The appeal of P&P

sergi balari Sergi.Balari at uab.es
Tue May 1 16:33:06 UTC 2001


Dear all,

This is a long message. My apologies in advance.

I would like to expand on Andreas ideas about the appeal of the P&P
approach.  I agree with him that a great deal of he appeal of P&P lies on
the way most of its proposals are framed within a putative cognitively
plausible set of ideas. This certainly gives the impression that one is
engaged ina kind of quest for the Holy Graal. In fact, if you let me
stretch a bit more the metaphor, note that this quest has something to do
with the Arthuric legend in the sense that it is posited in extremely
pessimistic terms. This is particularly clear in Chomsky's most recent
philosophical writings. For example, in a random search, I find the
following lines:

"Specifically, it is unknown whether aspects of the theory of mind --say,
questions about consciusness-- are problems or mysteries for humans,
though in principle we could discover the answer, even discover that they
are mysteries..." [N. Chomsky. 2000. 'Naturalism and dualism in the study
of language'. In New Horizons in the Study of Language and Mind. CUP, pp.
75-105.]

Thus, this mixture of feelings, this constant appeal to 'mysteries', all
give the impression that one is engaged in a kind of extremely important
adventure that, at the same time is extremely difficult, if not
impossible, and cover the whole thing with the necessary patina of
heroism that provides  the enterprise with its appeal. There lies its
weakness and, at the same time, its strength, as it is, in many ways,
almost undefeasible.

To the extent that Chomskyan thinking is the most visible part of
linguistics, I believe this is driving our field away from cognitive
science and, in my opinion, this is bad. Let me go back to this in a
moment, and get first into some more sustantial remarks.

Andreas points out that some evience is needed from fields outside
linguistics that is capable of challenging the Innateness Hypothesis
(IH). I think he is right, bu we can do more than that.

Carl Pollard, in an earlier posing to this list, expressed some
skepticism about the possibility of showing that a 'program' (as opposed
to a 'theory') is wrong. (I may be not very precise her, but with the
double postings I started erasing messages and I seem to have lost
Carl's. So, Carl, sorry if I'm not being faithful to your original
words.) Well, one may not be able to demonstrate that a research program
is wrong, but it is possible to point at incoherences that might render
it inviable.

In his message, Andreas focuses on the IH. I think this needs to be
qualified in the following way:

1.- The IH *as Chomsky seems to have expressed it* is an extremely naive
one, directly adapted from the original Cartesian thought (with some
Romantic seasoning provided by Humboldt and Herder). It is naive in the
sense that it presupposes an extremely strong version of genetic
determinism (to express things in contemporary terms): there is no actual
flux of knowledge from the environment to the organism (or the organism's
mind, if you prefer); stimuli just serve as triggers for this
pre-existing knoledge to unpack into a finished product at the end of
some develompental process. Here, the argument from the poverty of
stimulus (which is probably flawed and empirically untenable, as Andreas
notes in his reference to the work by Culicover, for example) plays an
important role in the conceptual scaffolding of these hypotheses.

But there is more to innateness than just that, and here is where, as
noted by lara Taylor in her message, the work by Elman and many others
becomes particularly relevant. Chomskyan nativism is naive because the
sort of genetic determinism it presupposes is hard to reconcile with most
current views within developmental psychology and biology.  The point is
that strong nativism invokes a view that gives almost all the poser to
the gene, whereas the developmentalists's work (both in biology and in
psychology) is providing very interesting evidence that ther's no much
sense in this conception of innateness. It doesn't make much sense to say
that genes  code for such and such capacity, as it appears that
development of such capacieties is best seen as a complex, dynamical,
dialectical, process, where the end product is a function of the complex
interactions that occurred among environmental and genetic factors. I
believe that this is beautifully (and accurately) exposed in Susan
Oyama's book The Ontogeny of Information (2nd Edition 2000, Duke UP), but
for other, less philosophical discussions, I would also recomend reading
the work by Elman et al cited by Lara Taylor (Perhaps I would add: J.
Elman. 1995. 'Language as a dynamical system'. In R.F. Port & T. van
Gelder, eds. Mind as Motion. MIT Press, and, for other areas of cognitive
science, the work by E. Thelen and L. B. Smith 1994. A Dynamic Systems
Approach to the Development of Cognition and Action. MIT Press.)

But I contend that these weaknesses extend well beyond the area of
language acquisition. For example, to all that has to do with conceptions
of the Evolution of Language (EoL). This is my second point.

2.- It is well-known that Chomsky has expressed his skepticism towards
any speculation concerning EoL, in particular when these speculations
show any trace of what has been termed 'adaptationist thinking'. Somewhat
paradoxically, one of the main sources for Chomsky's anti-adaptationism
is Richard Lewontin. I say paradoxically because Lewontin's arguments
against adaptationist views are tightly connected to his attacks against
strong genetical determinism; in fact, he is a great supporter of, for
example, Oyama's ideas. I beleive this is already incoherent per se, but
things get worse when Chomsky himself tries to speculate on EoL. Briefly:
if you look at his 'evolutionary fable' in Minimalist Inquiries, you will
see that the structure of the argument (which is an argument for the idea
of perfection, by the way) is 100% adaptationist; it has exactly the same
structure that Lewontin has criticised is, for example, his paper
'Adaptation' (several published sources. Mine is R. Levins & R. Lewontin.
1985. The Dialectical Biologist. Harvard UP).

