Flouting the convention of rational agency

Behaviorism as an ethnomethodological experiment: Flouting the convention of rational agency

Place, U T


ABSTRACT: As interpreted here, Garfinkel’s “ethnomethodological experiment” (1967) demonstrates the existence of a social convention by flouting it and observing the consternation and aversive consequences for the perpetrator which that provokes. I suggest that the hostility which behaviorism has provoked throughout its history is evidence that it flouts an important social convention, the convention that, whenever possible, human beings are treated as and must always give the appearance of being rational agents. For these purposes, a rational agent is someone whose behavior is controlled by a logically consistent body of means-end beliefs (“rules” in Skinner’s terminology) and complementing desires which between them provide a basis for predicting how the individual will behave and for suggesting what arguments will persuade the agent to modify his or her beliefs and the behavior based upon them. The behaviorist flouts this convention by suggesting that its fictional character makes it unsuitable for the purposes of scientific explanation of behavior. The hostility that this suggestion provokes is evidence of the importance attached by the verbal community both to preserving a consistent and rational connection between what is said and what is done and presenting it as part of the natural order of things.

The Ethnomethodological Experiment

Ethnomethodological sociology and behavior analysis have a number of things in common. One of the most striking is a shared suspicion of statistics and a preference for observing what happens in the single particular case. For behavior analysts, the case against statistics and in favor of the single-subject experimental design is stated by Murray Sidman (1960) in his classic Tactics of Scientific Research. To my knowledge there is no comparable text in the ethnomethodological literature arguing the case against statistical studies and in favor of observing and recording what happens in the particular case, despite the fact that that is the most distinctive feature of the ethnomethodological approach to sociological issues.

In a paper published in Philosophy of the Social Sciences in 1992, I argued that statistical studies of the incidence of a particular type of behavior within a social group, however accurate, cannot distinguish between a type of behavior such as sexual promiscuity, which has high natural probability of occurrence for biological reasons but whose incidence is reduced by the aversive social consequences of excessively conspicuous indulgence in it, and a type of behavior such as removing or putting on headgear or footwear when entering a place of worship, which has a low natural probability of occurrence as a biological response but whose incidence is vastly inflated by the aversive social consequences of the failure to conform to convention is this respect. Both types of behavior can yield the same statistical frequency despite the fact that the social contingencies are operating in the opposite direction in the two cases. In cases such as these, our contingency-shaped intuitions as members of the social group in question leave us in no doubt as to the direction in which behavior is being driven by the contingencies of social convention. But in other cases, the existence and nature of the social conventions which are constraining behavior may not be apparent, either because it is so familiar to a member of the group that the social forces maintaining it are ignored, or because, to an outsider, they are invisible. In these cases, and in any case where the evidence of intuition is challenged as unsatisfactory from a scientific perspective, the only way to demonstrate the nature and existence of the social contingencies that either enhance or reduce the incidence of a particular form of social behavior is to perform an ethnomethodological experiment.

As interpreted here, an ethnomethodological experiment takes as its starting point the observation of a form of human social behavior that is regularly emitted by members of a particular social group under particular circumstances, is peculiar to that social group, and does not readily lend itself to explanation as an innate biological propensity. From this observation we proceed to formulate the hypothesis that this regularity in behavior is sustained by a social norm or convention, in other words, it is maintained by the negative reinforcement provided by the non-occurrence of the aversive social consequences of failing to conform to it. Having formed this hypothesis the next step is to test it, either by flouting the supposed convention oneself or, as Garfinkel (1967) did, by persuading his students to do so. If the effect of so doing is to evoke from other members of the group verbal aggressive behavior of a kind designed to constitute an aversive consequence for the perpetrator of this misdemeanor, we have all the evidence we need to show that an important social convention has been transgressed.

Garfinkel (1967) describes his use of ethnomethodological experiment as follows:

Students were asked to spend from fifteen minutes to an hour in their [own] homes imagining that they were boarders and acting out this assumption. They were instructed to conduct themselves in a circumspect and polite fashion. They were to avoid getting personal, to use formal address, to speak only when spoken to.

