INTRODUCTION

In this chapter we will use the methods introduced in previous chapters to clarify and extend Kuhn's mature account of incommensurability. We will show that incommensurability between conceptual structures is created by changes that are neither total nor instantaneous. We will also draw out various conclusions that Kuhn suggested but did not elaborate, for example, that incommensurability varies in degree or importance and that the degree correlates with the position of a concept in a hierarchy or conceptual structure as depicted by the corresponding frame. Throughout the next two chapters our main historical focus will be the Copernican revolution, an episode that Kuhn never treated satisfactorily (Westman 1994; Barker 2001). On the basis of our new account, we will suggest that incommensurability may occur even within what Kuhn and earlier writers have regarded as a single paradigm and that this kind of conceptual difficulty may in itself be a motive for conceptual revision. As we will see in Chapter 6, one of the most important motives for Copernicus' revision of Ptolemaic astronomy was a problem of just this type.

THE DEVELOPMENT OF KUHN'S CONCEPT OF INCOMMENSURABILITY

Kuhn significantly refined his philosophical account of science in the years after the publication of The Structure of Scientific Revolutions. The concept of incommensurability also underwent major revisions. From an initial description that emphasized similarities to visual gestalt switches, Kuhn moved in the 1980s to an account that described incommensurability in solely linguistic terms.

THE PLACE OF KUHN'S WORK IN STUDIES OF SCIENCE

Thomas Kuhn's Structure of Scientific Revolutions became one of the most influential books of the twentieth century, although its author suffered the fate of many prophets: he was ignored by the people he most hoped to influence. His technical terms became so widely known that a popular cartoonist could depict a newly hatched chick greeting the world with the cry “Oh! Wow! Paradigm shift!” (Taves 1998) and a best-selling guide to success in life and business would tell its readers, “[W]e need to understand our own ‘paradigms’ and how to make a ‘paradigm shift’” (Covey 1990: 26). But there is no Kuhnian school of history, and many philosophers of science remain skeptical about his ideas. At the close of the twentieth century philosophers generally rejected paradigm shifts and normal science as useful categories for understanding scientific change and were still arguing about another key idea, incommensurability (Curd and Cover 1998; Hoyningen-Huene and Sankey 2001). Meanwhile Kuhn's emphasis on the historical variability of scientific standards and the role of research communities in scientific change was embraced by a new generation of sociologists of scientific knowledge. The new sociologists of science adopted Kuhn as a founding father, if not an intellectual guide: Kuhn's emphasis on the cognitive content of science was marginalized.

THE CONCEPTUAL STRUCTURE OF PTOLEMAIC ASTRONOMY

In astronomy before Kepler the path of a planet was not its orbit but the pattern of its motion seen by an observer on a stationary earth against the hypothetical sphere of the heavens. It was recognized in antiquity that this pattern was not a real motion, but a complex outcome of the observer's viewpoint and a variety of circular motions that acted together. The task of astronomy was to define this pattern – to specify the path of the planet in this original sense. The real motion of the planet – its track in what we would now call three-dimensional space – was unknown, and possibly irrelevant. All other considerations – the causes of celestial motion, the actual dimensions of the heavens – were the business of a separate science, cosmology. It was well known that the goal of astronomy could be achieved, that is, the path of a celestial object could be predicted, without making specific assumptions about its distance from the earth, once appropriate rates of rotation were introduced (Evans 1998; Pedersen 1993).

The basic data of astronomy from antiquity to the sixteenth century – the explananda or, if you prefer, the ‘phenomena’ that needed to be ‘saved’ – were recorded observations of planetary positions. Sixteenth-century astronomy texts devoted most of their attention to motion in longitude.

After the appearance of The Structure of Scientific Revolutions in 1962 Kuhn attempted to develop a Wittgensteinian account of family resemblance concepts for a domain that other philosophers had found most unlikely: scientific concepts. In the 1970 postscript to Structure Kuhn even suggested that the variants of Newton's second law that applied to different physical systems showed family resemblance, but no single defining properties. If this proved to be the general case, the most important examples of scientific concepts might turn out to be family resemblance concepts rather than the well-behaved concepts, analyzable by necessary and sufficient conditions, expected by earlier philosophers of science. Kuhn returned to these themes again and again in his later philosophical writings.

