Meaning, Mind, and Action: Philosophical Essays

By Julia Tanney

Julia Tanney’s Meaning, Mind, and Action challenges widely held presuppositions within philosophy in its classical ‘analytic’, ‘naturalist’, and ‘cognitivist’ forms. Beginning with canonical views in the philosophy of language and logic, the arguments are then applied to discussions of knowledge, action, causation, the nature of the mental, consciousness, and thinking.
Responding to a tradition that harks back to Plato and was resurrected by Mill, Frege, Russell, Moore, and the early Wittgenstein, Meaning, Mind, and Action challenges today’s orthodoxy on its own terms, beginning with canonical views in the philosophy of language and philosophical logic. The arguments of these early chapters are then applied to the theory of knowledge, action, and causation, followed by those on the nature of the mental, consciousness, and thinking. The final section, on the logic of the mental, widens the arguments to include the subject of animal minds, the postulation of mental representations in cultural anthropology, the author’s intention in literary theory, and the philosophical problem of irrationality in psychiatry.

• Language: English
• Publisher: Anthem Press
• Release date: Jul 12, 2022
• ISBN: 9781839984846

Book preview

INTRODUCTION

Today’s study of philosophy in the analytic tradition—of language, mind, knowledge and value—is beholden to a history and tradition that is characterized by its respect for argument and, at its best, a striving for clarity. Although its concerns and methodology go back to antiquity, today’s practice in various subdisciplines has—whether we realize it or not—been sculpted by the advances in classical logic of the nineteenth and twentieth centuries by Mill, Frege and Russell and its embrace by subsequent philosophers of language hoping to account for the semantics of natural language expressions.

In his summary of fin de siècle philosophy of language and mind, Tyler Burge outlines two traditions that evolved from postpositivist philosophy.¹ The first, deriving from Frege and taking science, logic, or mathematics as the source of inspiration for linguistic and philosophical investigation, distrusted intuition and championed theory.² Frege’s contribution to mathematical logic, Burge recounts, reached its apex in the mid- to late twentieth century with attempts to provide an account of the truth conditions, logical form, and compositional structure of natural language. Metaphysics in the analytic tradition, rejected by positivists as meaningless, was resurrected by Quine as ontological questions were recast in terms of the theoretical advantages of admitting candidates as the objects of bound variables in a formal language. Eventually, topics in domains Quine thought to be recalcitrant from a scientific point of view—for example, mind and meaning—would be embraced by his adherents who sought to broaden his naturalist program. In the last third of the twentieth century, philosophy of mind replaced philosophy of language as the predominant field and conceptual analysis was in many quarters nudged aside as philosophers refocused their inquiries on the nature of the phenomenon itself, as opposed to the natural language expressions that served, so they alleged, imperfectly to identify it. Naturalism became the cri de cœur of philosophers in several subdisciplines. Conundrums brought to center stage in the seventeenth century about the nature of the mental and of knowledge were rewritten in a new key to suit the emerging computer paradigm and its prominent role in cognitive psychology.

Burge suggests that the second, ordinary language, tradition began with G. E. Moore and continued with important contributions by Austin and Strawson. Taking ordinary practice as the touchstone for linguistic and philosophical judgment, it distrusted principles and championed examples.³ Although providing some brilliant linguistic observation and new tools for dealing with philosophical perplexities, it left a rather thin legacy of specific contributions, and as a philosophical method, it faced numerous difficulties, never adequately dealt with, in deriving philosophical conclusions from linguistic examples.⁴ By contrast, philosophical approaches to the studies of logical form by the utilization of classical logic have been substantial and lasting.⁵ Frege’s work in mathematical logic, Burge maintains, has given subsequent philosophers a new framework for discussing old problems.