Typical reactions to this behavior on the part of the student are described as follows:

Family members were stupefied. They vigorously sought to make the strange actions intelligible and to restore the situation to normal appearances. Reports were filled with accounts of astonishment, bewilderment, shock, anxiety, embarrassment, and anger, and with charges by various family members that the student was mean, inconsiderate, selfish, nasty or impolite. (Garfinkel, 1967, p. 47)

Not surprisingly, Garfinkel’s example in inducing his students to perform this kind of in vivo ethnomethodological experiment has not been widely followed by other sociologists. It survives, nevertheless, in the form of the ethnomethodological thought experiment. In an ethnomethodological thought experiment, readers or listeners are invited to imagine what would happen if or has happened in their own experience when such a convention is flouted, it is extensively used by sociologists in the ethnomethodological tradition as a way of drawing the reader or listener’s attention to the nature and existence of such conventions. This is well illustrated by the following quotation from a book by the Canadian feminist sociologist Dorothy Smith (1987). She writes:

When I take my dog for a walk in the morning, I observe a number of what we might call “conventions.” I myself walk on the sidewalk; I do not walk on the neighbor’s lawns. My dog, however, freely runs over the lawns. My dog also, if I am not careful, may shit on a neighbor’s lawn, and there are certainly some neighbors who do not like this [italics added]. (Smith, 1987, pp. 154-155)

Hostility to Behaviorism as Evidence That it Flouts an Important Social Convention

Having introduced the notion of an ethnomethodological experiment, I now want to suggest that behaviorism, or rather those forms of behaviorism which have proposed the replacement of what is called “mentalistic language” by some allegedly scientific alternative such as the language of behavior analysis, have inadvertently carried out just such an ethnomethodological experiment. By rejecting mentalistic language in the explanation of behavior, particularly human behavior, behaviorists have flouted an important social convention. The odium that these forms of behaviorism have incurred through their history is the consequence of its breaking this convention.

The Myth of the Rational Man

The convention that behaviorism flouts when its principles are applied to human behavior is what I call “the myth of the rational man.” I use the word “man” here advisedly, because it is part of the myth in its most typical form that it is restricted in its application to the behavior of the adult human male. The behavior of women is irrational, governed not by reason, but by what is known in the more complimentary versions of the myth as “female intuition.”

According to the myth, the behavior of the normal adult human male is invariably controlled by rational consideration of the probable consequences of doing one thing rather than another, an evaluation of those consequences in the light of the individual’s goals and objectives, and a balancing of outcome probabilities against value to be secured by adopting one alternative rather than another. In the light of these considerations the individual freely chooses to do what he does.

The doctrine of the freedom of the will is used to justify the demonstrably inefficient practice of trying to deter an individual from emitting behavior that is classified as “criminal” or otherwise blameworthy by inflicting aversive consequences months, if not years, after the emission of the behavior to which those consequences are supposed to relate. Only those whose “normal” powers of rational consideration and ability to exercise free choice have been undermined by some disease of the mind can be exempted from such penalties, if their behavior is found to have deviated from the norms in this way.

But while those who are judged to have lost their reason may be exempted from having to face the full rigors of the law, their loss of reason, even when it does not lead to any obviously anti-social behavior, is liable to make such people into social outcasts. In the past they ended up isolated from society in what have been variously referred to as madhouses, lunatic asylums, mental hospitals and now psychiatric clinics; but in these supposedly more enlightened days of so-called “community care,” most of them end up on skid row.

Two Truths That Underlie the Myth

Describing this theory of the behavior of the normal adult human male as a myth implies:

1. that it is false, and

2. that there are important social institutions such as those I have described which depend for their survival on most people accepting it as true.

What I am suggesting is that behaviorism incurs the hostility it does because it threatens to expose the myth for what it is and because the exposure of the myth threatens important social institutions. But what is it that is false about the myth of rational man and how, if it is false, does it come to be believed?

