Mind as a Causal Structure: Causal-Theoretical Functionalism

Philosophy of Mind and Metaphysics Lecture XII: Mind as a Causal Structure: Causal-Theoretical Functionalism I. CAUSAL-THEORETICAL FUNCTIONALISM a. According to CAUSAL-THEORETICAL FUNCTIONALISM, a mental state is an internal state that serves as a causal intermediary between (a) causes, which consist of sensory inputs of a certain sort and mental states, and (b) effects, which consist of behavioral outputs (of a certain sort) and mental states. b. “What, according to functionalism, distinguishes one mental kind (say, pain) from another (say, itch) is the distinctive input-output relationship associated with each kind. Causal-theoretical functionalism conceives of this input-output relationship as a causal relation, one that is mediated by mental events. Different mental events are different because they are implicated in different input-output relationships. Pain differs from itch in that each has its own distinctive causal role” (p. 104). c. When we define mental states in terms of other mental states, as we have in I.a, doesn’t circularity threaten? To circumvent the threatened circularity, “causal-theoretical functionalism attempts to use the entire network of causal relations involving all psychological states—in effect, a comprehensive psychological theory—to anchor the physical/behavioral definitions of individual mental properties” (p. 105). i. Begin with an entire—and, we hope, true—psychological theory, one that defines psychological concepts en masse. Consider the following portion of such a theory: There are certain internal states M1, M2, and M3 that are related to each other in a certain way and also to physical/behavioral states like tissue damage, wincing, groaning and yelping. It’s clear that this defines certain concepts—e.g., the concept of being in internal state M2—in terms of their causal/nomological relations. Moreover, among the causes and effects of M2, in addition to physical and behavioral events and states, are other mental states (e.g., M1 and M3). However, we have not specified those states as mental states but only as some states of the psychological subject. Notice also that there is a sense in which the three mental concepts—namely, M1, M2, and M3—are interdefined but 1 without circularity; all the defined expressions—e.g., “is in pain”—are completely eliminable. ii. We can then select a certain internal state to which the theory refers—for example, M2—and see how it is related to sensory inputs, physical/behavioral outputs, and other internal states. iii. If M2 is related to those things as pain is, then we will have selected the internal state that plays the distinctive causal role of pain. That is, we will have selected pain. d. WHICH THEORY SHOULD WE CHOOSE? i. SCIENTIFIC PSYCHOLOGY: We might choose our best scientific psychological theory. 1. PROBLEM: Psychology is a comparatively young science, with its methodological foundations very much in dispute, and it is fair to say that psychology has yet to produce any theory that we can agree to call our best scientific psychological theory. Psychology “has yet to produce a robust enough common core of generally accepted laws and theories” (p. 109). ii. COMMONSENSE PSYCHOLOGY: We might build a psychological theory out of our shared commonsense psychological platitudes. 1. ADVANTAGE: Commonsense psychology seems to be more stable than scientific psychology. 2. PROBLEM: Is commonsense psychology true? II. PROBLEMS WITH FUNCTIONALISM a. QUALIA i. Qualia are the characteristic qualitative features of sensations, e.g., the way a rose smells, or the way a ripe tomato looks. ii. “But by identifying sensory events with causal roles, functionalism appears to miss their qualitative aspects altogether” (p. 114). For a state can play pain’s causal role, for example, even in the absence of the qualitative aspects that are distinctive of pain. Yet we tend to think that those qualitative aspects are not only distinctive, but also definitive of pain. Thus, functionalism cannot tell us what pain is since it says nothing at all about its distinctive qualitative aspects. 2 Certain thought experiments are supposed to show that a state can play a certain causal role even in the absence of the appropriate and distinctive qualitative feel. 1. The following seems perfectly conceivable: “When you look at a ripe tomato, your color experience is like the color I experience when I look at a bunch of spinach, or vice versa. That is, your experience of red might be qualitatively like my experience of green, and your experience of green is like my experience of red” (p. 114). Nevertheless, you and I might be functionally equivalent. Call this the inverted qualia problem. 2. “[W]e can imagine a system, like an electromechanical robot, that is functionally—that is, in terms of inputs and outputs— equivalent to us but to which we have no good reason to attribute any qualitative experiences. This is called the “absent qualia problem”” (p. 114). b. THE CAUSAL EFFICACY OF MULTIPLY REALIZED PROPERTIES i. We take it as a fact that instantiating a mental property can, and does, cause other events to occur. ii. The causal powers of a mental property, M, arise out of the causal powers of the physical properties that realize M. (For example, the causal powers of pain arise out of the causal powers of C-fiber activation.) iii. But given that M can be multiply realized, it is not possible to associate a unique set of causal powers with M. (Each instance of M is an instance of some physical realizer, P, and each different physical realizer of M will be associated with a different set of causal powers. From this it follows that we cannot associate a unique set of causal powers with M.) iv. Moreover, kinds in science are individuated on the basis of causal powers – and to be recognized as a useful property in a scientific theory, a property must possess (or be) a determinate set of causal powers. v. Thus, M is not a scientific kind – and the property of being in M cannot be a useful property in a scientific theory. 3