The difficulty with the tradition inherited from Frege is that the old problems have not disappeared; the embedded layers of technicality serve merely to hide the fact that the traditional puzzles have become even more intransigent. Although it is true that Moore challenged philosophical theories that did not at the outset respect commonsense truisms, his method of conceptual analysis—taught to all beginning students of analytic philosophy and still attempted by many—was Platonic at its core. As we shall see, the method was partly inspired by, and had much in common with, the formal approach. Indeed, philosophers today have inherited a largely referential framework that was adapted in importantly different ways by Frege, Russell, and the early Wittgenstein and continued in the work of Quine, Dummett, Davidson, Fodor, and their various admirers.

My first volume of essays presented an incipient challenge to this backdrop.⁶ It began by tackling its more recent incarnation in metaphysical and empirical approaches to the study of the mind and, more specifically, to the prevalent view that conceives the workings of the rational mind as the outcome of causal interactions between mental states that have their bases in the brain. Rather than construing the workings of the rational mind in quasi-theoretical terms as causal interactions between functional states of an information processor or physical states of the brain, rational explanation, I argued, is better construed as embedded in thick descriptions of actions, or sophistications upon ever-ascending levels of contextually oriented discourse.

I also proposed that the move from conceptual analysis to philosophical theory has failed to confront the problem that plagued analysis. I argued that there are no practice- and circumstance-independent conditions that govern the application of most, if not all, nontheoretical predicates of natural language—mental or otherwise. But then nor can there be identifications of real states, events, relations, or properties obliquely indicated by these predicates, since no such identification can transcend our considered judgments about the appropriateness of the application of the predicate in the first place. And these conditions change with the circumstances. Inspired by Wittgenstein and Ryle—who are omitted from Burge’s summary—I defended the practice of philosophy as a type of conceptual cartography that invites us to think of concepts not as abstract objects but as amorphous abstractions: collections of discernible features of natural language expressions with overlapping and crisscrossing patterns of similarity.

In this second volume, in which all but two of the essays were written in the last decade (and slightly modified here), these themes are brought home and developed further as they are applied to subject matters that include meaning, philosophical logic, knowledge, action, the nature of the mental, consciousness, animal minds, cognitive anthropology, the interpretation of literature and everyday action, and mental trauma. These chapters are self-contained and treat various themes. In this introduction, I shall emphasize the ways in which they challenge the precepts—often implicit—guiding Western philosophy’s methods today and their adoption by early analytic philosophers.

Meaning and Philosophical Logic

Let us begin to uncover these methodological presuppositions by rehearsing the platitudes that are frequently cited as setting constraints upon philosophical theorizing across domains. First, there is the idea that an inquiry into knowledge or consciousness, for example, depends upon a view about the supposed referent of these abstract nouns. This may be construed as an abstract object—a concept with fixed rules of application across different contexts (as in the first example), or as a natural phenomenon in need of a theory (as in the second). Both involve (near enough) the assumption going back to antiquity that the meaning of the abstract noun or the nature of the phenomenon to which it gestures can be captured by discerning that which is common among its instances. To grasp a concept or to attach to a word its meaning is to be equipped with rules for its application that fix the relation between a lexical item—such as a general term—and its reference or extension. Moore thought these rules could be revealed by breaking down complex concepts into simpler ones by definitional equivalences, which would lay down context-free, invariant conditions—individually necessary and jointly sufficient—for the application of the word that is thus defined. The goal of analysis, for Moore, was to identify the simplest qualities, known directly, which (he believed) were named by the constituent concepts of any given proposition. The approach inherited from Frege, Russell and later Quine and Davidson, required paraphrasing natural language expressions into ones with a more perspicuous, topic-neutral, logical form (so that, e.g., we might eliminate or vindicate the types of entities to which the theory is committed). In either case, it is these abstract objects, extensions, or natural properties that are the appropriate target of intellectual curiosity and not the natural language expressions we employ in everyday discourse. These expressions serve merely to indicate that which it is up to our best philosophical-analytic or scientific-theoretical practices to examine.

