Philosophical Review

A Note on the Creative Aspect of Language Use

Author(s): Noam Chomsky
Source: The Philosophical Review, Vol. 91, No. 3 (Jul., 1982), pp. 423-434
Published by: Duke University Press on behalf of Philosophical Review
Stable URL: https://www.jstor.org/stable/2184692
Accessed: 02-08-2019 18:37 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Philosophical Review, Duke University Press are collaborating with JSTOR to digitize,
preserve and extend access to The Philosophical Review

This content downloaded from 200.16.89.220 on Fri, 02 Aug 2019 18:37:32 UTC
All use subject to https://about.jstor.org/terms
 The Philosophical Review, XCI, No. 3 (July 1982)

A NOTE ON THE CREATIVE ASPECT OF

LANGUAGE USE

Noam Chomsky

In her article "The Creative Aspect of Chomsky's Use of

the Notion of Creativity" (Philosophical Review, January
1981), Margaret Drach accuses me of having performed "a
prestidigitation trick," of various contradictions, of changing
fundamental concepts "without warning or acknowledge-
ment" after they had served their propaganda function, and
of numerous other sins. A careful look, however, reveals
serious errors of reasoning and gross misreading of the very
passages she cites; I will keep to these for the most part, since
they suffice to refute her allegations. A few examples follow,
but first, a brief review of the theses that Drach considers and
an analysis of her basic argument.
One thesis is that:1

(1) "A person who knows a language has mastered a sys-

tem of rules that assigns sound and meaning in a defi-
nite way for an infinite class of possible sentences."

1The quotes (1), (2), and (Q) are from my "Form and Meaning in
Natural Language," in Language and Mind, enlarged edition (New York,
1972). This is identified in the preface as "a rather informal lecture given
in January 1969.. ." Virtually all of Drach's reference to Language and
Mind are, in fact, references to this "rather informal lecture," including all
the quotes from this book cited below with the exception of (8). See Drach's
article for exact page references for these and other passages cited below.
Note that Language and Mind post-dates the "decisive shift" Drach claims
to have discovered in my views on the two concepts she discusses: the
creative aspect of language use (CALU) and competence. With regard to
the CALU, she writes: this "decisive shift .. . occurs in Cartesian Linguistics"
(New York, 1966); and she claims that my "definition of 'competence'
underwent a change at about the same time..." The earliest work from
which Drach quotes is "Current Issues in Linguistic Theory," in The Struc-
ture of Language, ed., J. A. Fodor and J. J. M. Katz (Englewood Cliffs,
1964), henceforth CI. Therefore, the period through which "the concept
[of creativity] seems to have run its course," having been used (by implica-
tion, dishonestly) to "gather countless repentant behaviorists to the
Chomskyan fold," was rather brief: about two years. In fact, as we shall see
directly, the elapsed time was null.

423

This content downloaded from 200.16.89.220 on Fri, 02 Aug 2019 18:37:32 UTC
All use subject to https://about.jstor.org/terms
 NOAM CHOMSKY

This is intended as a first step towards an answer to one

central question of linguistics:

(Q) "What is the nature of a person's knowledge of his

language, the knowledge that enables him to make
use of language in the normal creative fashion?"

The "normal creative fashion" of language use involves un-

boundedness, novelty, freedom from stimulus control, co-
herence and appropriateness to situations. The word "en-
ables" is to be understood, as Drach correctly observes, in the
sense of providing "a necessary, but not a sufficient condi-
tion."
A second thesis is that:

(2) "We do not understand, and for all we know, we may

never come to understand what makes it possible for a
normal human intelligence to use language as an in-
strument for the free expression of thought and feel-
ing."

A third thesis, this time in Drach's words, is that:

(3) There is "some very basic and revealing link between

what was being done in transformational linguistics
and the 'creative aspect of language use' [CALU]; that
the latter was (or should be) somehow very intimately
related to the concerns of linguistics.... [In Choms-
ky's later writings] there runs a persistent suggestion
(not to say, at times, an explicit claim) that generative
grammar is in some important way concerned with
the CALU-in spite of a no less consistent emphasis
on its being competence, not performance, that
grammar is concerned with."

