Fred D’Agostino, in his excellent Chomsky’s System of Ideas, lays down six ideologies that he identifies in Noam Chomsky’s linguistic writing. As I am by no means a linguist and have no authority or expertise to judge these doctrines, I will not. However, for those who are just beginning to explore Chomsky’s linguistic work, I think it helps to be aware of these six positions to really appreciate the angle he is coming in from. I will here present briefly the doctrines, as quoted from D’Agostino’s work:
- Linguistic Subjectivism
- Methodological Individualism
- Linguistic Intellectualism
- Linguistic Mentalism
- Linguistic Rationalism
1. Linguistic Subjectivism
Linguistic Subjectivism says “the properties of any human language must be those ‘that are given to it by the innate mental processes of the organism that has invented it’”. Languages are subjective, not objective, because “languages, unlike atoms, do not exist and do not have properties independently of our (largely unconscious) beliefs about them. A language has the properties we believe it to have, then, not because these beliefs are accurately representational.”
Subjectivism also tells us that “a grammar of a language describes the psychological basis of linguistic competence.”
Objectivism, by contrast, “construes languages as (possibly abstract) objects, the properties of which are not entirely determined by their users”.
2. Methodological Individualism
Methodological Individualism says “that linguistics is ‘a subfield of psychology’.” Furthermore, that “all social phenomena, such as language, are to be explained, ultimately, in terms of the characteristics of individual human beings.” To put it another way, “individuals are the only causal agents in the formation, development, and functioning of social institutions such as language.”
Some confusion may be made between Individualism and Behaviorism, but this would be a mistake. Individualism is not concerned with “overt linguistic behavior” (performance), but rather with the “psychological characteristics of language users which underlie” such behavior (competence). Chomsky has criticized traditional behaviorism, particularly the work of B. F. Skinner. Skinner’s work does not translate well to human behavior, except in very trivial ways. (Michael Devitt and Kim Sterelny say that Chomsky rejects behaviorism in favor of “its successor”, functionalism. Functionalism means that mental states — beliefs, desires, being in pain — are constituted solely by their functional role — that is, they are causal relations to other mental states, sensory inputs, and behavioral outputs. I will not address this here.)
An “individualistic methodological orientation… presupposes and implies a thoroughly subjectivistic account of language.”
The opposing view is Methodological Collectivism, where “some characteristics of social phenomena, such as language, are sui generis ["of its own kind"] with respect to the characteristics of individual human beings, and must therefore be explained in non-psychological terms.” One proponent of this view is August Schleicher (1821-1868). Collectivists accept “objective grammatical facts” “which are independent of facts about language users.” Individualists would deny these facts.
D’Agostino references the related concept of “social monadism”, a view presented by collectivist E. Gellner that says “every social event must have its habitat in the individual psyche”. D’Agostino claims that Chomsky is “committed” to such a view, because “language is a social institution which is a ‘reflection of individual chgaracteristics’, and one which has ‘its habitat in the individual psyches’ of its users.”
Social Monadism seems to be little more than a “straw man” propped up by Gellner for knocking down. As such, and because D’Agostino quickly destroys it, there is little more to say.
3. Linguistic Intellectualism
Linguistic Intellectualism claims “that linguistic phenomena [or behaviors] are rule-guided, rather than law-governed” … “since a grammar is a system of rules and not a system of natural laws.” Intellectualists would say that “linguistic behavior crucially involves interactions between language users and their environment which are best described in rational, computational terms, rather than in strictly causal terms in which the behaviour, for instance, of atoms and planets is best described.” Intellectualism also says that “intelligent behaviour is to be explained in terms of agents’ rational calculations, according to rules or maxims, in the light of their goals and their beliefs about the situation in which they find themselves.”
Intellectualism “seems… to presuppose and to be implied by dualism” and may be incompatible with materialism. Chomsky, however, has made materialist claims. He calls the mind a “biological structure” and says that “properties of the mind that underlie the acquisition of language” are “biological properties of the organism”.
