The Cognitive Science of Social Science: Still Waiting for the Revolution

Review of Cognitive Dimensions of Social Science, Mark Turner, Oxford University Press, 2001.

Mark Turner’s book brings a welcome call for a partnership between or “merging” of cognitive science and the social sciences (11). Each chapter is a relatively independent effort to show the form such an intersection might take by applying recent work by Turner and others (e.g. Fauconnier and Turner, 1998) on conceptual “integration” or “blending” to the understanding of social science. Such “blending” is a general capacity of persons to construct new conceptual combinations out of different “conceptual influences” or “contributing spaces.” Just what the metaphors of “blending” or “spaces” come to can be difficult to pin down. Turner introduces this central idea with combinations of simple lexical concepts. For example, Turner suggests that the concept front-loaded IRA is a combination of the concept front-loaded and the concept individual retirement account (16). What Turner and his collaborators have been seeking is an account of the mental capacity that selects elements associated with contributing concepts and background knowledge to be associated with the new complex concept. Understanding this mental capacity - understanding the way that, for example, prototypical features of concepts are selectively retained or evoked as those concepts are combined - is of deep importance to clarifying how we come to view particular information as relevant or irrelevant to an issue or situation. Turner is especially concerned with the role blends play in supporting counterfactual and figurative reasoning and the crucial role such forms of reasoning play in the many projects of the social sciences. Readers seeking an entry to the theory of concepts, the literature on conceptual blending, or other recent efforts to understand psychological relevance should look elsewhere for Turner gives only sketches of the theory of conceptual blending in service of arguing for the relevance of the cognitive sciences to the social sciences.

The first and longest chapter (comprising nearly a third of the book) is devoted to illustrating this relevance by a reading of Clifford Geertz’s seminal “Deep Play: Notes on the Balinese Cockfight” (1973a). While qualitative approaches to culture are often thought to be antithetical to an emphasis on cognitive mechanisms (e.g. the dispute over the emotions discussed in Mallon and Stich 2000), Turner offers little but praise of Geertz’s methodological approach, taking Geertz’s work to be struggling “just below the surface, but not any less powerfully, to discern the basic cognitive operations that make the invention of the cockfight possible” (15). Turner proceeds to show how the capacity to integrate concepts is implicated in the explanation of this elaborate social institution. The structure of the argument here is one that is repeated in later chapters: show how the cognitive capacity for conceptual blending underlies the phenomena of interest to social scientists.

In the second chapter, Turner turns his attention to “qualitative social science,” a term that he uses not for Geertz-style cultural anthropology, but as a label to pick out recent efforts to introduce careful scientific methodology into research efforts that resist easy quantification (e.g. the sort of efforts suggested in King, Keohane, and Verba 1994) or by Tetlock and Belkin 1996). A special emphasis of such efforts has been to focus research design on the consideration of counterfactual situations in the course of sound causal reasoning. And, Turner argues, the consideration of such non-actual but possible situations requires the resources of conceptual blending to succeed (78). Turner’s discussion of blending in this chapter allows one to appreciate the role of cognitive metrics of relevance in thought and interpretation. Here, Turner also goes beyond merely emphasizing the importance of counterfactual reasoning in social scientific modeling to argue that qualitative theorists have constructed models that are problematic in light of the facts about cognitive blending.

The form of Turner’s argument will be familiar to those aware of the challenge to traditional economics apparently posed by cognitive studies of reason and behavioral economics in the “heuristics and biases” tradition of Daniel Kahneman and Amos Tversky (Kahneman and Tversky 1973, Tversky and Kahneman 1974, Kahneman, Slovic, and Tversky 1982). Traditional economics offers abstract models of phenomena based on the assumption of rational actors, while experimental results show time and time again that people deviate systematically from rational actor assumptions. Turner discusses models of choice in Chapter 3, and it is to this chapter that economists and philosophers of economics interested in the relationship between classical economics and empirical work may wish to turn first. Within the context of Turner’s project, Chapter 3 provides another example of the way in which conceptual blending (and the psychological mechanisms of reason and choice more generally) are implicated in the practice of everyday social scientific research.