Thesis (3) (which, again, is Drach's formulation) is implicit in

(1) and (Q). That is, if knowledge of language is mastery of a
grammar, and knowledge of language makes possible the
CALU, then some version of (3) is plausible.
But, Drach claims, thesis (2) denies (3), therefore (1) or the
presupposition of (Q). She writes: "without any indication
that the goal he had set himself, or one facet of the enter-

424

This content downloaded from 200.16.89.220 on Fri, 02 Aug 2019 18:37:32 UTC
All use subject to https://about.jstor.org/terms
 CREATIVE ASPECT OF LANGUAGE USE

prise, had not panned out, this verdict: the creative aspect of
language use is as much a mystery as it ever was." This is the
"cheating" to which Drach refers: in early work (1964) I had
claimed that generative grammar had "made [a] contribution
to, or thrown light on" the CALU, but later (1969) I stated
that the CALU remains a "mystery."
We see at once, however, that there is neither contradic-
tion nor deception in these various theses. The CALU does
remain a mystery, as stated in (2). The study of grammar
attempts to answer question (Q) (inter alia) by providing a
substantive version of (1). Insofar as it succeeds in this aim,
we have an account of the mechanisms that enter into the
CALU. In this sense-which has been clear from the
outset-the thesis that Drach formulates as (3) is correct:
there is an "intimate relation" between the CALU and the
'concern of linguistics" to discover the mechanisms of
grammar, though of course grammar is concerned with
competence; and "generative grammar is in some important
way concerned with the CALU." This remains so even
though the CALU remains a mystery (cf. (2)). Evidently,
there is no contradiction here. There would be a contradic-
tion if, in Drach's phrase, "language is what Chomsky says it
is-something that is accounted for in toto by the rules of
transformational grammar," if the CALU had been "said to
be accounted for by his rules and then reestablished as a
mystery." But what I actually said is something quite dif-
ferent: that the rules and principles of grammar "provide the
means" for the CALU, thus shedding light on it, but not
giving anything like a full account of it, and not resolving the
mysteries it poses.
Drach considers this resolution of her various "contradic-
tions," but rejects it, on curious grounds. She cites my obser-
vation that there is no inherent contradiction in the notion of
creativity constrained by rules:2

2j also distinguished between the CALU and "true 'creativity' in a higher

sense," discussing also "rule-changing creativity" as well as the creative use
of particular rules and canons. See Cartesian Linguistics and Language and
Mind.

425

This content downloaded from 200.16.89.220 on Fri, 02 Aug 2019 18:37:32 UTC
All use subject to https://about.jstor.org/terms
 NOAM CHOMSKY

(4) "There is no contradiction in this, any more than

there is a contradiction in the insistence of aesthetic
theory that individual works of genius are constrained
by principle and rule. The normal, creative use of
language, which to the Cartesian rationalist is the best
index of the existence of another mind, presupposes a
system of rules and generative principles."

But there is a "problem," Drach believes; namely, "that the

rules he provides leave untouched what is by his own avowal
the most important aspect of the creativity: the coherence
and appropriateness of ordinary speech." That is indeed a
problem, but it is not my problem. Rather, it is a point that I
insistently stressed, in the passages that Drach cites and else-
where. We can formulate the principles and rules-at least,
we can approach this task with increasing success-but that
will still leave fundamental aspects of the CALU a mystery,
though success in the task that seems within reach will yield
insight into the mechanisms that are crucially involved in the
CALU and that make it possible.
Drach states, however, that success in this venture "leaves
the CALU out completely." This is obviously false. An im-
portant connection is established between grammar and the
CALU if we can show how grammar "provides the means"
for the CALU, how mastery of a system of rules enables (in
the sense of providing a crucial necessary condition) the
speaker-hearer "to make use of language in the normal crea-
tive fashion." In fact, Drach recognizes the falsity of her cen-
tral conclusion, writing: "Still, it would seem that under-
standing the mechanisms that make something possible
should go some way toward dispelling the mystery that sur-
rounds it." It is true that understanding the mechanisms goes
some way towards shedding light on the CALU so that
Drach's major conclusion is false as she here concedes, but it
still leaves intact what I called the "mystery" in passages she
cites; see below.
Drach argues that what "legitimately emerges" from my
account is that "the role of language in the creative perfor-
mance of the speaker is no more essential than that of

