Author’s Introduction (Full Text)
- Can creatures who do not have a language think? On the one hand, many types of non-linguistic creatures behave in ways that seem to require treating the creatures in question as thinkers. The evidence is not simply anecdotal. Much of the most exciting and influential recent research in developmental psychology, cognitive archeology, and cognitive ethology explicitly assumes that the capacity for thought is not in any way tied to language possession. On the other hand, we do not really have any way of attributing thoughts to non-linguistic creatures other than by crude analogy with the attribution of thoughts to language-using creatures. We have no theoretical framework for understanding the content and nature of non-linguistic thought or the mechanisms of reasoning and reflection of which non-linguistic creatures might be capable. And in the absence of such a theoretical framework, the practices of explanation within which the attribution of thoughts to non-linguistic creatures seems so necessary remain without a secure foundation.
- In this book I develop a framework for thinking about the thoughts of non-linguistic creatures that acknowledges the differences between thinking without words and thinking with words while nonetheless enabling us to attribute to non-linguistic creatures thoughts with many of the characteristic features of language-based thoughts. As will emerge, the thoughts of non-linguistic creatures can have compositional structure — they have distinguishable components that can feature in further thoughts. They possess determinate contents, and it is often possible, despite the claims of many philosophers, to identify and spell out with reasonable accuracy the precise way that non-linguistic creatures are thinking about their environment. Yet there are limits to the range of thoughts that non-linguistic creatures can entertain because, as I argue in chapters 8 and 9, certain types of thinking (all those involving intentional ascent, or thinking about thoughts) require a linguistic vehicle.
- Theoretical discussions of non-linguistic thought tend either to deny that there can be any such thing or to speculate about the vehicles of non-linguistic thought. Each of these general approaches is associated with a particular conception of the relation between thought and language. One very influential philosophical conception of the nature of thought, derived from the writings of Frege, is based on the principle that the study of thoughts can only proceed via the study of the sentences that express them. Theorists impressed by this conception of the nature of thought will find the notion of thought without language deeply problematic. Many of the problems that appear insuperable from the perspective of the classical philosophical approach seem less threatening if we shift paradigm and adopt the dominant understanding of the nature of thought within cognitive science, construing all thoughts (those of language-using creatures as well as those of non-linguistic creatures) as relations to sentences in an internal language of thought. Yet, whatever the merits of the language of thought hypothesis as an account of the mechanics of cognition, it cannot provide a full account of non-linguistic thinking. In particular, it does not provide us with an appropriate epistemology for the thoughts of non-linguistic creatures.
- In this book I develop a different approach to thought without language. Unlike the two approaches just outlined, the account I offer of thinking without words emphasizes the epistemological and explanatory dimensions of the problem. The best way of approaching the problem of thought without language, I suggest, is through providing an epistemological basis for the practice of attributing thoughts to non-linguistic creatures and for the psychological explanations within which those attributions take place. Accordingly, the book contains considerable detailed discussion of empirical material from developmental psychology, cognitive psychology, and studies of animal behavior, both experimental and ethological. The theory that emerges is philosophical, in the sense that it is based on a particular philosophical conception of the nature of thought and thinking, but that philosophical view is developed and refined through detailed consideration of the practical implications of attributing thoughts to non-language-using creatures.
- Plan of the Book:
- The basic problems that set the framework for this book are set out in chapter 1, where I outline the different types of question posed by the forms of psychological explanations of the behavior of non-linguistic creatures given in various parts of the cognitive and behavioral sciences.
- Chapter 2 explores the differing responses to these questions given by the two approaches to the nature of thought outlined earlier, and shows how neither can provide a fully satisfying account of thinking without words.
- Chapter 3 considers a deflationary or minimalist construal of the nature of non-linguistic thought that might be deployed to finesse the apparent need to attribute thoughts to creatures that are not language-users. The aim of the minimalist proposal is to show that thinking behavior in non-linguistic creatures can be understood in nonpropositional and perceptual terms, rather than through the attribution of propositional attitudes such as beliefs and desires. In opposition to this I suggest that there are important types of non-linguistic thought that cannot be accommodated in the manner proposed by the minimalist. The requirements of psychological explanation often demand that we attribute to non-linguistic creatures thoughts that are structured, represent the world in a highly determinate way, and reflect the particular mode of presentation under which the creature apprehends its environment.