In Chapter 4, Turner makes the case that conceptual blending is crucial to analogy. While he does emphasize the importance of analogical reasoning to a variety of intellectual endeavors, he does not do enough to relate it to specific social scientific research programs.

Chapter 5 is entitled “Descent of Meaning.” Here Turner draws an analogy between the evolution of meaning via conceptual blending and biological evolution. While provocative, the chapter does little to engage existing work. For example, it does not discuss “meme”-based approaches to culture (e.g. Dawkins 1976, Dennett 1995), nor the evolution and transmission of “evoked” culture (e.g. Atran 1998, Boyer1994, Sperber 1996), nor does it mention recent efforts to formally model cultural evolution (e.g. Boyd and Richerson 1985). Although there are many differences among these thinkers, they share with Turner the important idea that cultural evolution and biological evolution may be understood as sharing certain general features. Unfortunately, Turner’s chapter raises more questions than it answers without engaging any of the substantial existing literature on this subject.

As they exist today, the social sciences have little more than a nominal unity. Perhaps for this reason, discussions of the social sciences tend to suffer from an excess of generality, or from an emphasis on what seem, in the light of the diversity of social scientific research programs, to be idiosyncratic examples. Turner wisely escapes this fate by structuring his book as a series of case studies, with each chapter relating conceptual blending to a different influential text or research program. Unfortunately, the chapters are not well integrated with one another, and the book reads more like a collection of papers than a sustained argument. In addition, nearly every chapter would have benefited from more engagement with existing research paradigms, more detailed proposals demonstrating the relevance of the cognitive sciences to the social sciences, and even a clearer exposition of the central ideas of the book, including the idea of conceptual blending.

One persistent problem that ought to be of special concern to anyone seeking to integrate work in the cognitive sciences with the social sciences is Turner’s recurring confusion of intentional or mental representations of the world and the world itself. Consider his discussion of the Balinese cockfight:

The Balinese social man … is disintegrated in the blend. The status part of him, his psychological security, and his social “face” project from the influencing space with Balinese social man to the cock in the blend. His internal aggressiveness and his impulse to explicit social affront – completely inhibited in his life outside the blend – not only project into the cock but also leave their inhibitions behind as they go. The owner’s virility and the virile activity of his penis also project into his cock. (33)

Conceptual blending is a mental operation involving the production of one object of thought by combining other objects of thought. But in passages like this one, Turner switches back and forth between objects of thoughts (e.g. concepts) and the objects or properties those concepts represent. It is not the concept of the Balinese social man that has “internal aggressiveness” and an “impulse to explicit social affront.” Nor does the concept have a “life outside the blend” or, for that matter, a penis. These are all properties of the men themselves.

In philosophical jargon, such passages confuse using a concept or word to think or talk about the world and mentioning it (which is what we do to talk about the word or concept itself). While he is sometimes careful to avoid this error (20), Turner often takes for granted the relationship between the meanings or concepts (which are blended) and the actions, institutions, and artifacts these meanings and concepts represent or shape (which, if they are “blended” at all, it is in a different sense than concepts are blended). This equivocation is significant because one wants to know how it is that blended spaces (understood as “conceptual blends”) give rise to blended spaces (understood as actual events, e.g. actual cockfights). What Turner seems to have in mind is that social institutions like cockfights are shaped by the meanings that participants associate with them – an idea that is familiar to many social scientists. But he never makes it clear how he thinks that the study of the capacity to have thoughts with such meanings is supposed to illuminate the character of the social institutions that result.

The various chapters also reveal a fundamental ambivalence in Turner’s aim. In stating his overarching conclusion, Turner alternates between considering cognitive mechanisms “complementary” to social scientific discussions (e.g. 16), and imagining the study of cognitive mechanisms as transforming the practice of the social sciences, ultimately creating a new discipline or set of disciplines. The problem is that the weaker conclusion is of little interest to ongoing research in the social sciences. In contrast, a successful demonstration of the stronger conclusion would require substantial shifts in theory and research strategies, but Turner never establishes anything so dramatic.