The widely accepted view today is that the meaning of a complex, natural language expression such as a sentence is determined by its structure and the meaning of its constituents. That is to say, the same mode of combination and the same assignment of meanings to the embedded expressions cannot determine more than one sentence meaning within a language. Such content—which, it is maintained, must remain stable across contexts and thus be impervious to local pragmatic concerns—is needed to accommodate logical transformations (embeddings in, say, negational, conditional, or propositional attitude contexts) and inferential reasoning. It is this semantic or propositional content—however it is to be unpacked (perhaps as truth conditions or sentences in a language of thought)—which is the bearer of truth and falsehood.⁷

Challenges to these suppositions begin with an examination of the meaning of meaning. Wittgenstein suggested, enigmatically, that [t]he meaning of a word is what is explained by the explanation of the meaning.⁸ Look for what we call explanations of meaning, he continues, if we want to understand how the word meaning is actually used. With these two suggestions, the philosophical question, What is the meaning of a word? along with its traditional response—the item for which it stands or that which it signifies or represents or even its (determinate) extension—is replaced by How do we explain the contribution a given word makes to what is said in the context of a particular linguistic exchange?

Well, how do we? Our assertions are claims on others’ beliefs. To assess these claims, our audience must understand what we are trying to say. Sometimes what we mean, if we are to be understood, is settled by factors that override any scope we might have for drawing boundaries on our part. Nonetheless, we also enjoy a great deal of latitude in instructing our interlocutors how our words are to be taken. Chapter 1, Explaining What We Mean, undertakes to answer Wittgenstein’s transformed question by examining a puzzling, because at first sight inconsistent, claim. Once the speaker explains how she wants her statement to be taken—the particular logical inflections she wishes to rule in and out in speaking as she does—what she says makes sense in the newly revealed context. As the example continues, it becomes apparent that, as Ryle insisted, the logical significance of most, if not all, expressions of natural languages is indefinitely elastic, including, of course, those expressions used in explanations of what one means. Or, to put it another way, rather than being united by sharing the same thing in common, such expressions belong to a family of structures, more or less related to one another.⁹ Not only is the contribution of a word or phrase to what is said apt to change in multifarious ways, sometimes even within a given conversation, it is likely to continue to change in unforeseen ways as different uses become established. In unraveling the logical threads of what a speaker says—the point in saying it, what would count as evidence, justification, warrant (or as implied, consistent, inconsistent) as an appropriate or inappropriate uptake or reaction—she can spell out what she means. Doing so will reveal how the constituent expressions make their contribution.

In one stroke, several of the presuppositions fall.

First, that the meaning of an abstract noun or predicate expression, for example, or the (alleged) nature of the phenomenon to which it gestures can be captured by discerning that which is common among its instances, for there appears, at least for most, if not all, natural language expressions, to be nothing in common between their instances. If true, this would put paid to the suggestion that such natural language expressions can be assigned semantic interpretations in a formal language in such a way as to respect their saying power.

Second, that the meaning of complex expressions (e.g., a proposition or truth condition) is fully determined by the meanings of their constituents (the abstract or natural objects given by their interpretation) and the syntactic rules of combination, irrespective of other factors. On the contrary, it is in virtue of having contextual features in view that we can determine what a speaker is trying to say and thus the contribution her words make or what she means by them.

Third, and relatedly, the suggestion that a competent speaker can (within limits, of course) direct how her words are to be understood by clarifying their logical ties, and therefore what she means in saying what she does, puts under pressure the oft-cited platitude that content is needed which—impervious to local pragmatic concerns—remains stable and available for reasoning. This gets things back to front, for in order to determine the correct inflection of any given use of an expression, its inferential relations in the context of an utterance must already be discerned. To explain what one means is to make the logical implications of what one says evident. Unraveling the logical threads of an utterance, then, may be necessary for, and thus prior to, an assessment of what one has said, in terms, for example, of its truth, falsity, probability, or cogency. A rejection of semantic content thus construed would present no mystery about language use, however, for the ability to understand what is said is explained not only by shared practices and common interests but also by the capacity of interlocutors to ask questions and explain what they mean.

