3.3.2: Empiricism leads to Logical Behaviorism

Following Hume in the 18th century, the philosophy of science takes a sharp empirical turn in the latter 19th and early 20th century. During this time, what is scientifically knowable is taken to be limited to what can be defined in observable terms. This puts the mind and psychological phenomena generally on epistemically shaky ground. Mental states like beliefs, desires, perceptions, and anxieties are not the sorts of things we can examine under a microscope. If all things knowable are supposed to be knowable through sense experience, then it begins to appear that minds and mental states are not knowable.

The philosophical behaviorism of Gilbert Ryle is an attempt to salvage talk of minds and mental states and make such talk empirically acceptable. Mental terms like belief or fear can often be associated with observable behavior. Anger and fear, for instance, often seem to be observable. Suppose we identified the mental state of being angry with displaying angry behavior. On this proposal, anger just is stomping around, cursing a lot or generally throwing a fit. The obvious problem here is that some people can be angry without displaying it and some people, good actors for instance, can engage in convincingly angry behavior even though they aren’t really angry. Or to take another example, my desire for chocolate ice cream might be observable in my rummaging around in the freezer, or it might not be observable at all because the usual behaviors are checked by my (also unobservable) desire to shed a few pounds. So mental states like anger or many beliefs and desires sometimes show in terms of behaviors, but perhaps only under the rights sorts of conditions. To make mental states empirically respectable and yet avoid the obvious problems we’ve seen in identifying mental states with observable behaviors, Ryle proposed to analyze mental states as dispositions to behave.

We are disposed in one way or another when we would behave a certain way given certain conditions. The behavior is not the disposition itself, but a manifestation of the disposition. The disposition can be identified in terms of a certain kind of “if. . .then. . .” statement. To help get clear on the idea, consider simpler physical dispositions like solubility or flexibility. To say that a spring is flexible is not to say that it is currently flexed. It is rather to say that if you were to stress it in the right way, then it would absorb the force placed on it and bend. To say that sugar is soluble is not to say that it is dissolved. But it is to say that if you were to submerge it in water (under the right conditions), then it would go into solution. So dispositions are described in terms of stimulating conditions and responses or manifestations. Ryle’s idea is that talk of mental states, like beliefs, desires, perceptions, or emotions can be fully explained as talk of very complex dispositions where the stimulating conditions and the manifestations are observable conditions and behaviors. So, my desire for chocolate ice cream might be understood as a complex disposition to exhibit behaviors like rummaging around in the freezer if I think I’ll find chocolate ice cream there, and I’m not too worried about my weight, and . . . . If this project works out, then we can understand talk of mental states in terms of empirically respectable stimulus-response dispositions.

The project of defining talk of mental states in terms of observable environmental stimuli and behavioral responses faces a number of difficult challenges though. We normally understand simple physical dispositions as being grounded in some further physical basis. Sugar is soluble in water because of its molecular structure, for instance. So we associate the disposition of solubility with a physical state of the sugar. In his eagerness to avoid positing unobservable mental states, Ryle wanted us to understand talk of dispositions merely as defining mental terms in terms of empirically respectable stimulus response “if. . then. . .” claims. He wanted to avoid positing any unobservable states of the brain, for instance, as the basis of mental dispositions. So Ryle’s talk of dispositions is limited to mere “if . . then . . .” without any appeal to underlying states of minds.

A second problem is that while we might be able to formulate plausible stimulus response conditionals for some mental state terms like fear or anger, in many cases the subtle links between stimulus and response that we might associate with a belief or a desire are simply too complex to allow for an analysis of the mental state talk in terms of observably defined disposition talk. What “if . . then . . .” claim, for instance, analyzes talk of my belief that my brother lives in Arizona?

A distinct problem, one that will continue to dog subsequent theories in the philosophy of mind, is the problem of conscious experience. However Ryle’s project worked out, we could imagine some kind of mindless robot that satisfies all of the relevant stimulus-response dispositions we associate with beliefs, desires, and emotions. And yet, by hypothesis, the mindless robot lacks any subjective conscious experience. When we think of our own case at least, our subjective conscious experience seems to be quite central to having a mind. This is an issue we will return to after a brief look at a few other 20th century approaches to understanding the mind.