Reading 2.2 – What I hear You Saying

Chapter 2 – It’s Just Verbal Behavior

Reading 2.2 – What I hear You Saying

One of the more exciting aspects of being a behavior analyst is our take on verbal behavior, widely viewed as the centerpiece of the human experience. It’s not just the scientific evidence supporting our views that bolster our enthusiasm – after all, we certainly need lots more science. It’s the conceptual implications of what our science says that seem especially satisfying. We have a way of talking about the most complex feature of human behavior that both respects the scientific evidence and avoids mentalism, and that’s not an easy task, as Skinner’s classic volume, Verbal behavior, demonstrates (Skinner, 1957). The price of avoiding mentalism is costly, however, given its pervasive role in everyday dialect. Mentalism is the tie that binds us together as a society, it seems, and considering human nature without respecting this apparent foundation is asking for trouble.

And yet, our ability to avoid mentalism in talking about verbal behavior is half the point and the source of our delight. For everyone else, what remains is a tortured rendition that loses all sense of what someone was trying to say. It is not that our professional dialect is any more obtuse than that of our colleagues in the natural sciences. A meteorologist can take the beauty out of a sunset just as well as we can dispense with the illusion of choice, but no one is likely to be offended by the meteorologist’s explanation. Insisting that words don’t have meaning, on the other hand, can get you attention on your Facebook page that you don’t want. The difference is that understanding a meteorological version of a sunset does not require people to abandon their personal reactions. In contrast, our technical dissection of a conversation seems to eliminate its everyday function.

What about content?

The function of any dialect is, in common parlance, to communicate. Scientific dialects, ours included, encourage a very particular style of communicating. Guided by Skinner’s model, behavior analysts focus on talking about verbal behavior in a way that emphasizes its operant features and that avoids distractions associated with vernacular dialect that are unavoidably the larger context for all scientific discussions. This is the kind of communication our science requires.

Everyday dialect, on the other hand, is all about meaning in the usual sense. The operative elements seem to be words, sentences, ideas, and so forth, all of which are said to convey meaning. It is relatively informal, vague, and convenient, though these characteristics come at the price of imprecision, misunderstanding, and confusion. Nevertheless, our vernacular dialect obviously works well enough, given that this repertoire has survived with only superficial changes over countless generations.

What makes every day dialect work is the shared history of our verbal community. We all grow up speaking this dialect. It is all we know, but we all know it well. We don’t notice its many shortcomings because we are used to overcoming the inevitable lack of clarity with more verbal behavior, which is a low effort solution. In other words, we just keep talking until we get the desired outcome. In the end, we understand each other, at least most of the time. We can talk about all kinds of things with reasonable confidence that others will respond appropriately.

And what do we talk about? Our verbal behavior seems to have content, which is another word for meaning. We talk about things and experiences as if our talking is aimed at something. We talk about what we did yesterday or are going to do tomorrow. We argue about issues of the day. We discuss ideas. In these traditional characterizations, we seem to talk about something in the sense that there is talking itself and, separate from that, what we are talking about. We view our verbal behavior as something we “use” to express ourselves.

As behavior analysts, we know how to avoid the implicit mentalism in these characterizations. In our model, our verbal behavior and its “content” are not separate. We do not “use” the former to communicate meaning. The meaning of what we say lies in a mutual history of each other’s reactions. The meaning of our words is not a distinct phenomenon that listeners translate or understand. What we are talking about is entirely in the nature of the talking itself, though that is devilishly hard for others to grasp.

If we observe that the sunset is beautiful and a listener responds in agreement or by commenting on its colors, we understand at a technical level what is going on in this interaction and can offer the operant details if anyone is interested. The speaker and listener are not likely to be on that list, however, because they are interested in the sunset, not their verbal behavior about it. In other words, when people are talking about stuff, they are focusing on the stuff. When we are talking about people talking about stuff, we are focusing on the talking because that is how we figure out the stuff being talked “about.” To behavior analysts, the stuff does not mean what the speakers/listeners think it means. To them, of course, it very much does.

In one sense, this disconnect between the technical level of our analysis of verbal behavior and the expansive character of everyday verbal interchanges is only apparent. Our analytical model is capable of addressing more than elementary verbal response classes or the challenge of teaching children with ASD to talk. The larger the sample of verbal behavior, however, the more challenging the application of our model. When the task is to “analyze” complex and extended samples – such as this essay, for example – we tend to abandon a focus on basic response classes and instead emphasize the nature of the more general environmental influences. This shift is pragmatic and fully consistent with our model.