Instead, an exploration of any given concept would involve comparing and contrasting the way the relevant expressions contribute, on a case-by-case basis, to what is said in various situations. Siding with John Wisdom, I suggest that in examining the grounds for our inclinations to apply or refuse a particular expression, we thereby reveal affinities and differences between the new case and others. In so doing, we are tracing connections that language obscures.¹⁰ Or, as Austin suggests, to reflect on what we would say when is to make genuine discoveries about the corresponding realities.¹¹

The following two chapters continue to explore questions in philosophical logic. While the first chapter focuses on the complex and shifting commitments in the use of subsentential expressions such as general nouns and live verbs, the second chapter turns the spotlight on sentences themselves. Rule Following, Intellectualism, and Logical Reasoning: On the Importance of a Type Distinction between Performances and ‘Propositional Knowledge’ of the Norms that Govern Them begins with quotes from Wittgenstein and Ryle. The former reminds us of the countless kinds of uses of what we call sentences as well as symbols and words; the latter reminds us of the various employments of sentences besides those of describing or reporting facts. I concentrate here on sentence employments that, in the contexts I examine, are explanatory, justificatory, or action guiding. Among these, we find the expression of laws, rules, dispositions, definitions, principles, and norms. Because such sentences are classified by logicians as statements, and thus as truth apt, philosophers wearing the spectacles of logical theory tend to look in the wrong place for the facts they are alleged to state; namely, to those which are assumed to render the sentences true. Because awareness of these facts is also supposed to be motivating, they hypostasize entities that at the same time are thought to be imbued with the ability to guide our judgments and actions. A metaphysical tangle ensues as the postulation of impossible objects leads to its own philosophical problems. A way out of this maze becomes apparent when we notice that there is a logico-categorical difference between a sentence employed to provide, for example, a warrant and its employment in a description. The truth value of the former comes not from tracking causal links, logical rails, or action-guiding normative facts but rather from the warrant’s pragmatic role in proscribing or supporting time- and context-bound inferences. The chapter discusses this distinction in the light of Ryle’s insistence on the dispositional use of mental-conduct concepts, of Wittgenstein’s discussion of rule following, and of Lewis Carroll’s famous paradox about logical reasoning.

Explaining What We Mean will have already made us suspicious of the idea that in speaking and understanding a language, we employ a decision-making procedure or a calculus that we operate in accordance with fixed rules. For in this first chapter, we will have seen that to the extent that there are rules of meaning and grammar, they are violable if there is a point in overriding them. But there are other reasons as well. Chapter 3, Wittgenstein on Rule Following and Interpretation, takes a close look at what are arguably the most discussed and disputed passages of twentieth-century Anglo-Saxon philosophy: Sections 201 and 202 of the Philosophical Investigations. This chapter elucidates—in my own way—the steps in the famous, elliptical argument that begins Here was our paradox and concludes Hence, ‘obeying a rule’ is a practice and hence, it is not possible to obey a rule privately.

The gist of these chapters is that there may be nothing in the saying power of natural expressions themselves, at any level of elucidation, that uniquely determines how they are to be applied. How, then, can expressions of rules be otherwise? The mythical picture behind Platonism, however, suggests that we somehow grasp moral injunctions, mathematical formulae, and semantic and syntactic rules that lay out the moves to be made in advance and independently of the relevant normative practices in which we participate. Their independence is supposed to be necessary to explain the objectivity of the rules, and our grasp of them is supposed to explain how we act in accord with them. If this picture is to be rejected, however, then we are nonetheless required to make sense of normative practices such as acting rationally or morally and speaking and understanding a language that has objectively right and wrong ways to proceed. I argue in this chapter that adopting this Platonic framework requires asking too much of the tools we use to guide us in learning, teaching, or adjudicating moves in practices in which the notions of error and success have their place. Such normative practices or linguistic activities in which natural reactions reveal what counts as acceptable or not must be up and running before the formulation of rules and what counts as their application can be discerned.

