Concept Nativism: The Origins and Limits of the Human Conceptual System

Concepts are the representational components of thoughts. In everyday life it's easy to take our concepts for granted and overlook the astounding breadth and power that the human conceptual system must have in order to support the range of thoughts available to human beings. The conceptual system must encompass the representational resources necessary to support all the thoughts expressed in our everyday dealings with one another, as well as in all of the intellectual achievements of the human race from Plato to Shakespeare to Einstein. Any given person possesses an enormous number of different concepts-tens, if not hundreds, of thousands-across a multitude of different conceptual domains.

This raises the question of how our minds come to possess all the concepts that they do. What allows the minds of human infants to develop the vast number of concepts human adults possess? Are the mechanisms and processes of conceptual development constrained in interesting ways? What kinds of variation in conceptual resources are possible across different cultural groups? Are there limits to the flexibility of the human conceptual system? The aim of my proposed research is to come to a better understanding of how the human conceptual system develops, what aspects of this system are innate, and what types of learning and cultural variation are possible within this system.

One fundamental divide in the study of the mind is the divide between empiricism, historically advocated by such philosophers as Locke and Hume, and nativism, historically advocated by such philosophers as Plato and Descartes. Recent research in cognitive science has provided powerful support for a general nativist picture of the mind-one in which the human mind innately possesses many domain-specific representations and cognitive mechanisms (Carruthers, Laurence, and Stich 2005, 2006, 2007). This research suggests that a nativist approach to the human conceptual system is extremely promising.

However, concept nativism is widely rejected in philosophy. This is partly the result of general philosophical scepticism about nativism and partly the result of the prominence of the philosopher Jerry Fodor's highly implausible version of concept nativism. Fortunately, the general philosophical concerns about nativism stem more from misunderstandings than legitimate difficulties, and nativists need not adopt Fodor's particular account. In fact, the key to developing concept nativism lies in abandoning the central tenet of Fodor's account: his denial that concepts can be learned. Though not all concepts are learned, there is no incompatibility between nativism and learning. In many ways, the proposed research project is an extended reflection on this fact.

A large part of the proposed research project involves providing detailed accounts of the general structure of a variety of positive models of conceptual learning. Philosophical work on conceptual development has been misled by a focus on one or two idiosyncratic models of conceptual learning, and has done little to explore the space of possible options. One theme in the proposed research is that conceptual development is a heterogeneous phenomenon, with a variety of distinct types of mechanisms and processes contributing to the overall development of the conceptual system. A second theme is that it is not only complex concepts that can be learned (as is widely supposed by both nativists and empiricists). Primitive concepts (concepts not themselves composed of other concepts) can also be learned-and again, through a variety of distinct types of mechanisms. A third theme is that, surprisingly, nativists are actually better placed than empiricists to explain cross-cultural conceptual variation. Finally, the research examines some of the important philosophical consequences of this version of concept nativism, especiallyin connection with the question of the limits of the human conceptual system.

The proposed research is in the first instance basic foundational research and so to the extent that it has non-academic impact, this impact is likely to be indirect and long term, rather than direct or immediate. That said, work on the nature of conceptual development, its innate bases, and the likely patterns of conceptual variation and universality across cultures, has potential to impact such domains as: (1) education, (2) translation and cross-cultural understanding, (3) treatment of conceptual deficits, and (4) innovation. Accordingly, it could in principle have impact on policy makers or practitioners whose work impinges on these domains.

(1) Education. One part of education involves developing or learning new concepts. Foundational work on conceptual development might therefore lend itself to a greater understanding of which ways are more natural to learn or transmit new concepts, which types of concepts are more easily learnable, and why. These issues are not the direct focus of my research, but foundational work on the development of the human conceptual system is likely to eventually have this type of impact. The impact would likely come more directly from scholars working in psychology, linguistics, and computer science, and for this reason, I will aim to bring my work to the attention of scholars working in these disciplines.

(2) Translation and cross-cultural understanding. Translation and cross-cultural understanding effectively involve coordinating conceptual systems across languages and cultures. Accordingly, understanding the nature of conceptual variation could help to alleviate cross-cultural misunderstandings, facilitate the productive exchange of ideas across cultures, and enhance appreciation of conceptual diversity. Work on conceptual development is still at an early stage, and so these types of results are more likely in the long term. But my proposed work aims to deepen our understanding of the patterns of cross-cultural variation and universality.

(3) Conceptual Deficits. Conceptual deficits might arise through the dysfunction or absence of particular mechanisms of conceptual development/learning. It is possible that concepts that cannot be acquired through the normal route might nonetheless be acquired through an alternative route. Given the disunity of conceptual development, multiple routes to conceptual learning exist, though they are not all equally well placed to acquire all concepts. Work on the treatment of conceptual deficits is not well developed at present, and so this type of impact is perhaps even more long term than others. Ultimately, however, a clear understanding of the available mechanisms of conceptual learning is likely to be essential to the treatment of conceptual deficits.

(4) Innovation. Conceptual learning and development is also a key ingredient in much scientific, technological, and cultural innovation. Models of creativity and innovation are often premised on empiricist models of conceptual development. Nativist models of conceptual learning have the potential to suggest interesting new models of creativity and innovation. The development of these models could in principle lead in turn to greater and more productive forms of conceptual innovation.

As noted above, to the extent that there is potential for impact in these cases it would likely come more directly from scholars working in psychology, linguistics, and computer science. I will bring my work to the attention of scholars working in these disciplines.