D’Agostino puts Chomsky in this camp because Chomsky said a “grammar is a system of rules and principles that determine the formal and semantic properties of sentences … [which] is put to use, interacting with other mechanisms of mind, in speaking and understanding language.” D’Agostino believes that Chomsky thinks “linguistic behaviour is transcendentally free”, by which he means that “it is determined by rules which language users themselves construct.”
Opposing intellectualism is naturalism, where “intelligent behaviour is to be explained casually, in terms of the structural properties of agents and their environments.”
D’Agostino says Chomsky subscribes to Linguistic Representationalism because “language users ‘mentally represent’ the grammar of their language.” This “ism” is not explored in great detail, and as such I will not say any more on the matter.
4. Linguistic Mentalism
Linguistic Mentalism claims “that language users ‘know’ the grammars of their languages” and “individuals’ knowledge of a grammar provides the (psychological) basis for their linguistic competence. Linguistic mentalism is, then, just that version of linguistic subjectivism according to which a grammar provides the psychological basis for linguistic competence because it is known by the language-user whose competence it describes.” However, it is quite possible that they have “never been aware of that grammar and may be unable to formulate or even understand formulations of the knowledge they are alleged to have.” As they are not necessarily aware of grammar, “the knowledge of it attributed to them… must be tacit knowledge rather than any ordinary kind of propositional knowledge.”
Leiber defined mentalism as “the view that psychology, particularly when describing human linguistic activity and thought, must talk in terms of structures and processes that can be tied to observed physical behavior only in a very indirect way.” [Leiber: 15]
Linguistic mentalism is “merely a species of linguistic intellectualism.” D’Agostino rejects this view he ascribes to Chomsky and accepts what he calls a realist approach: “language-users’ linguistic competence is described by a grammar of their language and is based on their incorporation of a mental mechanism the operations of which are described by that grammar.”
The term “mentalism”, in general, is a terrible word to use, as it does not express a clear meaning and it may have different meanings to different people. Alex Orenstein, for example, says mentalism is “the view that meanings are ideas”, which is seemingly completely unrelated to any sense it is used here. [Orenstein: 122] Others use it synonymously with idealism, that mentalism is “a doctrine that mind is the true reality and that objects exist only as aspects of the mind’s awareness”. Chomsky is by no means denying the physical world.
Chomsky’s Response to “Mentalism”
As the mentalism-realism dispute was the primary point of contention between D’Agostino and Chomsky, I asked Chomsky (May 11, 2009) to comment on why he felt D’Agostino was wrong about accepting “linguistic realism”. He responded:
He is wrong for reasons pointed out by Wittgenstein. He uses the word “know” not as it is used in ordinary English (as in the statements of mine to which he refers), but in a technical sense invented by philosophers. It’s an accident of English that we speak of “knowing a language” (instead of having a language, or speaking a language, or other locutions). I’ve discussed this decades ago, but it seems to have no effect. Philosophers are wedded to that error, and insist on investing the English locution “know a language” with the associations of their invented technical usage. Exactly the mistake that Wittgenstein tried to induce philosophers to avoid, as did “ordinary language philosophers” — all of them studied, but forgotten, it seems.
Another error is to fail to recognize the systematic ambiguity of the term “grammar,” another matter I discussed probably 40 years ago or more. The term is used either to refer to linguistic competence, or to a linguist’s theory about it. I therefore suggested about 25 years ago that to avoid confusions like D’Agostino’s, we keep the term “grammar” to its use as “linguist’s theory” — similar to the traditional usage — and refer to the system attained (the state of the language faculty, competence) as “internal language” (“I-language”).
I certainly don’t accept the “bridge thesis” (if it is even coherent). As for “mentalism,” I know of no such notion, at least since Newton’s demolition of the only coherent notion of “physical”.
Further expanding on his understanding of the term “mentalism” (which was applied to him by D’Agostino, not his self-described position), Chomsky writes:
In brief, I don’t know of any sense of “mentalism” except that there are what we loosely call mental aspects of the world just as there are chemical, optical, etc. aspects. I suppose one could invent a sense in terms of something like Cartesian dualism, but I don’t see any way to make sense of that either — not for the reasons usually given, but because there is no coherent notion of “physical,” post-Newton.