426

This content downloaded from 200.16.89.220 on Fri, 02 Aug 2019 18:37:32 UTC
All use subject to https://about.jstor.org/terms
 CREATIVE ASPECT OF LANGUAGE USE

brushes in the painter's creative performance"; thus on my

account, she alleges, "the role of language in the CALU is
parallel to that of brushes in painting, or the CABU." Her
argument appears to be that since brushes play a role in
painting, it therefore follows that the study of the principles
and rules involved in some "creative" performance (but see
note 2) can in principle shed no more light on this perfor-
mance than the study of brushes sheds on "the painter's crea-
tive performance." In particular, the study of aesthetic can-
ons, conventions, principles and rules can provide no under-
standing of the work of the artist (cf. (4), though the analogy
drawn there should evidently not be pressed too far (see
note 2, and observe also that elucidation of the rules and
principles that make possible the CALU is quite different in
other respects from discovery of specific aesthetic canons), a
conclusion that is plainly incorrect. And the study of the
mechanisms of grammar yields no understanding of the
CALU. Or to construct an "argument" of the same sort closer
to the point, since the tongue musculature enters into lan-
guage use but yields no insight into the CALU, therefore the
study of the mechanisms of grammar yields no insight into
the CALU. This argument can hardly be taken seriously, but
perhaps there is no point in considering it further since, as
noted above, Drach recognizes that it is invalid, and that her
central conclusion concerning the CALU and the mech-
anisms of grammar is false.
Elsewhere Drach writes: "Having added these properties to
the CALU, it must at one point have become painfully clear
to Chomsky-as indeed it was to the reader-that his concept
of competence and the generative grammar that was sup-
posed to characterize it had nothing to say about them." The
central property that was allegedly added-"with particular
emphasis," Drach states-is "appropriateness." Note that this
property was "added" in the earliest passage she cites; cf. (6)
below. Furthermore, as already noted, it would be wrong to
conclude that the "concept of competence and generative
grammar" has "nothing to say about" the CALU, though it
has nothing to say about appropriateness of use, freedom
from stimulus control, etc. Finally, nothing new became

427

This content downloaded from 200.16.89.220 on Fri, 02 Aug 2019 18:37:32 UTC
All use subject to https://about.jstor.org/terms
 NOAM CHOMSKY

"painfully clear," since it was all clear, and not painfully, in

the earliest passages Drach cites, and consistently through-
out.
The basic confusion that runs through Drach's account can
be summarized as follows. She begins with my observation
that (A) the study of grammar can bring to light the
mechanisms that enter into the CALU. She concludes from
(A) that I have claimed that (B) all of language, including the
CALU in its entirety, "is accounted for in toto by the rules of
the transformational grammar." Then, she continues, after
the latter claim had "more than done the job it was called
upon to do-gather countless repentant behaviorists to the
Chomskyan fold," I silently withdrew it, noting that (C) the
CALU remains a mystery. Given (C), it then follows, she
concludes, that (D) my "account of language and of the
knowledge of it leaves the CALU out completely." The er-
rors of reasoning are transparent. (B) does not follow from
(A) (nor have I ever claimed that (B)), and (D) does not
follow from (C). Furthermore, I have held to the same posi-
tion throughout, namely, (A) and (C), along with the obvious
thesis (which, as noted above, Drach accepts) that (E) exhibit-
ing the grammatical mechanisms involved in the CALU
sheds important light on it, while not accounting for it. Fur-
thermore, all of this is clear and explicit in the passages she
cites.
Drach claims to be "confused" by such conjunctions of pas-
sages as (2), and a few pages later, (5) (see note 1):

(5) "We cannot now say anything particularly informative

about the normal creative use of language in itself. But I
think that we are slowly coming to understand the
mechanisms that make possible this creative use of lan-
guage, the use of language as an instrument of free
thought and expression" (emphasis added).

