The Sleeping Chess Player

To the Editors:

An important historical point in John Searle’s excellent article on Chomsky’s linguistics [NYR, June 29] needs correction, not only because it seems to have misled Searle, Quine, and other philosophers, but also because of its central bearing on the doctrine of innate ideas upheld by Chomsky and attacked by empiricists and behaviorists. Searle quotes a passage from Leibniz in which it is said that “ideas and truths are for us innate, as inclinations, dispositions, habits, or natural potentialities, and not as actions….” Searle then comments that if this is the “correct model” for innate ideas, as Chomsky has said that it is, “then at least some of the dispute between Chomsky and the empiricist learning theorists will dissolve like so much mist on a hot morning. Many of the fiercest partisans of empiricist and behaviorist learning theories are willing to concede that the child has innate dispositions, inclinations, and natural potentialities.”

This assimilation of empiricism and behaviorism to Leibniz’s innatism is, however, based on a misunderstanding of the latter’s statement. As his further discussions in the New Essays show, Leibniz draws two distinctions where the empiricists draw only one. Like Locke and other empiricists, Leibniz distinguishes between powers and acts, i.e., between the capacity for having or perceiving ideas and the actual occurrence or perception of ideas. But unlike the empiricists, Leibniz also distinguishes between powers and dispositions or potentialities (Leibniz’s French word here is virtualités). The difference between these is that powers are passive, indeterminate, and remote, while dispositions, in Leibniz’s view of them, are active, determinate, and proximate. Powers as such require the stimulation of external objects both in order to be activated and in order to receive their perceptual or ideational contents; hence, they have no specific contents of their own. Dispositions, on the other hand, already have determinate contents which the mind can itself activate, given appropriate external occasions. Both powers and dispositions may be called “capacities,” but then they are capacities of two quite different sorts. The difference may be illustrated by the way in which both the normal human infant and the sleeping, avid chessplayer may be said to have the “capacity” to play chess. In these terms, then, Leibniz and Chomsky affirm, and the empiricists and behaviorists deny, that the human mind has dispositions and not mere powers, and that basic ideas and linguistic rules are had in the latter way and not in the former.

The distinction among powers, dispositions, and acts is crucial to Leibniz’s view of substance, including the dynamic substances studied by physics; and since he regards the mind as a substance to which innate ideas pertain as dispositions, the distinction is also crucial for the understanding of innate ideas. It is this Leibnizian view of substance which seems to me to underlie Chomsky’s conception of the mind as having inherent linguistic competences. In Leibniz’s important essay On the Emendation of First Philosophy and the Concept of Substance, he writes that all substances have “active force” (vis activa) which differs from the “bare power” (potentia nuda) which the Scholastics had held to characterize substance. “Active force contains a certain determinate actuality or activity; it is intermediate between the capacity for acting and action itself and it involves dynamic effort (conatus), so that it is brought into operation by itself, nor does it need external aids but only the removal of impediments. This may be illustrated by the example of a heavy hanging body, or a tense bow.”

According to this model, then, for ideas to be innate as dispositions means that the mind has quite determinate contents of its own which it is itself able to activate and perceive; whereas for ideas to be innate merely as powers would mean that the mind has only diffuse mechanisms whose contents are exhaustively derived from the impact of external stimuli. As Leibniz frequently emphasizes, the latter model, unlike the former, is unable to explain how the mind can attain the sorts of necessary and universal truths found in logic, mathematics, and other disciplines. And a comparable sort of necessity and universality are attributed by Chomsky to the basic rules of grammar.

This conception of the mind and of innate ideas as involving dynamic dispositions is clearly brought out in the following passage by Leibniz in his elaborate critique of Locke’s Essay Concerning Human Understanding:

Philalethes (i.e., Locke): But suppose there are truths which could be imprinted upon the understanding without its perceiving them; I do not see how, in relation to their origin, they could differ from the truths which it is only capable of knowing.

Theophilus (i.e., Leibniz): The mind is not only capable of knowing them; but further of finding them in itself; and if it had only the simple capacity of receiving knowledge, or the passive power therefore, as indeterminate as the wax has for receiving figures and the blank tablet for receiving letters, it would not be the source of necessary truths, as I have just shown that it is; for it is incontestable that the senses do not suffice to show their necessity, and that thus the mind has a disposition (active as well as passive) to draw them from its own depths; although the senses are necessary to give it the occasion and attention for this. [New Essays, pp. 80-81]

Leibniz then goes on to say that, in its relation to necessity, innate truths, the mind “is not, then, a naked faculty which consists in the mere possibility of understanding them; it is a disposition, an aptitude, a preformation, which determines our soul and which makes it possible for them to be derived from it. Just as there is the difference between the figures which are given to the stone or the marble indifferently and those which its veins already indicate, or are disposed to indicate, if the workman profits by them.”

Searle’s subsequent discussion captures part of Leibniz’s distinction between powers and dispositions when he emphasizes that, according to Chomsky, the child “must have a specific set of linguistic mechanisms at work.” But the Leibnizian background, correctly understood, gives much stronger support to Chomsky’s innatism. Far from being compatible with empiricist and behaviorist learning theories, as Searle and Quine hold that it is, Leibniz’s doctrine shows how the mind can itself be the exhaustive source of its linguistic competence, for which external stimuli serve only as occasions for activating what is already dispositionally contained in the mind’s own structure. Leibniz’s doctrine therefore explains, as the behaviorist theory cannot, the necessity and universality of the linguistic rules for forming and interpreting sentences, to which Chomsky’s theory has called emphatic attention.

Alan Gewirth

Department of Philosophy

University of Chicago

John Searle replies:

I agree with Professor Gewirth that there is a distinction to be found in Leibniz between powers and dispositions. But I fail to see that this “Leibnizian background, correctly understood, gives much stronger support to Chomsky’s innatism.” Chomsky’s innatism is an empirical hypothesis that will have to be verified or refuted on empirical evidence; its truth or falsity does not derive from its Leibnizian ancestry.

Furthermore, as Gewirth explains the distinction between powers and dispositions, neither term is quite appropriate for Chomsky’s own innatism. For Chomsky, the human mind prior to an exposure to language is not quite like either that of the infant, who has the “capacity” to play chess only in the sense that he will someday be capable of learning how, or that of the sleeping chess player, who already knows how. Obviously the human child unexposed to a language does not yet know how to speak a language; whereas the sleeping chess player does know how to play chess. The child’s relation to language, according to Chomsky, is unlike either the infant’s or the sleeping chess player’s relation to chess in that the child has a language-specific capacity to acquire a language, and that capacity is such as to determine the form of the grammar. I think in fact that Chomsky’s view is close to Leibniz’s notion of a disposition but I cannot accept all of Gewirth’s explanation of the notion of a disposition.

Finally Gewirth confuses my view with Quine’s. Quine holds that Chomsky’s innatism is compatible with behaviorist and empiricist learning theories. I don’t. I maintain that the claim that the child has language-specific mechanisms that determine the form of the grammar goes far beyond traditional empiricist and behaviorist learning theories.