Perhaps the problem arose as far back as 1965 when Chomsky wrote, “The problem for the linguist… is to determine… the underlying system of rules that has been mastered by the speaker-hearer… Hence, in a technical sense, linguistic theory is mentalistic, since it is concerned with discovering a mental reality underlying actual behavior.” [Chomsky 1965: 4] (emphasis mine) Clearly this was not as simple a concept as Chomsky intended.
What is a Grammar?
A grammar is a very basic concept in linguistics, particularly when reading Chomsky. But, perhaps you’re not familiar with the technical definition.
A “grammar is a mathematical function which induces a mapping between linguistic stimuli (acoustic sensations) and their corresponding percepts (structural descriptions of sentences).”
5. Linguistic Rationalism
Leiber says that “most philosophers, psychologists, and scientists generally have taken it for granted that rationalism is as outmoded as alchemy, the divine right of kings, or the view that the world is supported by a giant standing on a tortoise.” [Leiber: 14] Why is Chomsky not following the modern trend?
Linguistic Rationalism claims “that human language learning is mediated by innate mental schemata which define the range of possible human linguistic experience.” Or, alternately, that “research in linguistics supports a rationalist account of learning”. For Chomsky, we have to accept at least some parts of rationalism if we accept universal grammar, innate ideas or that psychology relies in part on mentalistic notions.
Chomsky contends that innate ideas are not present in conscious thought, and some will never be present in conscious thought. What is confusing, and why Chomsky may not be considered a rationalist by some, is that what philosophers post-Descartes called “rational” (a priori knowledge or truth) is not what Chomsky means by the term.
Limitationism says “there are biologically determined limits on the kinds of systems of beliefs or aesthetically significant objects which human beings can construct or find intelligible.” Or, another way to say it, “biologically fixed innate cognitive faculties ‘set limits on human intellectual development’ — limits within which human creativity is manifested.”
Chomsky says “that true creativity ‘takes place within — presupposes, in fact — a system of constraints and governing principles’.” D’Agostino calls this the “limits thesis”. The limits thesis says “there are systems of beliefs and objects of aesthetic value which human beings must fail to find intelligible or which they must be unable to construct. But the limits thesis does not specify which systems of beliefs or objects of value are inaccessible to human intelligence in this way.” The limits are “biologically determined”.
Some have objected to limitationism, saying that it makes human behavior predictable and determined. D’Agostino refutes this by comparing human activity to a game of chess. There are a certain set of rules that must be followed during chess — we cannot suddenly decide to move a rook diagonally, for instance — but this does not in any way mean that the moves or outcome are determined or predictable, despite having limits.
Chomsky accepts the limits thesis because without it “we have arbitrary and random behaviour, not creative acts”. For him, it is the limitation that creates a structure within which to express ourselves, not an open-ended possibility.
D’Agostino tries to draw a connection between Chomsky’s linguistic beliefs and his political belief in libertarianism (or anarchism, if you prefer). As Chomsky himself has pointed out on many occasions, there is no direct connection and I feel that any points D’Agostino makes are a stretch.
For Further Reading
Obviously, the best external source of material on Chomsky’s linguist ideologies is Fred D’Agostino’s Chomsky’s System of Ideas. However, that being said, I have also found John Lyons’ Noam Chomsky to be quite helpful. Lyons has the benefit of writing on Chomsky before his political work overshadowed his linguistic work, which provides for a more pure and unbiased examination.
As for Chomsky’s own writings, Syntactic Structures is his groundbreaking work and deserves attention. Other notable volumes are Cartesian Linguistics and Language and Mind. His books and papers are numerous, but these provide a sensible overview.
Chomsky, Noam. Aspects of the Theory of Syntax MIT Press, 1965.
D’Agostino, Fred. Chomsky’s System of Ideas
Devitt, Michael and Kim Sterelny. Language and Reality: An Introduction to the Philosophy of Language MIT Press, 1997.
Leiber, Justin. Noam Chomsky: A Philosophic Overview St. Martin’s Press, 1975.
Lyons, John. Chomsky (Modern Masters) Fontana Press, 1997.
Orenstein, Alex. W. V. Quine (Philosophy Now) Princeton University Press, 2002.