To appreciate the indefinite elasticity of most, if not all, nonartificial, natural language expressions is to accept that there may be nothing in common by which we call a thing by the same name. Philosophically interesting abstract nouns, for example, do not typically refer to abstract objects, determinate extensions, or concepts whose essence can be revealed by contextually transcendent application conditions for their cognate expressions. The difficulty in Socratic- or Moorean-style analysis should have taught us that. A fortiori, nor do they function to signify, or even gesture in the direction of, facts, properties, or relations whose nature is to be revealed by metaphysical or philosophical-scientific speculation—that is, if the manifold saying power of the expressions in which they are embedded is to be respected. In examining this power by looking at the contribution the relevant expressions make to diverse types of linguistic exchanges, we will find that their role is often other than that of description or representation.

Knowledge, Action, and Causation

As the following chapters continue to press these points, we begin by focusing on the pursuit of a theory of knowledge. It should be evident by now that we must be chary of the distinction between language, on the one hand, and the world, on the other, if this is meant to suggest we can talk meaningfully about language’s adequacy to the world.¹² Even those who embrace this distinction, however, and purport to be interested in the nature of knowledge as opposed to our use of language cannot sustain this for long, since they are bound to admit that philosophical or theoretical investigations of knowledge, like those of other abstractions, cannot be conducted without attention to what we say and understand when using the relevant expressions. Chapter 4, What Knowledge Is Not: Reflections on Some Uses of the Verb ‘To Know,’ reinforces this point and demonstrates that an examination of first- and third-person singular uses of to know and their comparison with to believe and to be certain reveals complex and shifting commitments. As we should now be able to anticipate, these logical threads cannot be discerned by studying these expressions, or even the sentence-types in which they are embedded, in isolation. The fact that situation-bound utterances comprising I know or S/he knows may function, logically, in any number of ways presents difficulties for any theory that fails to respect these differences. Indeed, it casts into doubt the thought that the standards that govern the use of know do so, as truth conditions are traditionally conceived to do, outside time. Here, I show how it undermines the doctrine that knowledge is a type of mental state; in particular, a belief that is true and justified.

Our investigation into the nature of knowledge—as it is unveiled, for example, by the shifting commitments in our employment of to know and its contrast with related expressions—reveals anomalies that infect traditional arguments for skepticism. Are claims to knowledge, like mathematical statements and their proofs, supposed to be independent of time and context? Should the kinds of inferences, explanations, and justifications that we make in everyday reasoning aspire to, or be made to conform to, this standard? It would seem so, if philosophy’s fascination with secure, analytic, or deductive entailments is anything to go by. Philosophical skepticism is predicated on this supposition and, I argue, should be disregarded because of it. This chapter—influenced not only by Wittgenstein and Ryle but by the work of Austin, Toulmin, Urmson, and Wisdom—ends by suggesting that our claims to knowledge cannot meet the standards imposed by the skeptic without undermining the multifaceted expressive power of to know and its kin. I go on to sketch how the debates between realism and reductionism in various domains presuppose acceptance of this impossible standard.

It is suggested in Chapter 2 that the tendency to interpret declarative sentences as statements of fact hinders an appreciation of the saying power or the logic of the utterances in which they figure. Indeed, in our comparison of the contributions made by to know, to believe, and to be certain, we encounter other, nondescriptive functions such as hedging, endorsing, and other forms of signaling which are left unaccommodated by the attempt to offer an analytic definition or a general account or theory of knowledge.

Other nondescriptive roles of sentences often classified by logicians as statements include the expression of laws, rules, and dispositions. When, however, such statements are construed as having a descriptive or reporting role, a metaphysical-epistemological mystery arises. Searching for the facts and their constituents that are supposed to render these statements true while at the same time recognizing their normative force, we postulate abstract objects, necessary relations, causal links, or rules as rails and then puzzle over their power to move us. Chapter 5, Remarks on the Thickness of Action Description begins with a consideration of the difficulties for the supposition that intentions, acts of will, and reasons for acting, construed as mental events, are the causal factors that would render mere bodily movement into voluntary or intentional actions. Given the problems with this view—regarding, for example, the nature of mental states and our knowledge of them—what motivates us to accept it?