The first, weaker reading of Turner’s project is most fully on display in Chapter 1, but recurs to a lesser extent in every chapter that follows. On this reading, to show that cognitive mechanisms are complementary to processes discussed by social sciences is simply to show that they operate in those processes, underlying and explaining crucial aspects of human action and interaction. For example, Turner argues that the capacity for conceptual blending allows social agents to engage in complex social enterprises like the Balinese cockfight. But such a conclusion is far too weak to be of central interest to working social scientists, for it offers no critique or illumination of existing disciplinary practice (just as Turner offers no critique of Geertz’s anthropological approach). Rather, it simply points to the compatibility of independent endeavors. While some social scientists, particularly those engaged in various sorts of qualitative research (and including Geertz himself (1973b, 37ff)) decry the relevance of any universal cognitive mechanisms to cultural research, few go so far as to say that such mechanisms do not exist! Rather, they believe that cognitive mechanisms do not explain the differential cultural content that is at the heart of such qualitative inquiry.

Consider vision. Cognitive research into vision has resulted in a fascinating and fruitful set of research programs. Moreover, the mechanisms these research programs investigate and theorize about so successfully are deeply important to culture and behavior. Without vision, or with a radically different sort of vision, human life across cultures would be radically transformed. Moreover, the capacity for vision underlies the very processes that cultural anthropologists care deeply about: processes like the attachment of meaning to actions, objects, and events, the negotiation of understandings, the coordination of behavior, the distribution of resources, and so forth. No social scientist should deny this. But none of this gives any reason to think that the social sciences are impoverished by the failure to incorporate more fully insights from the cognitive study of vision. Rather, the picture here – as in Turner’s discussion of Geertz – is one of complementary, but independent research programs.

Calls to forge some closer connection between cognitive science and social science are compelling only when accompanied by demonstrations that some or another social or cognitive investigation would be enriched or transformed by the data, theories, or methodologies from another field. This is what, for example, recent advocates of evolutionary psychology (e.g. Tooby and Cosmides 1992) have attempted to do. They aim to show that social science informed by evolutionary psychology should transform or replace existing theories and research paradigms across the social sciences.

Turner offers a more critical approach to existing research practices in Chapter 2 and to a lesser extent in Chapter 3, attempting to show that facts from the study of human cognition call for revision of social scientific practice. His discussion here is intriguing, full of provocative examples, yet much of the discussion is a promissory note regarding the potential for cognitive research to influence social scientific practice. Unfortunately, where Turner’s discussion turns to specific criticisms, the discussion suffers from a failure to show that the empirical model of blending he is proposing actually undermines the assumptions of contemporary social science. Consider Turner’s attack on Gary King, Robert O. Keohane, and Sidney Verba’s model of “thought experimentation” in the social sciences. King, Keohane, and Verba require that counterfactual thought experiments compare scenarios in which “‘everything remains the same, except’ for a single explicit change”(Turner 82). But Turner suggests this is psychologically unrealistic for it assumes “that we already understand how counterfactual reasoning works merely because we are so good at it” (82). He continues:

the counterfactual space in a thought experiment involving the real political world is almost never a mere copy of one of its influencing spaces (in this case, the real election) with one explicit change. On the contrary, the projection to the counterfactual space is typically highly selective. … this is a shaky requirement to impose on all counterfactual reasoning and hence on all causal reasoning. (82)

The criticism is misconceived, because Turner fails to appreciate the normative character of King, Keohane, and Verba’s requirement. The requirement that a thought experiment involve only one change is similar to the requirement that the independent variable be altered between test and control groups in any experiment. In both cases, the requirement arises out of the demands of causal reasoning. And if the demand is out of place in thought experiments, it is out of place in actual experiments as well. This sort of requirement is fundamentally normative. In cases in which we cannot manipulate the independent variable alone, we are not inclined to think there is something wrong with the requirement that we do so. Rather we are inclined to say that an experiment has not succeeded in eliminating alternate possible causes.

