tlhIngan-Hol Archive: Wed Aug 24 12:35:40 1994

Back to archive top level

To this year's listing

[Date Prev][Date Next][Thread Prev][Thread Next]

Re: Duj pIm

Hu'tegh! nuq ja' William H. Martin jay'?

Oy, the trouble I get into. This will be tricky...

=According to Nick NICHOLAS:
=> ... To clarify: there are two major
=> schools of linguistics at the moment; each school tends to think the other
=> is a waste of time, and they don't communicate between each other two well,
=The irony stops me in my tracks.

Dunno; seems rather natural to me. ghomDI' cha' Human, pa' wej vuD tu'choHlu'.
(Paraphrasing a saying about Greeks). Ultimately, it's all about politics.
There was a rather nasty flame-war on the normally academic and sedate
LINGUIST mailing list (for linguistics academics) back in May, where several
functionalists were claiming the Chomskians were keeping them out of
university jobs, and they had to go over to English or Literature or
Russian to get a job (*). All things considered, there's a lot of bad blood.
Frederick Newmeyer (generative (Chomskian) linguist, and author of the
excellent _Linguistics In America_ (or whatever his history of the science
is called) is organising a conference to bring the two camps together;
it'll be interesting to see what comes of it...

(*) My supervisor is sort of a data point in favour of this, having gone off
to Swaziland and thence Australia to work --- but I don't feel like dragging 
US academia politics into this.

=> The majority school
=> (at least in the US) is the formalists; it includes the Chomskian tradition.
=> These linguists seek to explain language as a formal system, and are
=> sympathetic to positing a language organ genetically determined in the brain;
=> languages are different, because a baby's brain selects 'switch' (parameter)
=> positions when it hears a language, and works out what goes where.
=Am I right to take this as the "mind as brain" approach. The
=mechanism must exist in order to provide the function. Language
=is then the execution of a preprogrammed facility in response
=to an environment.

Yes. I'm sure Holtej will correct me if I'm wrong, but I believe the formalist
answer to the question "why is language the way it is" is "because of
genetics and the evolution of the human mind." As you'll have seen Holtej
already say, they believe this is why babies pick up language so quickly
based on so little input...

=> I'm with the functionalists, who prefer to think of language not as a formal
=> system, but as a communicative resource. In particular, we seek to explain
=> particular features of language, not in terms of genetics and parameters,
=> but in terms of how a linguistic form best communicates a meaning. 
=Language, then, is more comparable to tools, once stone or
=bone, then flint, then metal and now plastic. If someone comes
=up with a better tool, competitive forces adopt the technique,
=though, as with tools, the manufacture of competitive products
=requires extensive refinement such that a new tool tends not to
=spring from an individual so much as within a company that
=regularly produces it. Some elements are fashion statements,
=and so last but briefly, though others are fundamental to the
=form and function of the tool and become long term. (Cars may
=stop having fins, but they almost always have four, not five or
=three, wheels.)

Almost... Um, consider. English used to look like Latin; it's increasingly
looking like Chinese (reduced morphology and increased reliance on word
order to convey what does what in a sentence.) I wouldn't say that Old
English was a stone knife, and Modern English is steel. Linguists take
it as a matter of faith that all languages are equally as powerful; if
you're looking for a stone knife in language, you'd probably have to
go back to the Neanderthal.

But your take on how *bits* of language change is... yeah, essentially
correct, in my view. The thing is, that functionalists will explain these
changes in terms of function --- which formalists are hesitant to. The
local functionalist guru here likes to say that formalism is like anatomy
(working out how the body works), and functionalism is like physiology
(working out why the body is the way it is.)

Hm. Here's an example. 100 years ago, American Sign Language was mostly SOV. 
Now it's mostly SVO. Now, it's SVO because of pressure from English. It
used to be SOV because French Sign Language was SOV. If you ask *why* the
change happened... a functionalist answer would go like this: there's a
reason you'd like sign language to be SOV:

Boy (signed to your left) Girl (signed to your right) Loves (move sign from
left to right).

If you do SOV, then you know what left and right means before you move
'LOVES' from left to right. This is taken to be easier to understand than
BOY LOVE GIRL --- more iconic, for starters: the sign for 'loves' moves
from the lover to the lovee, it doesn't just sit there. As a result, SOV
took a long time to die in ASL, and isn't quite dead yet. (Word order is
rather free in ASL; there's topicalisations galore).

What I've just said is a functionalist story; it ties form to function.
A formalist account won't do that; the formalist venture is to work out
the *rules* of language. A YACC parser doesn't ask why += is a grammatical
operand in C, but /= isn't (it isn't, is it?) And from what I know, formalists
view functionalist argumentation as rather woolly; Geoff Pullum, in his
damn funny _The Great Eskimo Vocabulary Hoax_, talks of the Berkeley
'Flat Earth' Functionalists (who believe there is no grammar as such, no
rules *distinct from* the pressures of communication.)

The more moderate line I subscribe to is that changes can happen quite
accidentally --- but once they do happen, people will find uses to put them
to. If bits get chipped off your blade, you make a feature out of the
bug: you make the blade edge jagged, and use it to tear at flesh. (May I
be forgiven for the analogy!)

=> (This, incidentally, is why Krankor's argumentation on the basis of 'rintaH'
=> in HolQeD 3:2 is so shaky. 
=I don't think ANYBODY should base ANYTHING on the use of
=rIntaH. 

Well... not true. Okrand could have chosen a particle like rItlh to do
the work of rIntaH, instead of a proper verb. And things like rIntaH *do*
pop up in languages --- including the languages Okrand works on. The point
is that Okrand as probably looking at rIntaH, *not* as a productive
Sentence As Subject construction, but as a particle leftover from Klingon
History. You need only compare the Melanesian Pidgin "em i dai pinis"
= "he VERB-MARKER die finish" = "Heghpu'/ Hegh rIntaH".

rintaH, like I said, is classic grammaticalisation --- and one could envisage
a scenario like this:

1. Hegh ghaH. rIntaH wanI'.
2. Hegh ghaH. rIntaH.
3. Hegh rIntaH ghaH.
4. Hegh rItlh ghaH.
5. HeghrItlh ghaH.

Indeed, we all know how our least favourite verb taH came about; but if
Klingon was a real language, in all probability the verb taH would have
come first, and the suffix -taH would have come about as the result of
a process like what I've speculated for rIntaH.

So yes, Okrand *is* making these things up, but he's also echoing things
that human languages do historically...

=The others [decisions] DO reflect decisions made here, based upon
=the lack of guidance from Okrand, which perhaps may soon be
=filled in something like an indirect dialog.

I admit: I'd rather he didn't fill anything in. I find our process of
working things out ourselves fascinating.

-- 
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>.<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
Nick Nicholas. Linguistics, University of Melbourne.   [email protected]  
        [email protected]      [email protected]
            AND MOVING SOON TO: [email protected]

Back to archive top level