Preface and Guide to the Book

Professor of Philosophy and Director of the Hang Seng Centre for Cognitive Studies Professor of Philosophy This book argues for a rationalist account of the origins of human concepts—that is, for a version of concept nativism. While this type of account comes in many varieties, they all take the mind to possess a rich innate structure that plays a central role in explaining the origins of concepts. Our own version of concept nativism holds that many concepts across many conceptual domains are either innate or acquired via learning mechanisms that involve innate representations or other innate special-purpose elements. Drawing on a broad range of evidence from many different disciplines, we argue that there is a powerful case to be made in favour of this view. However, we are also keenly aware of the fact that the rationalism-empiricism debate is widely seen as being irrelevant to contemporary theorizing about the mind—or worse, as being fundamentally confused—and that this has led many philosophers and cognitive scientists to dismiss it altogether. If this scepticism regarding the value and coherence of the rationalism-empiricism debate were warranted, our project would be doomed from the start. So in addition to making a case for our rationalist view over competing alternatives (both rationalist and empiricist), it is essential that we also address the fundamental challenges that call the debate itself into question. Part IV of the book provides a comprehensive rethinking of the theoretical foundations of the rationalism-empiricism debate which clarifies what exactly the debate is about—as well as what it is not about—and at the same time makes clear why it remains central to the study of the mind. In our view, the rationalism-empiricism debate should be understood to be about the differing views that rationalists and empiricists hold regarding the collection of innate psychological structures which constitutes the ultimate psychological basis for the acquisition of all further psychological traits. Likewise, the more specific debate about the origins of concepts should be understood to be about the differing views that rationalists and empiricists hold regarding the collection of innate psychological structures which constitutes the ultimate psychological basis for concept learning. This way of understanding the rationalism-empiricism debate is not new. But it has never been fully articulated and is frequently conflated with (or rejected in favour of) a number of prominent alternative ways of understanding the debate that turn out to be intellectual dead ends, especially the view that it is about nature versus nurture (or the relative contributions of genes versus the environment). Both critics of the rationalism-empiricism debate and its proponents and participants frequently conceptualize it in these mistaken and unproductive ways, often conflating several incompatible interpretations of the debate without realizing it. We see the widespread scepticism regarding the value and coherence of the debate as stemming directly from such misunderstandings. While critics have rightly regarded these ways of understanding the debate as unworkable, they have been wrong to conclude that the debate itself should be abandoned as a result. Instead, what’s needed is a better understanding of the debate. This understanding should be built around the idea that we started with—that rationalists and empiricists differ in terms of the ultimate psychological basis that they posit for acquiring all further psychological traits. By systematically developing this interpretation of the debate—and sharply distinguishing it from unproductive alternatives—Part IV establishes a sound theoretical foundation for the debate, providing a detailed framework for understanding the diverse range of possible rationalist and empiricist theories and how they relate to one another. In Part IV, we turn to our positive case for concept nativism. As we see it, there is an overwhelming case to be made in favour of our view that many concepts across many conceptual domains are either innate or acquired via learning mechanisms that involve innate representations or other innate special-purpose elements. In making this case, we distinguish and clarify seven distinct types of argument supporting concept nativism, many of which have been poorly understood or insufficiently appreciated. Since our view is that a rationalist view about the origins of concepts is the right view to hold for many concepts across many conceptual domains, our discussion needs to cover a broad range of concepts from different conceptual domains. An exhaustive treatment of each of our seven arguments for concept nativism as it applies to every candidate concept and conceptual domain is out of the question. Instead, to make the discussion manageable, we have chosen to illustrate the breadth of our account—the range of concepts and conceptual domains that it covers—by bringing in new examples as we introduce each new argument. To illustrate the depth of the case for concept nativism—the fact that often many of these arguments apply to a given type of concept or conceptual domain—we examine a selection of conceptual domains from the vantage point of a number of these different arguments. While each of these seven arguments individually supports