It will be useful to recall the recent moments that have informed today’s practice of analytic philosophy. Those following Quine have by and large accepted that theoretical purposes require paraphrasing natural language expressions into others with a more perspicuous logical form. Like Davidson, they have rejected Quine’s project of reforming language. In discarding his strictures on what is to count as appropriate topics for theorizing, they have enlarged his naturalist program to include the study of meaning, mind, and action. The naturalist philosopher’s preference for a topic-neutral ontology of events, states, properties, and causes seems to be inherited from the desire to understand these notions as figuring in statements that describe phenomena in the world that occur independent of our conceptual or descriptive resources. The problem is that in such a scheme, crucial explanatory differences are elided as we very earnestly employ for concepts of two (or more) very different levels the same umbrella-words.¹³ The chapter ends by suggesting how if reasons, beliefs, desires, or intentions are not prior or concomitant mental causes, they nonetheless perform their pivotal explanatory function.

The examination of the logic of sentences used to report actions and the connections between the concepts of agent, action, event, and cause—especially prevalent in the 1960s—made use of the topic-neutral ontology of events, properties, states and causes and the language-world divide that underpins it. The wide acceptance of Davidson’s argument that reason explanation is a species of causal explanation and, correlatively, that reasons are causes, provided a paradigm shift away from the positivist and antipositivist traditions, as it combined views from both.¹⁴ Welcomed by philosophers, psychologists, and cognitive scientists, Davidson’s work in turn led to a resurgence of interest in metaphysical questions about the mental but also served to rewrite the traditional puzzles in the philosophy of mind. One of these problems is that of the explanatory role of the mental. I have argued elsewhere and at length that the thesis that reasons are causes is untenable.¹⁵ In Chapter 6, Prolegomena to a Cartographical Investigation of Cause and Reason, largely inspired by the work of R. F. Holland and John Wisdom, I take a somewhat different tack by engaging in my own conceptual-cartographical investigation and comparing and contrasting the use of terms associated with the concepts of cause and reason. There is no doubt that the conceptual domains overlap; nonetheless, there are substantial differences. Our causal knowledge derives from the contextually embedded use of verbs—melt, stain, catch, explode, break, etc.—which are implicated in answer to questions such as What made this happen? or Why did this occur? The concept of reason intermingles with those of forward- and backward-looking motives, character traits, intentions, plans, and so forth. Our search for reasons tends to exploit thicker notions of, say, revenge, lust, and power, and, as Anscombe famously pointed out, involves considering plausible answers to questions such as Why did you act thus? The ideas of agency and self-determination presupposed in this question overlap with the notions of freedom, responsibility, blameworthiness, and praiseworthiness. This sophisticated network of logical-implication threads is part and parcel of our concepts of cause and reason. They are, in other words, what give the sentence utterances that embody these concepts their explanatory power. Once these logical threads are eviscerated from such sentence utterances—as they are, for example, when we are asked to consider causation as a relation between events in the world independently of our conceptual-descriptive apparatus for individuating them, or, with Davidson, when we are asked to consider reasons as causally efficacious events under physical descriptions but explanatory under mental descriptions—we must face the inevitable paradoxes about the nature and the explanatory power of the mental.

Mind, Consciousness, and Thinking

We continue to explore central topics in the philosophy of mind with Chapter 7, The Location of the Mind, proffering an up-to-date survey of the attempt to find a place for the mental in a physical world. After reviewing several endeavors to unpack the notion of in the head, I turn to recent proposals to locate mental properties either in the head, not in the head but somewhere else, or some combination of the two. I hope to show that these suggested theories are saddled with reiterations of the fundamental puzzles they were introduced to solve. The epistemological problem of other minds, the metaphysical problem of mental-physical identity, as well as that of mental explanation that they bring in train, not only reemerge, but, I allege, prompt new absurdities. Supersizing the mind and extending the bounds of cognition or phenomenological space to include the environment are two examples. The idea of introducing consciousness as a fundamental element of the universe is a third.

