Chimalpahin

[Galen Brokaw]

Swanton, M. wrote:

>In a descriptive grammar of some language L, all posited lexical categories (verbs, nouns, positionals, etc) should be defined based on their linguistic behavior in language L. In other words, lexical categories are language-specific and should be empirically, not "logically", defined.
>

Mike,
Thanks for your comments. On most of the points you mention, I'm not
sure we really disagree.
First, when I oppose empirical to logical, I am talking about the
difference between sensorial experience which is empirical and ideal
categories which are logical. Of course, you can't have one without the
other. But if we are just talking about language and the linguistic
system itself, lexical categories cannot be defined empirically because
there are so many empirical differentiations: they occur at different
times, in different places, with different pronunciations, etc. So they
all have to be synthesized into a single ideal, logical construct, even
though it depends upon empirical experience for its raw material.
Radical idealism emphasizes the logical dimension of this relationship
while radical empiricism emphasizes its empirical dimension. And there
are numerous other positions in between. The best place to stand is on
this continuum is debatable, but it seems to me that the terms of the
debate itself are pretty sound. I'm not denying that you have to define
lexical items based on their linguistic behavior in language L. The
point is that, although empirical experience is necessary, the
conceptualization of that experience is a logical ideal.

>When speaking of "verbs" in Nahuatl, we should not be applying a "metalanguage of European grammar to Nahuatl as if it there were an isomorphic relationship between the languages". Rather we should be referring to a category of lexical item that demonstrates certain linguistic behavior (in Nahuatl) distinguishing it from other lexical items (in Nahuatl).
>

I completely agree with you here. You are citing me out of context. The
sentence from which you lifted that quote says:  "...this illustrates
the kind of difficulty in applying the metalanguage of European grammar
to Nahuatl as if it were an isomorphic relationship between the
languages." I admit that my own grammar here is a little convoluted; it
was late when I was writing that. But I was making the same point you
are, which is why I said that this was a "difficulty" and "as if it were
isomorphic." What I was saying was that languages are not isomorphic.

>The fact that we use the term "verb" to describe lexical classes in English, Nahuatl, French, Mandarin etc does not mean that we take "verbs" in these languages to be in some sort of "isomorphic relationship". Nor does it mean that there's some sort of well developed cross-linguistic definition of this term outside of particular theories of grammar. It's just a convenient label. At best, the label "verb" captures something about the category being grammatically sensitive to time (whether manifested through tense, aspect, temporal adverbials, mood distinctions).
>

Again, I was saying that we cannot assume an isomorphic relationship.
Like Fran, however, and as I explained perviously, I do believe that
there are certain universals, and this is why all languages appear to
include certain linguistic categories, as Fran pointed out.  I would
argue that these categories are not universal themselves but rather are
derivative of whatever is universal about human experience. The fact
that not all linguistic categories appear in all languages would seem to
corroborate the idea that the categories themselves are not universal
but rather something else that underlies them and provides for the
possibility of variation.

>On another subject, I find your use of the word "grammaticalization" somewhat confusing. In linguistics grammaticalization refers to a diachronic process by which a lexical morpheme becomes a grammatical one. I don't think this is what you're referring to.
>

You are absolutely right. I am aware of the standard use of the term in
lingiustics, and I should be more careful about adopting it for use with
a different meaning, especially on this list where there are so many
linguists.  In this context, what I mean is merely the analysis of
language that produces a metalanguage we call grammar. Perhaps I should
use another term. As you can probably tell, I'm still developing these
ideas (with the help of Jose's Socratic prodding :-)).

>On a more personal level, no Nahuatl speaker I've ever met has translated chichiltic as "it became a red pepper". In our efforts to segment morphemes we should be careful not to confuse diachronic and synchronic analyses.
>

I was not suggesting that a native speaker would translate "chichiltic"
as "it became a red pepper." What I was saying is that this is the way
it is constructed according to the linguistic system. The point that I
was trying to make is that in the project of developing an organic
metalanguage, an analysis of this type of construction would have to
figure in. So, for example, one option might be to call this an
adjective as we do with words like "anaranjado" and "pointed." Another
option might be to call it a verb phrase and not even have an adjective
category in the grammar.

Galen