Many philosophers and cognitive scientists are now familiar with a traditional debate between two accounts of folk-psychology (the intuitive framework of beliefs, desires and intentions we use everyday to understand each others). According to the first one–the theory-theory account, or "information-rich"–we apply a psychological theory to other's actions and infer, on that theoretical basis, the reasons that motivate their actions. To the contrary, the simulation account, or "information-poor", holds that folk-psychology is essentially imitative and imaginative: we use ourselves as simulators of others agents' mind in order to gain information about their reasons to act. The debate was particularly vigorous in the 90's (philosophers, psychologists, primatologists and cognitive scientists participated) but seemed to vanish in the recent years. In his new book, Rediscovering Empathy: Agency, Folk Psychology, and the Human Sciences, Karsten Stueber is not only putting forth a welcome revival of the contention, but is also deepening it.

Beside the mechanistic questions of how we process information in order to explain and predict actions, the disagreement between both parties is also epistemological. Simulationists do not only argue that we use different skills, but also that there is something radically different in the way we understand other agents (vs. the way we understand the rest of the physical and world). When we interpret other persons, we conceive them as minded creatures that have a first-person perspective, just like we do. When we interact with, or think about, other physical objects, we don't use our imagination to simulate their subjective point of view, since we don't take them to have one. Thus, for the simulationists, agents and objects are structurally different. There are no common inferential mechanisms that apply to both.

In his book, Stueber makes a strong case in favor of simulation, or more generally, empathy. His thesis is that empathy has a central epistemic role: it is the default mode of interpersonal understanding. We first and foremost comprehend actions by putting ourselves in someone else's shoes, not by relying on a psychological theory of human cognition. Cognizing other agents is essentially an 'engaged' task, not a 'detached' one: we use ourself–our emotions, sensations, and thoughts–as mindreading tools, not an external device such as theory. Stueber compares mindreading and judging whether someone is the same height as yourself. You can either use an external, neutral standard–a measuring tape for instance–or use yourself as a standard: see if your head and hers is at same level. In this case, we use a subjective, non-neutral and egocentric point of view.

Stueber draws a distinction between what he calls basic and reenactive empathy. The first one is a quasi-perceptual mechanism (implemented in so-called 'mirror neurons'): when we see someone scared, we can easily sense her feeling. We see that someone is scared, but not why. Understanding why requires a more complex kind of empathy or reenactment. This second type of empathy, realized through a deliberative process, allows us to understand the reasons of actions. Since thought is essentially contextual and indexical, understanding someone else's thoughts requires that we see others' thoughts as thoughts that, had they been ours in the same context, would give use reason to act.  Thus it is by inner imitation that we really grasp others' intentions, not by theoretical deduction.

It is impossible, however, to do so without viewing each other as rational beings. Rational agency is a condition for reenactive empathy: when we take others to be normatively assessable, we can reconstruct the thought processes that govern their actions. By 'rational agency', Stueber does not imply that humans are good logicians or rational-choice theorists. A rational creature, he argues, is a creature whose assertions and actions are motivated by reasons, and whose reasons can be evaluated in the light of normative theories of rationality.  Thus, all empirical studies of showing that humans do not comply with logic and rational-choice theory do not undermine the role of rational agency in folk-psychological interpretation. Hence empathy is inherently based on a rationality assumption, what other philosophers such as Davidson called the principle of charity: interpreting others' beliefs as coherent. 

Having established empathy as the central mindreading device and rejected theory-theory and other detached accounts, Stueber goes on to claim that empathy also has a normative role: it justifies our beliefs-desires attributions. Here again, the author uses a vivid analogy. We can justify a prediction that it will rain tomorrow only by using an information-poor background: the barometer says so, and the barometer is a reliable tool.  Thus the prediction is inductively justified. Similarly, as long as we are in the domain of psychological interpretation, empathy is a reliable predictive tool that doesn't require a rich theoretical background. Of course, as many objected, empathy might be fallible, since it can be influenced by cultural and social background. Yet empathetic reenactement is still, Stueber contends, the principal mindreading strategy. It is fallible but it can be supplemented with auxiliary information.

Rediscovering Empathy is not just another book about folk-psychology. It is a systematic enquiry into the structure and function of mindreading that goes beyond the traditional exposition of recent cognitive theories. As Stueber shows, the debate between the engaged and detached conception of interpretation is not new and has roots in 19^th century discussions of hermeneutic Verstehen (understanding) and aesthetic Einfühlung (empathy). The nature and function of empathy is relevant for a diverse array of empirical and theoretical inquiries: beyond cognitive science and philosophy of mind, the debate concerning the nature of folk-psychological understanding impacts upon foundational debates in hermeneutics, aesthetics, anthropology, neuroscience, philosophy of language and philosophy of social science (mainly philosophy of history). Interpreting other agents, artifacts, texts, historical events or different cultures requires some kind of mechanisms that reliably indicates why individuals do what they do.

The greatest strength of this book is its ability to guide the reader through many important philosophical and scientific debates: the simulation vs. theory-theory debate, the rationality debate, the significance of mirror neurons and the nature of historical explanation, principally. In every case, the author uses an acute terminology and provides a clear presentation of competing theories, their empirical basis, their conceptual significance and their position in the history of thought. He exposes complex problems but never looses the reader (except in the last chapters, where the arguments is less clear). One notes also the unity and coherence of the book. The only problem with the book is that the author takes certain claims (e.g., the contextuality of thought) to be purely 'conceptual': he accepts them without much justification and thus it seems more dogmatic than conceptual.

This book will be of interest for any scholars interested in interpretation, generally speaking, but might be more accessible for philosophers of mind and social science. Cognitive scientists, social psychologists and social scientists will also found many discussions in the book relevant for their field.

Note: The introduction of the book can be freely downloaded on the publisher's website:

http://mitpress.mit.edu/books/chapters/026219550Xintro1.pdf

© 2007 Benoit Hardy-Vallée

Benoit Hardy-Vallée, PhD, Department of Philosophy, University of Toronto (Canada). The author is a philosopher of science interested in cognitive science, biology and economics.  His research focuses on the naturalization of rationality and the theoretical issues raised by empirical approaches of economic rationality.

Categories: Philosophical