Of course, if our best cognitive psychology tells us that people cannot successfully engage in a certain sort of reasoning, a methodology that requires them to do so might be in trouble. But to see that cognitive psychology has such a result, we would need a much closer and more careful study of particular counterfactual claims and the character of the distortions that result than any Turner suggests. Even so, in undertaking such a project, we do not abandon the normative project of understanding ideal causal reasoning. Rather, such a project requires an understanding of the norms described by King, Keohane, Verba and others in order to understand the ways in which our reasoning may fall short of such norms in particular cases. Nor is this sort of problem isolated. Distinguishing normative, descriptive and other research aims is essential to clarifying the relationship between the results of cognitive science and the many projects of the social sciences (e.g. see Samuels, Stich, and Faucher 2002 for a careful discussion of various research aims in the study of reasoning and rationality).

Having said this, there is also much that is admirable in Turner’s book. It is written in an engaging style and full of thoughtful insights. It covers a great deal of intellectual ground in an academic environment in which even furtive steps outside safe disciplinary boundaries require great efforts, efforts that usually seem inadequate to those entrenched in ongoing intradisciplinary conversations. Moreover, the prospect of principles guiding what Turner calls conceptual blending is intrinsically interesting, and a better understanding of such principles would no doubt have broad implications for how we understand cognitive practices across the social sciences as well as in the humanities.

But when one agrees with Turner, as I do, that disciplinary intersections between the cognitive and social sciences will likely be fruitful for both, one cannot help but be disappointed in the absence of the sort of close argumentation and concrete examples of cognitive research that require social scientists to take notice. In this way, Turner’s illustrations involving conceptual blending compare poorly with, say, work by behavioral economists undermining (some construals of) rational choice theory or work by evolutionary psychologists arguing for the importance of cultural universals. Turner is correct that there is much fruitful and interesting work to be done in this area, but one leaves this book wishing he had done more of it.

-Ron Mallon, University of Utah

Bibliography

Atran, S. 1998. “Folk Biology and the Anthropology of Science: Cognitive Universals and Cultural Particulars.” Behavioral and Brain Sciences 21: 547-609.

Boyd, R. and P.J. Richerson. 1985. Culture and the Evolutionary Process. Chicago: University of Chicago Press.

Boyer, P. 1994. The Naturalness of Religious Ideas: A Cognitive Theory of Religion. University of California Press.

Fauconnier, G. and M. Turner. 1998. “Principles of Conceptual Integration.” In Discourse and Cognition: Bridging the Gap. Jean-Pierre Koenig (Ed.) Stanford: Center for the Study of Language and Information. Pp. 269-283.

Geertz, Clifford. 1973a. “Deep Play: Notes on the Balinese Cockfight.” The Interpretation of Cultures. New York: Basic Books. Pp. 412-453.

Geertz, Clifford. 1973b. “Thick Description: Toward an Interpretive Theory of Culture.” The Interpretation of Cultures. New York: Basic Books. Pp. 3-30.

Kahneman, D., P. Slovic and A. Tversky. Eds. 1982. Judgment Under Uncertainty: Heuristics and Biases. Cambridge: Cambridge University Press.

Kahneman, D. and A. Tversky.  1973. “On the psychology of prediction.” Psychological Review, 80, 237-251. Reprinted in Kahneman, Slovic, and Tversky (1982).

King, G., R.O. Keohane, and S. Verba. 1994. Designing Social Inquiry: Scientific Inference in Qualitative Research. Princeton, NJ: Princeton University Press.

Mallon, R. and S. Stich. 2000. “The Odd Couple: The Compatibility of Social Construction and Evolutionary Psychology.” Philosophy of Science 67: 133-154.

Samuels R., S. Stich, and L. Faucher. 2002. "Reason and Rationality". In Handbook of Epistemology. I. Niiniluoto, M. Sintonen, & J. Wolenski, (Eds.) Dordrecht: Kluwer Academic publisher.

Sperber, D. 1996. Explaining Culture. Cambridge, MA: Blackwell Publishers.

Tooby, J., & Cosmides, L. 1992. “The psychological foundations of culture.” In The Adapted Mind. J. Barkow, L. Cosmides, & J. Tooby (Eds.) New York: Oxford University Press. Pp. 19-136.

Tversky, A. and D. Kahneman. 1974. “Judgment under uncertainty:  Heuristics and biases.”  Science, 185, 1124-1131. Reprinted in Kahneman, Slovic, and Tversky (1982).