- Chapters 4 and 5 are devoted to showing how we might go about attributing such determinate and structured thoughts to non-linguistic creatures. In chapter 4 I explain how a theorist might fix an ontology in a way that will allow the theorist to determine what objects a particular non-language-using creature is capable of thinking about — or, in other words, that will elucidate how the creature "carves up" its world into bounded individuals.
- Chapter 5 explores how a semantics1 can be provided for non-linguistic thoughts in a way that both does justice to philosophical constraints on acceptable theories of content and provides the ethologist or developmental psychologist with a workable method of assigning contents to the beliefs and desires of non-linguistic creatures.
- A theory of non-linguistic thought is incomplete without an account of non-linguistic reasoning and the norms of rationality by which such reasoning is governed. In chapter 6 I show how an account of non-linguistic rationality emerges when we pose the question: What could count as evidence that a non-linguistic creature is behaving rationally? There are several different forms of evidence that can come into play here. At the most sophisticated level, a creature is behaving rationally when it is sensitive to the consequences of different courses of action, but there are types of rationality that do not involve such consequence-sensitivity. Different forms of rationality are appropriate to different types of explanation, and I draw a distinction between level 1 rationality and level 2 rationality that maps onto the distinction between explanations of the type proposed by the minimalist and explanations that make use of belief-desire psychology.
- Reasoning and rationality are, of course, correlative notions, and in chapter 7 I pursue the question of the forms of inference available at the non-linguistic level. I offer an account of protoinference that respects the differences between linguistic and non-linguistic thought (and in particular the impossibility of explaining non-linguistic reasoning in formal terms), while at the same time offering analogues at the non-linguistic level of some basic forms of inference.
- In chapters 8 and 9 I address in more general terms the scope and limits of non-linguistic thought. I argue that there are certain types of thinking for which a linguistic vehicle is essential — and by this I mean a public language rather than a private language of thought. Roughly speaking, a linguistic vehicle is required for all types of thinking that involve intentional ascent, or what is sometimes called metarepresentation. I argue in chapter 8 that intentional ascent requires semantic ascent, on the grounds that intentional ascent requires the ability "to hold a thought in mind" in a way that can only be done if the thought is linguistically vehicled.
- In chapter 9 this argument is developed and the limits of non-linguistic thought plotted in the context of a range of different types of thought. These language-dependent cognitive abilities range from second-order reflection on one's own beliefs and desires and the capacity to attribute thoughts to others to the ability to entertain tensed thoughts and to deploy logical concepts. Many of these language-dependent cognitive abilities, however, have primitive analogues that do not involve intentional ascent and hence are available at the non-linguistic level.
- A final word. In previous work I have mapped the distinction between linguistic and non-linguistic cognition onto a distinction between conceptual and non-conceptual content. In The Paradox of Self-Consciousness2 I proposed that genuine concept mastery involves an ability not simply to make judgments involving those concepts but also to justify those judgments and to reflect on the grounds for them. Since these are paradigmatically language-dependent activities, it follows that concept mastery requires the possession of a language. This line of thought still seems to me to be fundamentally correct, and the discussion of the differences between linguistic and non-linguistic cognition in chapters 8 and 9 provides tools for developing it further. However, I have deliberately written this book in a manner that does not involve or require a particular interpretation of the distinction between conceptual and non-conceptual content. It seems to me that a clear view of the distinction between conceptual and non-conceptual content is more likely to emerge from a clear understanding of the differences between linguistic and non-linguistic cognition than vice versa.
- Note to the Reader: The problems with which this book is concerned are of broad interdisciplinary interest, and I have written the book with an interdisciplinary audience in mind. I hope that the general line of argument will be of interest to anyone with a theoretical, experimental, or practical interest in non-linguistic creatures and how they represent and think about the world. The book does not presuppose a philosophical background, but there are certain sections that will perhaps be most directly of immediate concern to philosophers. Readers without a philosophical background could omit sections 2.1, 2.2, and 2.3 on an initial reading, moving straight from the end of chapter 1 to the final section of chapter 2. In chapter 3, section 3.2 could initially be omitted.
Text Colour Conventions (see disclaimer)
- Blue: Text by me; © Theo Todman, 2020
- Mauve: Text by correspondent(s) or other author(s); © the author(s)