Brains, Practices, Relativism: Social Theory after Cognitive Science (review)

Turner's book is a collection of previously released work that holds out the promise of providing an alternative path for social theory. Turner, a philosopher, combines an outsider's perspective (on sociology) along with a solid understanding of such key social theorists as Mead, Durkheim, Weber, and Geertz. Including the term cognitive science in the title suggests an approach taking into account the recent findings in that field for constructing a new social theory. However, it is not always clear whether his proposed "new approach" intends to build a new model of social theory based on a cognitive science viewpoint of individual action (that is, counterposed to the more common rational choice model often employed), or whether it simply wants to use the findings of recent cognitive science work to provide a new metaphor for viewing social life.

Regardless, Turner argues forcefully that theory must begin at the level of the individual. His model of individual action builds on the cognitive psychology of connectionism: "humans are multilayered neural networks that learn, under the continuing pressure of experience, by the gradual modifications of the strengths or 'weights' of their myriad synaptic connections." The importance of this viewpoint is that "every mind is the product of a distinctive and individual learning history." This historical view of individual thinking processes suggests problems for a rational choice model, but, other than occasionally echoing Weber's view that rational thinking occupies only a small part of a social world shared with emotions and habit, Turner's real target is theories positing units at higher levels of analysis than individuals. He argues that the proper goal is to develop a theory of emergence where higher-level concepts emerge from individual activity.

His justification for a theory built at the individual level is parsimony, arguing that adding "logics" unnecessarily complicates things conceptually. That is, higher levels of theorizing require a whole set of causal links from these abstract objects to the actors who actually do the proximate explanatory work. But while Turner desires to avoid the theoretical complexity of James Coleman's "boat" showing the linkages between the micro and the macro, the burden of proof then lies upon his effort to show that such structure is not necessary for explaining individual action, and it is not clear that he succeeds.

For Turner, the perspective of connectionism views the brain as a parallel distributed processing system that is given a learning algorithm (but no detailed rules) and is then trained up by feeding it data and giving feedback for "correct" answers. Thus we can view attitudes as nodes in connectionist learning networks. The key question then is whether this model can account for higher mental processes. This suggests interesting implications, not the least of which is that it problematizes the neo-Kantian "shared premises" approach of much social theory: There is no need for shared premises from some initial starting point, but instead societal norms and values simply arise from interaction over time. Thus we are left with a path-dependent model of social structure.

Chapter 1 does a nice job of laying out Turner's argument. In chapter 2 he strongly critiques Searle and models requiring a "higher level" of intentionality: we-intentionality rather than simply I-intentionality. Chapter 3 displays his emulationist model, arguing that interactions over time can explain the emergence of culture. In chapter 4 he critiques the necessity for the shared premises model, arguing that other models are possible, including his path-dependent approach. Chapter 5 argues that relativism ultimately results in presuppositions and, as a result, a model where change is difficult to explain. In chapter 6 he dives into the question of normativity, pointing out the difficult distinction between a "norm" and "training" and providing the provocative conclusion that we simply read normativity into the world. The later chapters seem to stray from the model, rather than pushing forward the argument. Chapter 7 provides a discussion of presentism and contextualism...