It is sometimes thought that the aim of achieving explanatory adequacy--as against mere observational and descriptive adequacy--only makes sense on the view that linguistics is branch of psychology. I argue that it’s possible to motivate this reasonable aim solely by an appeal to the theoretical virtues of generality and explanatory unification. Two cases from the history of generative linguistics are discussed and cited as examples of this strategy: (i) Chomsky's path-breaking shift, in Syntactic Structures, from pure phrase-structure grammars to a transformational formalism. (ii) The turn from the transformational grammar of the Extended Standard Theory (EST) in the 1970s to the Principles and Parameters (P&P) approach in the 1980s and onward. I argue that Chomsky's well known cognitivist position with regard to the metaphysics of language derives no leverage over its platonist and nominalist rivals on this issue, for the platonist and nominalist can both supply compelling grounds for pursuing explanatory adequacy. Conceived as the maximally general, simple, and unified theoretical coverage of all human languages, explanatory adequacy is desirable whether or not cognitivism is true.
I begin by outlining the goals of Linguistics, taking into account all of the various subfields. I then summarize Jerry Katz’s history of the field, with a focus on his preoccupation with matters of ontology. This provides a springboard into a broad critique of Katz’s Platonist metaphysical and epistemological views. I argue against his contentions that Linguistics studies a set of abstract entities and that it does so in a non-empirical fashion. I rebut Katz’s direct argument against the opposing view that all rational inquiry is empirical, including logic and mathematics. Next, I turn to Katz’s considerations concerning the infinitude of language. Following Michael Devitt, I propose that we treat claims concerning the infinitude of language as claims about the modal import of the lawlike, counterfactual-supporting claims of linguistic theory.
Can the qualitative character of a sensory state be nonconscious? Is qualitative character an intrinsic property of a sensory state, directly and exhaustively revealed in conscious experience? The answers to these questions put substantive constraints on any account of the metaphysical nature and epistemic status of conscious experience. To address these questions, I take as case studies two strikingly similar theories of consciousness, articulated by David Rosenthal and Daniel Dennett, respectively. Though they share foundational commitments, key differences set them apart. Specifically, they diverge in their treatments of the consciousness of qualitative mental states. Rosenthal’s Higher-Order Thought model supports a distinction between conscious and nonconscious qualitative states, and a complementary distinction between the appearance and the reality of such states. Dennett’s Multiple-Drafts Model collapses both of these distinctions. I argue that sustaining these distinctions is crucial for formulating an adequate account of consciousness and sensory qualities. Because Dennett’s position collapses them, Rosenthal’s HOT model is in a better position to resolve longstanding questions in this area. I argue, furthermore, that Dennett’s own methodology presupposes the coherence of the very distinctions that he seeks to dissolve. His position is thus inherently unstable. I locate the instability in two features of his methodology. First, Dennett imposes unwarranted restrictions on the sorts of evidence that may legitimately be taken to bear on the truth of hypotheses about the nature of conscious experience. Second, he is committed to the thesis that a subject’s first-person reports regarding his or her mental states constitutively determine the character of those states. This “first-person operationalism,” I argue, is both implausible and theoretically unmotivated.
I am concerned here to elucidate the nature of our conscious sense of agency. I develop my view by contrasting it with three accounts in the current literature. In his book, The Illusion of Conscious Will (MIT, 2002), psychologist Daniel Wegner argues that our sense of agency is systematically in error, as it mischaracterizes the causes of our voluntary actions. In the first half of this paper, I distinguish between a benign interpretation of this thesis and a radical interpretation. I argue that the benign reading closely resembles a view espoused by David Rosenthal. The radical version, by contrast, rests on an implausible eliminativist view, which I argue should be rejected. In the second half of the paper, I turn to the implications of the view shared by Rosenthal and, on a charitable reading, Wegner. I begin by discussing Harry Frankfurt’s famous compatibilist account of free will, with a particular focus on his notions of an agent’s identifying with his or her effective desires, or feeling alienated from them. After examining J. David Velleman’s critiques of Frankfurt’s analysis, as well as his own alternative account, I argue that Velleman’s motivation for his account is faulty and that his positive account fails to meet the very criteria that it was designed to satisfy. I then offer my own analysis of identification and alienation, which draws on the higher-order thought theory of consciousness.
A Look Ahead...
In future work, I aim to pursue a longstanding interest in the debate between holism and atomism with regard to intentional content. Atomistic theories, especially the purely referential atomism defended by Jerry Fodor, are often seen as promising alternatives to the unbridled holism of W. V. Quine, Donald Davidson, and, more recently, Daniel Dennett and Robert Brandom. I find the standard arguments against content holism unconvincing and I believe that atomistic theories face a number of insuperable difficulties with regard to their treatment of the mental attitudes and their accounts of concept acquisition. Furthermore, much of the appeal of atomistic theories derives from the conviction that Saul Kripke’s famous arguments in Naming and Necessity establish an atomistic direct reference theory, or at least point in the direction of a causal or informational metasemantics. I am skeptical that Kripke’s arguments show anything of the sort. In future work, I hope to show how content holism—particularly inferential-role semantics—can avoid the objections that have been leveled against it, and to defuse the Kripkean arguments that animate the atomist position.