Linguists tell us that the sentence ‘I enjoyed yourself’ is ungrammatical because it violates structural constraints on English sentences. Is this a fact about the psychology of English speakers, or a fact about some mind-independent state of affairs? If it is indeed a fact about the speaker’s psychological makeup then is it so in virtue of the speaker’s linguistic beliefs or knowledge, or can we leave intentional states like these out of the story?

My answers to these questions take the form of these two theses: (T1) The Psychological View is preferable to the Autonomous View as a conception of what linguistics—generative grammar, in particular—is about. The Psychological View holds that linguistic theories are best seen as psychological (ultimately biological) in nature, that is, as concerned with certain aspects of the mind/brain. In contrast, “autonomist” views claim that linguistics deals with an extra-psychological, and extra-biological, “linguistic reality”. (T2) Within the Psychological View, a Non-Intentional view of linguistics is preferable to an Intentional View (IV). Though linguistics, as a psychological discipline, studies a mental structure, this structure is to be understood in a non-intentional way, in particular, without making reference to propositional attitudes. Thus, the view I call “Propositional Attitude Intentionalism” (PAI) must be rejected.

To establish T1 I argue that the Psychological View can foster, and make sense of, episodes and even programs of interdisciplinary exchange, which are themselves important sources of scientific progress, due to the cross-fertilization between the disciplines involved. In sum, the Psychological View, in virtue of the integrationist stance it involves, makes linguistics more fit for some important sources of progress than the Autonomous View does, and this gives us a prima facie reason for preferring the Psychological View.

Jerrold Katz, Scott Soames, and Michael Devitt have all argued for the Autonomous View. I reply by showing that their arguments rest on unjustified, and ultimately mistaken assumptions, in particular, on an overly restrictive conception of psychology and on the assimilation of linguistics to the “formal science” model.

As for T2, I reject some of the main arguments offered by some PV authors in favor of PAI (“Propositional Attitude Intentionalism”), focusing on what I call the “Explanatory Indispensability Argument”. This argument relies on the premise that we must attribute propositional attitudes (knowledge, belief) about grammatical principles to competent speakers in order to explain (the content of) linguistic intuitions. I question the intellectualistic view of linguistic intuition that the Explanatory Indispensability Argument presupposes, and then suggest an alternative “perceptual/central” view. I also cast doubt on the plausibility of PAI; the states that carry “grammatical information” seem to be subpersonal, subdoxastic and non-conceptual, contrary to what PAI itself demands.

As an alternative to PAI, I present a broad outline of a non-intentional view, which construes linguistics as the study of the language faculty, which is an “organ of computation”. Linguistics then, is a higher-level study of the structural and formal (as opposed to dynamic and functional) aspects of the language organ.