Drach omits the first sentence of (5) here italicized. She then
states that the second sentence "seems to be in direct con-
tradiction" to (2). Note that the sentence she omits recapitu-
lates the relevant part of (2). A more accurate version of her
statement, then, would be that the second sentence of (5)

428

This content downloaded from 200.16.89.220 on Fri, 02 Aug 2019 18:37:32 UTC
All use subject to https://about.jstor.org/terms
 CREATIVE ASPECT OF LANGUAGE USE

"seems to be in direct contradiction" to the first sentence of

(5) which she omitted. But it is quite clear that there is no
contradiction, and no reason to be confused-particularly,
when the sentence Drach omitted is restored. We can come
to understand the mechanisms of grammar that provide a
crucial necessary condition for the CALU while we do not
(and may never) understand what makes it possible for hu-
mans to employ these mechanisms as they do. This is a sim-
ple and straightforward point, which Drach appears to mis-
understand throughout.
Let us now turn to Drach's account of how I allegedly
made strong claims to entice the unwary behaviorist to the
fold, then silently withdrew them when the job was done.
The earliest passage she cites (from CI), to which she re-
peatedly refers in support for her claims, is this:

(6) "The central fact to which any significant linguistic

theory must address itself is this: a mature speaker
can produce a new sentence of his language on the
appropriate occasion, and other speakers can under-
stand it immediately, though it is equally new to them.
Most of our linguistic experience, both as speakers
and hearers, is with new sentences; once we have mas-
tered a language, the class of sentences with which we
can operate fluently and without difficulty or hesita-
tion is so vast that for all practical purposes (and, ob-
viously, for all theoretical purposes), we can regard it
as infinite. Normal mastery of a language involves not
only the ability to understand immediately an indefi-
nite number of entirely new sentences, but also the
ability to identify deviant sentences and, on occasion,
to impose an interpretation on them ... it is clear that
a theory of language that neglects this 'creative' aspect
of language is of only marginal interest .... Clearly the
description of intrinsic competence provided by the
grammar is not to be confused with an account of
actual performance.... Nor is it to be confused with
an account of potential performance. The actual use
of language obviously involves a complex interplay of
many factors of the most disparate sort, of which the

429

This content downloaded from 200.16.89.220 on Fri, 02 Aug 2019 18:37:32 UTC
All use subject to https://about.jstor.org/terms
 NOAM CHOMSKY

grammatical processes constitute only one. It seems

natural to suppose that the study of actual linguistic
performance can be seriously pursued only to the ex-
tent that we have a good understanding of the genera-
tive grammars that are acquired by the learner and
put to use by the speaker or hearer.

Drach cites these passages, apart from the final sentence.

From them, she concludes that I am referring the CALU to
competence, not performance. In fact, the passages are per-
fectly explicit in attributing the CALU to performance, not
competence; the CALU is an aspect of the "use of language"
(i.e., performance), which is "not to be confused" with com-
petence. (Note also that the passage stresses at once the rele-
vance of "appropriateness" to the CALU; Drach acknowl-
edges this, but dismisses it, claiming that this notion was only
"added" in later discussion). Drach's sole reason for her claim
that these passages refer the CALU to competence is that the
phrase "the creative aspect of language" (her CAL) appears.
But surely this is a curious way to read. The passage is
explicit that it is language use that is being discussed, and
stresses the importance of distinguishing the study of compe-
tence from the study of language use, which obviously in-
volves "factors of the most disparate sort" apart from the
mechanisms of grammar. The term "language" is clearly
nontechnical in this context; the term is surely used conven-
tionally with sufficient breadth to include language use, and
the passage leaves absolutely no doubt as to the intent, which
Drach grossly misconstrues. The conclusion, then, is that in
the earliest passage Drach cites, I was quite explicit in distin-
guishing competence from performance and attributing the
CALU to performance. The same is true throughout.
Some of Drach's readings are really quite remarkable. For
example, in a further effort to show that in early work I re-
ferred the CALU to competence, not performance, she cites a
passage that "clinches it-it is linguistic competence." The
passage reads:

(7) "The most striking aspect of linguistic competence is

what we may call the 'creativity of language,' that is,

430

This content downloaded from 200.16.89.220 on Fri, 02 Aug 2019 18:37:32 UTC
All use subject to https://about.jstor.org/terms
 CREATIVE ASPECT OF LANGUAGE USE

the speaker's ability to produce new sentences, sen-

tences that are immediately understood by other
speakers although they bear no physical resemblance
to sentences which are 'familiar.' The fundamental
importance of this creative aspect of normal language
use has been recognized since the seventeenth century
at least" (emphasis added).