Kuhn's theory of concepts focused on a restricted class of terms, namely, kind terms. As Kuhn defined kind terms they are “primarily the count nouns together with the mass nouns, words which combine with count nouns in phrases that take the indefinite article. Some terms require still further tests hinging, for example, on permissible suffices” (Kuhn 1991: 92). Thus, kind terms include natural kinds, artifactual kinds, and social kinds.

EXEMPLARS

An important source for Kuhn's theory of concepts was his early reflection upon science teaching. He made the observations that, first, science education is based entirely on prepared teaching materials, and, second, this teaching confers the ability to recognize resemblances between novel problems and problems that have been solved before.

A frame is a hierarchy of nodes (Figure 10). The origin of this notation may be traced to the British psychologist Sir Frederic Bartlett, who introduced the notion of a schema in his famous study of memory (Bartlett 1932). During the 1970s researchers in artificial intelligence developed and applied frames for a variety of purposes, including computer-based representations of everyday human activities (Schank 1975; Schank and Abelson 1977) and vision (Minsky 1975; Brewer 2000). During the 1980s, the American cognitive psychologist Lawrence W. Barsalou introduced frames in his studies of ad hoc categories (Barsalou 1982, 1991), autobiographical memories (Barsalou 1988), and contextual variability in concept representations (Barsalou 1987, 1989; Barsalou and Billmann 1989). He extended and refined previous presentations of the frame notation to represent concepts (Barsalou 1992b; Barsalou and Hale 1993), calling his new approach ‘dynamic frames’. The present authors adopted his techniques in the 1990s and began to apply them to conceptual change in science and the implications of Kuhn's mature work (Andersen, Barker, and Chen 1996; Chen, Andersen, and Barker 1998; Barker, Chen, and Andersen 2003).

CONSTITUENTS OF DYNAMIC FRAMES

In the explanations that follow we will draw increasingly complex diagrams to represent frames, and we distinguish the concepts appearing in frames by capital letters when we mention them in the text. To represent a concept by a frame, one layer of nodes is selected to represent attributes of the concept.

In this final chapter we attempt to do three things. First, we review the results we have presented. Next we consider the implications of our position for one of the major controversies within philosophy and sociology of science, the realism debate. Finally, we consider the significance of our results for wider debates in the history, philosophy, and sociology of science.

RESULTS

Our goal throughout this book has been to recover and extend Kuhn's account of scientific change by showing that its most important features are consequences of the nature of concepts, as currently understood in cognitive psychology and cognitive science. An important subsidiary point is that Kuhn's own theory of concepts has been shown to be independently supported by work in cognitive psychology and cognitive science.

We have shown that there is a defensible distinction between normal science and revolutionary science, but that the difference between them is not a question of the historical rarity of one process versus the other. Viewed as conceptual changes both processes may occur at any time. In the case of revolutionary change, whether the result is a big revolution or small one depends on other factors – the status of the conceptual structure that changes (for example, whether or not the changes affect a fundamental item in the ontology of the field), as well as the speed and completeness with which the changes are adopted.

In Chapter 2 we gave an account of concepts and conceptual structures based on family resemblance. We showed that on this account possession of a conceptual structure implies knowledge of ontology, as objects not belonging to any of the known similarity classes are assumed not to exist. Likewise, we showed that through the relations of similarity and dissimilarity, possession of a conceptual structure implies knowledge of regularities, that is, expectations of the different situations that nature does and does not present.

In Chapter 3 we have seen how conceptual structures of the kind introduced by Kuhn may be represented by dynamic frames, a form of representation developed in cognitive psychology and independently supported by empirical research. Frames not only accommodate the most important features of Kuhn's account, such as family resemblance, but may also be used to represent graded structure, the most important empirical phenomenon documented by studies of categorization supporting the reality of family-resemblance categories. The frame account allows us to display details of conceptual structures that are otherwise difficult to examine, such as the patterns of attribute-value sets that characterize concepts, and it allows us to locate constraints between elements of the structure that correspond to knowledge of ontology and knowledge of regularities.

We have already suggested a developmental perspective: a particular conceptual structure is always given by the preceding generation, which passes it on to the next.