A broader analytical focus

This distinction between everyday dialect and operant dialect about everyday dialect becomes personally important when we are interested in talking about ordinary things with ordinary folks, but as behavior analysts. What if we want to engage a friend or even a group of friends on a political topic, for example, but while also viewing the discussion from a professional perspective? How might we do that in a way that accommodates their understandable tendency to focus on the political issue of interest, while not giving away our rather odd perspective on what they are saying? Can we still talk about content without abandoning our behavior analytic sensibilities – or straining our friendships?

Accomplishing this requires balancing the need to speak in the vernacular dialect with our conflicting interest in considering the sources of control over what is being said. If we are to join a discussion of politics as just another participant, we must speak plain English, lest we be discovered. We are going to be talking about politics, not tacts, mands, and intraverbals. It’s not as if this level of analysis would be feasible anyway, given the pace of a discussion. Even if we could push a pause button and examine a transcript, analyzing discrete verbal samples in operant terms would be a relatively laborious exercise.

The challenge is to retain our focus on discerning meaning from contingencies by retreating to a more superficial level of analysis. Our goal should not be to assess the function of individual words or phrases, but to consider the sources of control over larger points, arguments, ideas, and so forth. In other words, we must focus on the same “stuff” that other participants are talking about. While doing so, we can consider the past and present contingencies influencing the verbal behavior of everyone else – although there is no need to leave our own influences out of the picture.

Naturally, such analyses are necessarily speculative, as they must be outside of controlled experimental conditions, but our guesses should at least be constrained by the scientifically established understanding of how behavior works. In this level of analysis, each point, statement, position, or argument offered by a speaker becomes a very loose kind of behavioral unit. Such “units” should be defined by conversational rather than scientific criteria. That is, we should be trying to understand the sources of control over the salient content in what others are saying – what they are talking about.

If our friend Mary is arguing that the country would be better off with a smaller and less intrusive government, for instance, we might focus our analysis on understanding the contingencies influencing her argument. Instead of taking her position at face value in a colloquial context, we might think about what variables in their history or in the present circumstance leads her to say this. What does she mean by “small government?” To what extent is what she is saying the result of what she learned from her parents, or her experiences as the owner of a business, or her social circle, or the TV channel she watches? Has she thought through her argument? Does she understand what this political view would mean if it were implemented? Is she just repeating what she has heard others say? How might what she is saying be influenced by immediate social variables?

Attempting to answer these and other questions helps us get at the meaning of Mary’s verbal behavior. Just as we examine the meaning of “happy” by looking for its social influences when a friend comments that Jules seems happy today (see Chapter 2), we can scale up this application of Skinner’s position on operationism to larger samples of verbal behavior defined in more traditional ways. Although this application is admittedly weaker than those anchored by elementary response classes, it retains the spirit of Skinner’s interpretation and perhaps a fair bit of its utility as well.

Attempting to understand the meaning of verbal behavior at a macro level by identifying possible influences can help us be more effective participants in these conversations. How we respond to Mary may be informed by our guesses about the influences over what she is saying. If it seems that she is parroting what she learned from her parents or from her favorite news outlet, we might respond simply by offering alternative positions – assuming we want to change her views. On the other hand, if we realize that she is well read on the matter and finds some of the ramifications of small government appealing, we might articulate some of the consequences that she might find less attractive.

Carrying on this kind of covert contingency analysis while participating in a discussion is not easy. It requires keeping sufficient intellectual and emotional distance from the “content” of everyday conversations to allow us to simultaneously consider the operant influences that might be at work. Our professional history encourages this observational tendency, and with a bit of practice it can be honed into a useful skill. Others might view this tactic as little more than being a good conversationalist, but they would fail to appreciate the unique understanding of verbal behavior that behavior analysts bring to the exercise.

It is easier to apply this tactic to written material because we don’t have the demands of participating in a live conversation while hiding our analytical tendencies and can go at our own pace. Whether reading a political rant or an academic treatise, we should be able to analyze the important features of its content from an operant perspective while not losing contact with that content. Our reactions should focus on its meaning in terms of its controlling variables. With this perspective, we might even develop illuminating observations about its content that would not have emerged by simply responding in a vernacular context.


Skinner, B. F. (1945). The operational analysis of psychological terms. In Psychological Review, 52, 270-277.

Skinner, B. F. (1957). Verbal behavior. New York: Appleton-Century-Crofts.