no one questioning this as potentially problematic; it is assumed that if
> it is in print, it's true. In discussions between functional linguists,
> this happens only extremely rarely. If a formalist is careful about data
> from 'exotic' languages, it is because s/he personally believes that this
> is the responsible thing to do. If a functionalist is careful about such
> data, it is because of this but also because s/he is afraid of getting
> shown up or getting a bad reputation; you just can't get away with as much.


Uff!  That is a rather challengeable "slip of the fingers (on the
keyboard)". This is the sort of black-and-white judgements that have
hindered understanding in this discussion, as Givon pointed out yesterday.

At least two counterexamples can be cited that call for a softening of the
above statement: One is the much cited "impersonal, non-promotional
passive" of Spanish, which has given rise to the most fantastic theories
on the evolution of "passives", simply by paraphrasing traditional or
traditionally tainted grammars. The other is the celebrated "ergativity"
for some, "voice" for others, nature of Tagalog and sister languages,
analyses that always leak; with the most notable exception of one proposed
not precisely in the context of typological comparison, but from the
insights of a native speaker, who apparently did not feel pressed by any
straitjacket forcing him to abide to what was/is in the market:

Naylor, Paz-Buenaventura. 1995. Subject, topic and Tagalog syntax. In:
        Bennet, D. et al. 161-201. Object, Voice, and Ergativity.
        University of London: School of Oriental and African Studies.

This paper is so nice to read, especially because it puts a categorical
end to the phantasies existing on Tagalog syntax.

In view of the two counterexamples cited (and I'm sure people can come up
with more), John's above claim needs mending. I would suggest first of all
to speak in terms of tendencies rather than in terms of "discreet"
categories (the latter being an antifunctional precept).  Second, the
tendency that can be linked to the folks on the other sidewalk is that the
data are secondary to the model, while on our sidewalk the ***TENDENCY***
is to let the data speak first and then try to come up with some
explanation that need not be constrained by the structural configuration
of the phenomenon dealt with.  In other words, Chomskians (which is what
most people on this list have in mind when they say "formalists") are
deductive (a bit too much for some), while functionalists tend to be
inductive. The professional irresponsibilities incurred by either group
are not intrinsec to the approach, they simply reveal how lazy a linguist
can be when it comes to testing a retesting.

J. Diego Quesada
University of Toronto