Critter psychology: On the possibility of nonhuman animal folk psychology

Abstract

Humans have a folk psychology, without question. Paul Churchland used the term to describe "our commonsense conception of psychological phenomena" (Churchland 1981, p. 67), whatever that may be. When we ask the question whether animals have their own folk psychology, we're asking whether any other species has a commonsense conception of psychological phenomenon as well. Different versions of this question have been discussed over the past 25 years, but no clear answer has emerged. Perhaps one reason for this lack of progress is that we don't clearly understand the question. In asking whether animals have folk psychology, I hope to help clarify the concept of folk psychology itself, and in the process, to gain a greater understanding of the role of belief and desire attribution in human social interaction. To start, we can construct a simple argument in favor of animal folk psychology, based on a standard definition of the term. According to what I am calling the standard view, humans attribute specific mental states to a target, using a folk psychological theory, a mental simulation, or some combination of the two in order to generate predictions of intentional behavior. To count as a proper target for folk psychological analysis, the agent must engage in behavior that is predictable through the attribution of beliefs and desires. We see this commitment in Daniel Dennett's intentional stance, for example (Dennett 1987, 1991).1 That folk psychology centrally involves the attribution of beliefs and desires also seems to be endorsed by Alvin Goldman, who takes attribution of mental states via simulation to be the primary means for predicting behavior.2 These views reflect the standard notion of folk psychology, according to which intentional agents predict others' intentional behaviors via the attribution of mental states. Though one doesn't have to use belief/desire attribution in every instance of predicting behavior, to have a folk psychology one must be able to conceive of others as the sorts of things that have beliefs and desires, and be able to use specific mental state attributions to predict behavior; thus you have to conceive of others as intentional agents. On these views, folk psychology is not seen merely as a useful heuristic for making predictions, but rather it is thought that the primary function of folk psychology is the prediction of behavior.3 Any other role for folk psychology, such as the explanation of intentional behavior, is derivative of prediction. This assumption is clearly made in discussions of animal and child theory of mind, and in the debates between simulation and theoretical accounts of the subpersonal mechanisms driving our folk psychological behaviors (Andrews 2003). Dennett accepts this position when he writes, "-our power to interpret the actions of others depends on our power-to predict them" (Dennett 1991, 29). Folk psychology as the attribution of beliefs and desires is presented as the simplest predictive heuristic available for making accurate enough predictions across different domains. For this reason, advocates of the predictive power of folk psychology suggest that we use the attribution of beliefs and desires when we make all sorts of predictions, from the prediction that someone will duck if you throw a brick at him (Dennett 1991) to the prediction that you will arrive on the 3 p.m. flight if you say you'll arrive on the 3 p.m. flight (Fodor 1989). If there were quicker or easier heuristics we could use to make accurate predictions of behavior across domains, then there would be no reason to think that these examples are in fact examples of folk psychological prediction (rather than the result of using some other heuristic device). Since the standard view of folk psychology promised to help us understand the nature of beliefs and desires by presenting them as things that are used to make fast, easy, and accurate enough predictions of behavior, we can define folk psychology as how (non-Laplacean) intentional agents routinely predict the behavior of other intentional agents, namely through the attribution of beliefs and desires. The claim here is that without a folk psychology, predictions of behavior would not get made routinely, easily, and accurately (enough) across domains. In a world without folk psychology, we would not be able to predict that someone would duck if a brick were thrown at him, nor would we be able to predict that you will arrive on the 3 p.m. flight given that you said you will arrive on the 3 p.m. flight. Instead, everyone would be 'baffling ciphers.' Given such views, we may be tempted to think that though there may be more complex ways of predicting intentional behavior, such as the Laplacean super-physicist's method of following a deterministic causal chain, there are no more efficient and accessible heuristics available to us than the attribution of beliefs and desires. If so, then any non-Laplacean who routinely, easily, and accurately predicts behaviors across different domains has a folk psychology. This understanding of folk psychology allows us to construct a cheap argument for critter psychology: Argument C 1. Any (non-Laplacean) intentional agent who routinely, easily, and accurately predicts the behavior of other intentional agents has the ability to attribute beliefs and desires, and thus has a folk psychology. 2. Animals are (non-Laplacean) intentional agents who routinely, easily, and accurately predict the behavior of their conspecifics, competitors, predators, and prey. 3. From (1) and (2) it follows that animals attribute beliefs and desires, and thus have a folk psychology. Because the conclusion follows from the premises, and we can assume the truth of (2), any problem with this argument must rest with (1). The main problem with premise (1) is that we have good reason to think that there are methods other than the attribution of propositional attitudes for predicting behavior. Despite what others may suggest, even without appeal to people's beliefs or desires we can predict that someone will duck when a brick is thrown at him, because that's just one thing people generally do; they move to avoid large flying objects. And we can predict that you will arrive at the airport at 3 p.m., because you said you would and people generally do what they say they'll do. Without the ability to attribute propositional attitudes we probably wouldn't understand why people do what they say they'll do, but we could still make predictions by generalizing over past behavior. Since we can make predictions of intentional agents using such a method, perhaps the animals can as well. This, I think, is enough to establish that the attribution of beliefs and desires should not be described as the method we use to predict behavior. That is, though we may sometimes appeal to beliefs and desires when predicting behavior, predicting doesn't begin and end with the attribution of mental states. There may be other mechanisms that undergird our ability to anticipate behavior. This sort of view seems to be what Barbara Von Eckardt endorses, given her definition of minimal folk psychology as consisting of "(a) a set of attributive, explanatory, and predictive practices, and (b) a set of notions or concepts used in these practices" (Von Eckardt 1994, p. 300). Von Eckardt wants us to accept a wider conception of folk psychology that includes "any concept of generalization ordinary people use in their FP practices" (Von Eckardt 1994, p. 305), and her account leaves open the possibility that humans don't need to attribute mental states to make predictions of intentional behavior. I am sympathetic to Von Eckardt's account of folk psychology, and I think there is good reason to accept this critique of (1), for reasons I will present in Section 11.2. After we have in place a wider and, I believe, more satisfactory account of folk psychology, we can create a reconceptualized Argument C and examine what it might tell us about critter psychology. © 2007 Springer.