Chess vs. psychology of chess

Esa Itkonen eitkonen at UTU.FI
Tue Feb 25 12:10:48 UTC 1997


The discussion that took place on Funknet in January was not quite
as exhaustive as it seemed first. The 'chess = language' analogy came
up, but its real significance did not. In some respects this is a
bad analogy, and in others it is a good one. Because chess pieces,
unlike nouns and verbs, lack referential meanings (yes, verbs too
'refer'), chess it not of much use if one wishes to elucidate the
form - meaning relationship; nor is it of any use if one wishes to
elucidate the notion of functional motivation. (Instead, any genuine
instrument will do; but notice that the 'instrument = language'
analogy too has its pitfalls.) On the other hand, chess is a good
analogy, if one wishes to draw attention to the fact that any given
language (just like chess) is constituted by a set of socially valid
or public NORMS.

There is only ONE set of norms (or rules) of chess, which means that
all chess-players share the knowledge of this set: it is the object
of their common knowledge. And then there are millions and millions
of internalizations of this set of norms, i.e. exactly as many as
there are chess-players. Thus, chess as a social and normative
entity (= one) is distinct from the psychology of chess (=
millions). You have to know chess (although you need NOT have some
complete theoretical description of it), before you can start to
investigate the psychology of chess. This is a very simple truth
(and yet my personal experience of more than 25 years has taught me
that most linguists do not understand it). In any case, in the
Funknet discussion this simple truth was perfectly captured by the
person who claimed that investigating the psychological (or
biological) foundation of X presupposes that one understand X. Just
try to investigate HOW the Finnish relative clause is mentally
processed or is stored in the psychologically real, unconscious
competence (or, for that matter, HOW it has changed during the past
1000 years), without knowing WHAT is the Finnish relative clause.
This distinction between WHAT (= norm) and HOW (and, ultimately, WHY)
(= internalization-of-norm) is the cornerstone of my 1983 book
'Causality in Linguistic Theory'.

What is the relevance of all this to the Funknet discussion? The
relevance lies in the fact that, contrary to the lessons that might
be learned from the 'chess = language' analogy, the primacy of the
social-normative dimension is ignored in today's cognitive
linguistics (just try to find a systematic account of normativity in
the currect cognitivist literature); and it is flatly denied in
generativism (cf. Chomsky 1976: "As for the fact that the rules of
language are 'public rules', this is, indeed, a contingent fact.")
It is in order to oppose this line of thinking that, instead of
loose chess-type analogies, I have made use of Wittgenstein's
private-language argument in my 1978 book 'Grammatical Theory and
Metascience', as Kripke has in his 1982 book 'Wittgenstein on Rules
and Private Language'.

But why should this be brought up just now? For the following
reason. In the Funknet discussion on form and meaning it turned out
that nowadays functionalism (and even non-discreteness) is part and
parcel of generativism, and that - somehow - this has in 'reality'
always been so. (Never mind Chomsky's 1976 statement to the contrary:
"To account for or somehow explain the structure of UG, or of
particular grammars, on the basis of functional considerations is a
pretty hopeless prospect, I would think; it is, perhaps, even
'perverse' to assume otherwise.")

This had me worried. I suddenly remembered Hilary Putnam mentioning
a private discussion in which Chomsky had toyed with the idea that
maybe "our idealized or 'competence' description is a description of
correct thinking in the normative sense" (see "Reflective
Reflections", p. 216 in S. Silvers (ed.): 'Rerepresentation',
Dordrecht: Kluwer, 1989). Of course, this is exactly what the
'competence' description is (which means, in fact, that it is NOT
about any genuine or psychologically real competence); it is just
that this has been officially denied so far (for details, see my 1978
and 1983 books). However, if functionalism and non-discreteness are
any guide, this may soon change. Maybe it is changing right now. All
discoveries have 'always already' (= 'immer schon' as Heidegger and
the later Husserl would say) been made by generativists. It
seems to be an inherent feature of the very notion of 'discovery'
that, in a timeless sense, the (true) discoveries have been, are
being, and will be  made by generativists.

Does there remain any notion or issue which generativism just
cannot claim as its own? At least until recently, ANALOGY was
precisely such an issue. But who knows what will happen? Therefore,
before the wind turns and analogy too becomes a generativist
property, it might be of some interest to some people to get
acqainted with a recent rehabilitation (and formalization) of
analogy in syntax. It can be found in a paper by Jussi Haukioja and
myself. (Miraculously, some off-prints are still available; write
to the following address: jhaukioj at utu.fi )


Esa Itkonen



More information about the Funknet mailing list