The next two chapters treat what has become known as the problem of consciousness and its reliance on the notion that there is something that it is like to be a conscious human being that is left out of theories that attempt to reduce mental to physical or functional events or processes. It is understandable, of course, to suspect that something significant has been omitted by a physicalism or, indeed, a functionalism that tries to understand mentality as nothing over and above neurological, or, more abstractly, functional or computational processes of a living human being. Less clear is how invoking the notions of what it is like or conscious experience might clarify what is missing, especially when it is claimed that both these expressions point to something ineffable that cannot be articulated even by first-person narrative. Claims such as Red things appear to her as green or She seems to see red, by contrast, can be and are defended in various ways. Such a defense or backing, as we saw in the summary of Chapter 1, will help inform what is meant by the constituent expressions—here, what is to count in the circumstances as appearing green or red things. What tempts us, then, to invoke this mysterious, ineffable mental space? In Chapter 8, Some Absurdities in the Notion of ‘Conscious Experience’, I detect a sleight of hand when philosophers argue from such commonplaces as that we see the sky, hear the piano, and feel pain to the existence of inner experiences known only to the subject, for whether or not someone sees the sky, hears the piano, and feels pain is testable. Moreover, the suggestion that these newly hypostasized experiences are ineffable is self-defeating. The problem is not an epistemological one about imaginability: it is the logically prior one of the neo-Dualist’s attempt to convince us of the possibility of situations—involving inverted color spectra or philosophical zombies—for which she has and can have no justification for describing. This undermines her own attempt to explain what she means in mounting her own arguments and defending her conclusions.

Chapter 9, Ordinary Consciousness, continues the argument against neo-Dualists with the reminder that talented writers (if not philosophers!) are able to recount their unspoken trains of thought, memories, sensations, and feelings in such a way as to make them understandable. If these are not typical examples of conscious experiences, then it is not clear what would be. It would seem, however, that this understanding has been quietly dropped as expressions such as what it is like, conscious experience, and their kin transform into ones with a particular philosophical meaning with no clear connection to what we, in normal, nonphilosophical discussions understand by them. The conclusion that the neo-Dualist wants to support, however—that there is something we all recognize to be essential that is missing from a physicalist or functionalist framework—depends, as we have seen, on her using these terms with agreed application criteria. My arguments, however, are not intended to support the physicalist or functionalist who does, indeed, I agree, seem to be overreaching. The solution, however, is not to reinvent a private theatre within a functionalist framework or to suggest that consciousness is a pervasive and ultimate element of a physical one. We need to resist some of the key moves that figure earlier in the physicalist’s or functionalist’s argument. They should not be allowed, anymore than the behaviorist, to help themselves to full-blooded action descriptions or to riders that include mentions of other propositional attitude expressions. For once these descriptions are conceded, so too is the explanatory fabric in which rational, knowing, conscious, and even self-conscious beings are logically and grammatically presupposed.

It is true, however, that philosophers have not paid sufficient attention to, for example, situations in which we describe someone as thinking when there is no visible performance. Consider Ryle, who, far from being the representative behaviorist many still take him to be, was a severe critic of the very scheme in which the dichotomy between realism and reductionism has its home. After its publication, he conceded that The Concept of Mind’s emphasis on overt performances left him open to being misinterpreted as a behaviorist and that he needed to think more about contemplation. Taking Rodin’s famous sculpture, Le Penseur, as illustrative, Ryle spent his later years puzzling about scenarios that most tempt us to postulate an inner theatre to account for what is going on inside the agent’s head. Toward the end of his life, Ryle bemoaned the fact that he was unable to find a peg upon which to hang his various writings about thinking and thought. Chapter 10, A Peg for Some Thoughts, attempts to make some progress on Ryle’s behalf. In so doing, it continues the argument in Chapter 7, The Location of the Mind, that the bifurcation between the mind and the body, the inner and the outer, or the internal and the external is a mistake. Perhaps the idiom in the head—rather than signaling a