a rationalist perspective, the full force of the case for concept nativism comes from their collective impact and the recognition that they comprise what amounts to a single multifaceted inference to the best explanation argument for concept nativism. This argument not only demonstrates that a rationalist account of the origins of concepts should be adopted over competing empiricist accounts, it also shows why our version of concept nativism should be adopted over competing rationalist accounts (e.g., what are known as core knowledge accounts) which take there to be considerably less rich innate structure underlying concept learning. Part IV critically examines the empiricist opposition to concept nativism. Our critique of this opposition is organized around a representative selection of some of the most important and influential empiricist accounts of concept acquisition. One common theme of Part IV is that these empiricist proposals fail to do justice to the theoretical and empirical considerations that drive the rationalist accounts they are meant to be alternatives to. At the same time, however, we argue that work in the empiricist tradition contains valuable insights about conceptual development. We argue that not only are these insights consistent with concept nativism but that they can make a more significant contribution to explaining conceptual development when incorporated into a rationalist framework. Our discussion in Part IV extends both the breadth and depth of conceptual domains covered in relation to the arguments for concept nativism in Part IV by illustrating ways in which many of these arguments apply to new conceptual domains. We conclude that an examination of empiricist alternatives to concept nativism only serves to strengthen our case for rationalist accounts of the origins of concepts in general, and for our own version of concept nativism in particular. Finally, Part IV addresses what is perhaps the most famous contemporary position in the rationalism-empiricism debate regarding the origins of concepts, namely Jerry Fodor’s influential view that semantically primitive concepts (concepts that aren’t composed of more basic representations) can’t be learned and the corollary Fodor argued for that virtually all lexical concepts are innate (a view known as radical concept nativism). One of the reasons that Fodor’s arguments against concept learning have figured so prominently in this debate—despite the wildly counterintuitive conclusions they are associated with—is that it has proven to be remarkably difficult to say exactly where they go wrong. But perhaps even more importantly, many theorists see Fodor’s arguments as containing a deep insight about learning that imposes a fundamental constraint on any theory of concept learning; they just see Fodor as having drawn the wrong moral from this insight. While rejecting Fodor’s radical concept nativism, these theorists agree with Fodor’s claim that semantically primitive concepts cannot be learned and so must be innate. In fact, this view about conceptual structure and the limits on what can be learned lies behind a nearly universally accepted model of concept acquisition—endorsed in different ways by rationalists and empiricists alike—which we call the Acquisition by Composition model (or ABC model) of conceptual development. According to this model, concept learning requires that the learned concept be a complex concept which is formed from a compositional process that builds the new concept out of its semantic constituents. The heart of Part IV of the book is directed at showing why this model is mistaken. Our discussion encompasses an overview of the history of Fodor’s views on these issues, which changed substantially over a period of more than thirty years. By carefully analysing Fodor’s arguments, we show precisely how they go wrong, which in turn shows why the ABC model of conceptual development should also be rejected. The rejection of this model opens up a range of new possibilities for explaining how concepts can be learned which we explore in relation to a variety of different types of concepts and different theories of meaning for mental representations. This discussion further underscores a major theme of the book—that rationalist accounts of the origins of concepts not only are consistent with concept learning but also offer the best overall account of how concept learning works. We end Part IV on this theme by highlighting the depth of the connection between our own rationalist account of the origins of concepts and cultural learning. Since this is a long book, we have tried to arrange it in such a way that the four main parts of the book can be read on their own or out of order (though readers who do this may need to consult Chapter 2 and Chapter 6 for key terminology we use later on). Likewise, most chapters are sufficiently self-contained that readers who are interested in particular topics can jump ahead to the relevant chapter. However, it should be kept in mind that the theoretical framework in Part IV and the many arguments, examples, and empirical findings that are discussed in different chapters in Parts II–IV are meant to interact with and support one another as part of a single integrated argument for concept nativism that runs through the entire book.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.