Note that the passage refers explicitly to language use, that is,
to performance. It immediately follows discussion of the
"distinction that must be made between what the speaker of a
language knows implicitly (what we may call his competence)
and what he does (his performance)." The aim of a "linguistic
grammar" is "to discover and exhibit the mechanisms that
make this achievement possible," namely, achievement of the
CALU, where "make possible" is to be understood as before,
clearly. Drach might have made the point that reference to
performance as an "aspect of linguistic competence" might
have been misleading, despite the immediate clarification
and the preceding discussion of the nature and importance
of the distinction. But it is strange indeed to cite a reference
to the creative aspect of "language use" as a "clinching" ar-
gument that I am referring "not [to] language use," but to
competence, in discussing the CALU.
Comparable misreadings and confusions run through
Drach's entire article. For example, she writes that "It is the
creativity of human language, so eloquently extolled by
Chomsky and, in the eyes of his followers, decisively demon-
strated by his generative-transformational linguistics ... ." It
is difficult to imagine that anyone has taken the CALU to be
"decisively demonstrated by generative-transformational
linguistics"; a reference would have been helpful. To my
knowledge, the CALU has always been presented as in effect
an observation, and far from a novel one. The goal of
"generative-transformational linguistics" was, in the first
place, to exhibit the mechanisms that enter into the CALU,
and, more significantly, to discover the basis in innate en-
dowment for the development of these mechanisms in the
case of particular languages.
To cite another case, related to the latter concern, Drach

431

This content downloaded from 200.16.89.220 on Fri, 02 Aug 2019 18:37:32 UTC
All use subject to https://about.jstor.org/terms
 NOAM CHOMSKY

claims that I am evasive about whether human behavior can

be modeled by automata, thus allowing too wide a variety of
"converts to Chomskianity." As proof, she cites such passages
as those given above, which observe that the CALU remains a
mystery, while other passages, she claims, support the con-
clusion "that Descartes and his followers simply lacked the
foreknowledge to imagine sophisticated present-day com-
puters." She cites two passages which might, she claims, sup-
port this "logical conclusion":

(8) "There would be no difficulty, in principle, in design-

ing an automaton which incorporates the principles of
universal grammar and puts them to use to determine
which of the possible languages is the one to which it is
exposed."

(9) "This is not to deny that the method of explanation

suggested by La Mettrie may in principle be correct."

Consider first (8). This states that the problem of selecting

a grammar on the basis of evidence, given the principles of
universal grammar, might in principle be modeled in an
automaton. But this in no way conflicts with the belief that
the CALU cannot be modeled in an automaton, a point that
is quite obvious, and that is emphasized in the references
Drach cites, even in a passage she cites: namely, that among
the questions "that appear to be within the reach of ap-
proaches and concepts that are moderately well under-
stood-what I will call 'problems'," I class "questions of lin-
guistic competence and of language acquisition" (her phrase,
my emphasis); while among the questions that "remain as
obscure to us today as when they were originally
formulated-what I will call 'mysteries'," I class "questions of
performance" (her phrase; more accurately, questions of
production).
Consider next (9), a footnote added to a discussion of how
La Mettrie failed to consider "the problem raised by Des-
cartes," namely, the problem of the CALU. The footnote
merely expresses the open-mindedness that any rational per-
son should maintain when faced with an entirely open ques-
tion, one that remains a "mystery." To state that some form

432

This content downloaded from 200.16.89.220 on Fri, 02 Aug 2019 18:37:32 UTC
All use subject to https://about.jstor.org/terms
 CREATIVE ASPECT OF LANGUAGE USE

of "mechanism" might in principle be correct does not yield

the "logical conclusion" that the CALU can be modeled by a
computer program. To express doubts as to whether the
CALU can be so modeled is not to put forth the claim that
these doubts are unshakeable truth.
There are other examples, but perhaps this sample suf-
fices.
Drach finds my "appeal to Descartes's doctrine of innate
ideas" baffling, indeed, "doubly baffling."3 Note first that
there was no such "appeal," another curious misreading.
Why is my allusion to this doctrine "doubly baffling"?
Drach's reason is (i) that Etienne Gilson states that the Car-
tesians rendered "the existence of human language
philosophically incomprehensible and its very possibility in-
conceivable," and (ii) that he stated that Descartes's doctrine