Well, sorry because this is getting already too long. Just let me add
that there are several other areas where data and theories from other
disciplines may be used to challenge some misconceptions within the
Chomskyan program (by the way, this is the 'label' I would favor): one is
the function of the language faculty and its roel within a general theory
of communication (also from the point of view of evolution; J. Ginzburg
has some interesting proposals here in his paper 'The evolution of
Communicative Interaction Systems'. In J.L. Dassalles et al., eds.
Proceedings of the Paris Conference on the Evolution of Language.);
another could be, as somehow Andreas hints, the epistemological priority
of 'processing' over 'acquisition'.

To conclude, and picking up a thread I left loose above, this collection
of misconceptions and incoherences, in my opinion, are doing a very bad
service to the development of a linguistic science that can truly have
its roots on cognitive theories and evidences. it only favors conceptions
of language that are completely detached from cognition (e.g., Katz's
platonism).

This is all very loose, but I hope it provides some of the suggestions
Andreas requested.

Best,

Sergio

NOTE: This is the first time I express these ideas in written form, so I
apologize if they are not very well structured. In addition, I should
acknowledge the contributions from Raquel Fernández to most of what I
said here, which is part of a long-term, joint, unfinished project. She
should not be blamed, however, from any of the inadequacies and
inaccuracies in my presentation.

Andreas Kathol wrote:

> Dear all,
>
> I hate to raise some issues that will make poor Carl spend even more
> time responding to postings to this list ..., but here it goes:
>
> I'd like to inject another thought into this discussion, which I
> believe has been missing so far. If we want to understand why it is
> that Chomskyan linguists have been playing fast and loose with formal
> rigor and citation etiquette, I don't think it's because P&P attracts
> a greater proportion of jerks than other frameworks. Rather, I'd like
> to submit that it has to do with the fact that Chomskyan linguistics
> is at some level engaged in a different enterprise than many of its
> alternatives. More than questions such as whether LF or traces exist,
> the Chomskyan enterprise is characterized by the persistent belief in
> the innateness hypothesis and its implementation in the form of the
> principles and parameters model. Part of what has attracted so many
> people to the P&P model (and keeps them loyal to it, even in times of
> great intellectual uncertainty) is that as a practitioner of Chomskyan
> linguistics, one is made to believe that one is part of a Grand Quest
> for the very essence of what has been portrayed to be the Great
> Mystery, namely what makes language learnable. For many people, lack
> of formal precision (or even empirical mispredictions) seem to become
> relatively insignificant in light of the enormity of the task of
> uncovering the Big Picture behind human language. I think this also
> explains why so few Chomskyans are particularly worried about the
> time-dependent nature of truth (which Carl has pointed out).  If
> something is proposed/discovered as part of the Grand Quest, then for
> a lot of Chomskyan this immediately seems to endow it with a veneer of
> scientific legitimacy, whereas if the same proposal/discovery is made
> in a different context that does not pursue the greater purpose of the
> Chomskyan enterprise, then it looks accidental and without the stamp
> of noteworthiness that only a greater purpose can bestow.
>
> I'm not saying this to condone the reprehensible behavior that has
> been discussed on this list; I just want to point out to what extent
> human nature and the power of the communual experience of pursuing
> so-called "deep questions" have shaped the intellectual culture that
> has been bemoaned in the recent postings.
>
> At the danger of stepping on the toes of those adherents of HPSG that
> actually (still) harbor sympathies for the P&P view of linguistic
> knowledge, I think that Chomskyan linguistics will only begin to lose
> market share when it becomes clear that the P&P model is truly
> untenable as a theory of language acquisition. At that point, it will
> no longer be possible to defend the kind of hypertrophied analyses
> that characterize P&P/MP by appealing to innate linguistic knowledge
> that would make those analyses cognitively plausible. Chomskyan
> linguistics seems to have conceded long ago that it doesn't have
> anything interesting to say about language processing, but it still
> desperately clings to language acquisition as the one source for any
> claims re. cognitive plausibility.
>
> The upshot of this is that more than pointing out mispredictions or
> lack of formal precision, the real challenge (and one that Chomskyans
> of influence may be more likely to take seriously) will come from
> evidence that the P&P model is neither necessary nor sufficient for
> explaining language acquisition. I'm aware of some work that takes on
> this task from inside linguistics (for instance Culicover's recent
> work), but I suspect that ultimately it'll have to come from outside
> of linguistics, i.e., such fields as psychology and neuroscience. I
> for one would like to have a better sense of the work in those areas
> that already poses a direct challenge to the innateness hypothesis and
> P&P. Suggestions?
>
>   --A

--
_______________________________________________
Sergio Balari Ravera
Universitat Autònoma de Barcelona
Àrea de Lingüística General
Departament de Filologia Catalana
Facultat de Lletres, Edifici B
E-09193 Bellaterra (Barcelona)
Spain
Phone: +34 935 812 350 Fax: +34 935 812 782
_______________________________________________



More information about the HPSG-L mailing list