There are two important respects in which the myth follows what Skinner in The Behavior of Organisms calls “the natural lines of fracture along which behavior and environment actually break” (Skinner, 1938, p. 33). In the first place, developmental studies tell us that from a relatively early age, human children learn to solve problems by using verbal formulations of the contingencies involved, starting with formulae or “rules,” to use the term which Skinner introduced in his “An operant analysis of problem solving” (Skinner, 1966, 1969, 1988), which they have learned from others, but later adding their own rules based on their own experience of the contingencies. Very soon this becomes the invariably preferred human response to any problem situation so much that those whose assessment of human behavior derives from what they observe in the psychological laboratory, are sometimes tempted to conclude (Wearden, 1987) that all human behavior is rule-governed, governed, that is, by verbally formulated hypotheses as to the contingencies involved. This is because human subjects, once they emerge from infancy, invariably treat a psychological experiment as a challenge to work out what the experimenter is expecting them to do, that is, what behavior he or she is going to reinforce.

The other respect in which the myth corresponds to reality is in so far as all normal human adults are constrained by social convention to preserve at least the appearance of a consistent and rational connection between what they say on the one hand and what they do, how they otherwise behave, on the other.

Where the Myth Becomes Mythical

The myth becomes mythical:

1. in so far as it maintains that the rules or verbal formulation of the contingencies involved continue to control the behavior, once a solution to the initial problem has been found, and

2. in so far as it implies that such rule-governance of behavior is part of the natural order of things.

What tells the organism that a problem has been solved is that the predicted or anticipated consequences of behavior begin to coincide with the consequences that are actually experienced. As soon as that happens, the actual consequences of behavior begin to take over from the verbal formulations or other behavior involved in seeking a problem solution. The behavior becomes contingencyshaped. We act no longer out of rational consideration of the consequences, but from force of habit.

Because rational considerations were implicated in the initiation of the behavior in the first place and because getting someone to accept a different way of formulating the contingencies involved can in most cases induce them eventually to break even the most deeply entrenched of contingency-shaped habits, it is usually possible to predict what someone will do by observing:

1. what they say,

2. what they have observed, and

3. what they have been told about that situation. Such predictions are based on an inference as to what the individuals in question would have said to themselves in deciding what to do when confronted with that situation for the first time.

So important is the ability to predict how people will behave from a consideration of how, given the information available to them, they would talk to themselves in deciding what to do in that situation, that there are powerful social sanctions at work to ensure that a consistent and rational connection is maintained at all times between what people say and what they otherwise do. The aversive social consequences of deviating in this respect is a lesson that we learn from some of our earliest interactions with the peer group. That, after all, is what being denounced as “stupid” or “crazy” is all about.


If it takes that kind of social pressure to ensure that our behavior gives at least the appearance of being rationally controlled, what more convincing evidence do we need to show that it is the actual contingencies of reinforcement and punishment, not the way those contingencies are formulated in words, that are the ultimate determinants of human behavior, just as they are of the behavior of prelinguistic organisms.

I Paper presented at the Annual Convention of the Association for Behavior Analysis, Washington, DC, May 30, 1995.


Garfinkel, H. (1967). Studies in ethnomethodology. Englewood Cliffs, NJ: Prentice Hall.

Place, U. T. (1992). The role of the ethnomethodological experiment in the empirical investigation of social norms and the application to conceptual analysis. Philosophy of the Social Sciences, 22, 461-474.

Sidman, M. (1960). Tactics ofscientific research. New York: Basic Books.

Skinner, B. F. (1938). The behavior of organisms. New York: Appleton-Century-Crofts.

Skinner, B. F. (1966). An operant analysis of problem solving. In B. Kleinmuntz (Ed.), Problem solving: Research, method and theory (p. 225-257). New York: John Wiley.

Skinner, B. F. (1969). Contingencies ofreinforcement: A theoretical analysis. New York: Appleton-Century-Crofts.

Skinner, B. F. (1988). An operant analysis of problem solving. In A. C. Catania & S. Harnad (Eds.), The selection of behavior. The operant behaviorism ofB. F. Skinner: Comments and consequences (pp. 218-236). Cambridge: Cambridge University Press.

Smith, D. E. (1987). The everyday world as problematic: A feminist sociology. Boston: Northeastern University Press.

Wearden, J. (1987, April). Rules and rule-governed behaviour. Paper presented at the meeting of the Annual Conference of the Experimental Analysis of Behaviour Group, University of Manchester, England.

U. T. Place

University of Wales, Bangor

Copyright Cambridge Center for Behavioral Studies Spring 2000

Provided by ProQuest Information and Learning Company. All rights Reserved