3Drach claims "that critics have as much trouble pinning Chomsky down
concerning the meaning of his doctrine of innateness as it was difficult to
do so in the case of Descartes," citing no examples or reasons. Putting aside
the interpretation of Descartes, I am unaware of any difficulties in pinning
down the meaning of my views on innateness. Evidently, there is some
element of biological endowment that enables a human, but not a bird or
(so far as is known) a higher ape, to acquire a human language. This
"doctrine of innateness" is certainly not difficult to "pin down," and is in
fact hardly controversial. I also tend to believe, as do many others, that
there are some "special purpose devices" involved in what I think should
properly be called "the growth of grammar." This is, plainly, an empirical
issue, and the "doctrine" (i.e., the belief, for which there is fairly good
evidence) is again not difficult to "pin down." I have also made proposals,
as have many others, about what these devices might be, abstractly charac-
terized as properties (specifically, principles) that it seems plausible to at-
tribute to an innate "language faculty," realized somehow in as yet un-
known physical mechanisms. If the meaning of this "doctrine" is difficult
to "pin down," then the same must be true of similar studies of the visual
system (and in this regard, the reference to Descartes, and to current
research, is quite appropriate, as discussed in detail in references that
Drach cites).
In fact, there is no more difficulty in understanding what Drach calls my
"doctrine of innateness" than there is in understanding the "doctrine" that
the human and insect visual systems develop on the basis of distinct biolog-
ical endowment, or the "doctrine" that the human visual system incorpo-
rates a rigidity principle that is used in visual processing and is somehow
physically realized, etc. The belief that there is something deeply puzzling
here has been frequently expressed, but as yet, no one has succeeded in
explaining what the puzzle or problem might be, or why the puzzle (what-
ever it is) arises peculiarly in the case of language.

433

This content downloaded from 200.16.89.220 on Fri, 02 Aug 2019 18:37:32 UTC
All use subject to https://about.jstor.org/terms
 NOAM CHOMSKY

of innate ideas enlists that "side of his personality where he is

not truly himself, and to put it bluntly, one of his defects"-
he appealed to innate ideas to show how the mind could
''grasp the truths of religion"; "This is what the doctrine of
innateness was to allow him to do," Drach states, citing Gil-
son.
There is no reason to be baffled, even once. I would agree,
and have repeatedly so stated, that appeal to a second sub-
stance is not in order; in fact, Newton demonstrated that
Descartes's concept of "body" was far too limited, as is well-
known. If this is the import of (i), we may dismiss it. If (i)
refers to the CALU, however, then we may not dismiss it,
though (i) would be a most misleading way to express the
point. But while we need not follow Descartes in postulating
a second substance, nevertheless, as I pointed out in the work
to which Drach refers, some critical elements of the "Carte-
sian" framework4 can be adapted and reconstructed in ways
that make a good deal of sense-without, however, dispelling
some of the mysteries to which the Cartesians alluded. As for
the claim (whether it is Gilson's, as Drach alleges, or not) that
Descartes's notion of innate ideas was developed solely to
account for the truths of religion, that is simply false, as is
amply demonstrated in passages that I cited, which are those
relevant to my further discussion.
Drach concludes with some speculations about my "change
of mind about creativity, first said to be accounted for by
[my] rules and then reestablished as a mystery." When her
misreadings and errors are corrected, it becomes clear that
there has been no change of mind in this respect (there has
been in many other respects, as explained in the work Drach
cites and elsewhere, and I would hope that formulations be-
came clearer over the period that Drach surveys), hence no
need to speculate about the motives for it.

Massachusetts Institute of Technology

4Note that the concept "Cartesian" was explicitly given a rather broad
construal in the work Drach cites. Cf. Cartesian Linguistics, note 3.

434

This content downloaded from 200.16.89.220 on Fri, 02 Aug 2019 18:37:32 UTC
All use subject to https://about.jstor.org/terms