You Say Potato, I say Solanum Tuberosum (Let's Call the Whole Thing Something Else)

In my junior year at the University of South Florida, after taking all the prerequisite psychology courses, including statistics and research methods, I signed up for psychopathology. Finally, I would learn the mysteries of neuroses and psychoses and all the other arcane elements of my major subject. In opening the first class, my professor (Jack Sandler, who would become a huge influence in my life) said, "We won't be talking about neurosis or psychosis in this course. We'll be talking about operant learning and human behavior." What? But I took stats and methods!

I was outraged, but twenty minutes into the first lecture the sky opened. The scales fell from my eyes. I was a born-again behaviorist. Thorndike, Watson, Pavlov, and Skinner had brought us the law, and this man was going to teach it to me. Other prophets would follow: Sidman, Azrin, Wolpe, Baer, Wolf, Risley, Greenspoon, Kazdin, Bandura, Bijou, Lovaas.

The next two decades represented near total operant immersion as I worked in behaviorally based treatment settings serving children, youth, and adults with developmental and psychiatric disabilities. I loved the clarity of the behavioral approach. I loved its empiricism. I loved its commitment to addressing behavior without recourse to diagnostic name-calling. And boy did I love the terminology. Free operant rate, decreasing interlocking schedule, positive celeration, variable ratio, establishing operation, signaled avoidance, stimulus discrimination, backward chaining, ratio strain, negative reinforcement, stimulus delta, positive punishment, extinction burst. Medicine may be Queen of the Jargon Pageant, but Operant Psychology is First Runner Up.

Although my earlier training still informs my thinking in a profound way, I no longer adopt a rigidly operant frame, and unless I'm talking with another behavior analyst, I don't even use this kind of language. Partly because I think operant learning—the way we typically understand it—accounts for only a certain range of human behavior, but also because this sort of jargon is alienating and I believe promotes a kind of arrogance and insensitivity in the user.

This language can also be very confusing and problematic for people providing and receiving services in hospitals, group homes, clinics, schools, and other settings. Some words that have very specific technical meanings in the laboratory have been borrowed from our common language and have very different meanings in everyday speech. Behavior analysts themselves sometimes add to the confusion by using technical terms somewhat carelessly. So you'll hear things like "non-contingent reinforcement" or "placing a person on extinction."

Non-contingent reinforcement is oxymoronic for the same reason that contingent reinforcement is tautological; the contingent nature of reinforcement is implicit in its definition. Yet even behavior analytic luminaries use this language in peer-refereed journals. The usage isn't technically correct, but we know what they mean. They are referring to situations in which reinforcement is mediated or unmediated by an experimenter or in which reinforcement was programmed or adventitious. They're speaking a kind of shorthand, the same kind marine biologists might use when they say that a fish evolved certain colors as protection from predators. Fish didn't evolve colors as anything. Evolution isn't directed by intentionality, but we understand what is meant (although this usage presents its own set of problems—sit in on a local school board meeting and see what I mean).

"Reward" and reinforcement are related words with slightly different meanings. "Reward" is colloquial; reinforcement is technical. A "reward" is recompense for good behavior. A reinforcer is a stimulus event that follows a behavior and increases the probability of the behavior occurring again. Often, though not always, rewards function as reinforcers. Put simply, reinforcers are rewards that work. It is fairly easy to understand how these terms differ from each other, and if confusion arises it is easy to resolve. This is helped immensely by "reward" and reinforcer being two completely separate words. That is not always the case with behavioral terminology.

One example—the one that causes the most distress and ill will—is punishment. In everyday speech, punishment is understood as penalty for wrongdoing. Synonyms for "punish" include penalize, censure, fine, chastise, castigate, berate, scold, rebuke, rebuff, and retaliate. This is the way most people understand the word. Because we wish to provide services and supports that respect individuality, preserve dignity, seek understanding, and support freedom and choice, we shun these notions and renounce their use.

In behavior analytic settings punishment doesn't have the emotional connotations it holds in a clinic, developmental center, or school. It is simply a stimulus that when presented or removed following a behavior, reduces the probability of that behavior's subsequent occurrence. Now you can insist till the cows come home* that punishment doesn't technically mean "punishment", but misunderstanding and conflict will often arise because, unlike "reward" and reinforcement, here the colloquial word and the technical word are THE SAME WORD. Consequently, use of the term punishment in therapeutic or educational settings is alienating, confusing, and yes, punishing.

But if, as we say, we are committed to exclusively positive, collaborative, and respectful methods of behavior support, why use procedures that suppress behavior at all? Couldn't this entire semantic controversy be rendered irrelevant by simply eliminating all suppressive strategies from our practice? Can that be done?

I've seen severely dangerous and puzzling behaviors of longstanding duration give up their mysteries to reasoned, persistent investigation. I've seen patient, sensitive, committed support professionals use encouraging, strength-based, and quality of life enhancing strategies to help people discontinue very disturbing behavior and make huge strides in positive new directions. But I've also seen catastrophic self-injury (deep tissue laceration, nerve and vascular damage, bone fracture, retinal detachment, etc.) and interpersonal violence that, despite comprehensive functional assessments and genuine and sustained efforts at positive approaches, remained intractable.

The vast majority of people with dangerous behaviors fall into the first
group. People in the second group are very rare. Most of us will go our
entire professional lives and never meet one. The trouble is, it is difficult to differentiate the first group from the second without effort, and far too few of us are willing to go to the lengths required to tell them apart. Out of laziness or impatience, or our own inflexibility, we often assume that the majority of people fall into the second group, but this is manifestly wrong.

Even among the second group—the people with seemingly intractable behavior—I believe, given time and motivated helpers, their behavior can be understood and supports designed that are built upon that understanding. But sometimes during the time required to reach such an understanding, vision loss, brain damage, and other irreversible injury can occur. That's why I concede that in rare instances there could be a time and circumstance where suppression or disruption of life-threatening behavior may be required. Sometimes an extremely dangerous behavior may need to be suppressed in order for any of the other supports we are employing to take effect. Each person's situation must be addressed individually and not considered as an abstraction. It's easy for us to take expedient political positions when we are not immediately facing such problems ourselves.

In order to use suppressive strategies in a considered way, with wisdom and humility, our motives and rationales must be clear to the person, to the person's family, to each other, and finally to ourselves. Because the language we use to describe such strategies is also the language we use to think them out and think them through, our language should be very clear. That's why I strongly suggest that we discontinue use of the word punishment in service and support settings.

Since the goal of a punishment procedure is suppression of a particular behavior, let's call such procedures suppression. I know how radical this sounds. We have an experimental-pedagogical-clinical tradition several decades old that uses the term punishment to describe a procedure that suppresses unwanted behavior, but it came into widespread use in service settings during the late 1960s and early 1970s, a time when disability advocacy was in its infancy and today's celestial chorus of protest was unimagined. I think it's time our language caught up to our consciousness.

Replacing punishment with suppression does a couple of things. One, calling it suppression describes functionally what the procedure does without contaminating it with the emotionally charged associations of "punishment." Two, it separates the terms semantically, which aids in understanding. The basic operant concept of punishment is already complicated, start talking about positive and negative punishment and people stare at you like you're a Scientologist.

In providing training and technical assistance in educational, mental health, and developmental settings across the country, I've found people very accepting of this idea, but it's the kind of argument that has prompted one critic to label what I teach "Fisher Price behaviorism." That was meant as a dig, but I'll accept it as a compliment. Fisher Price produces high quality, durable products that are safe, colorful, entertaining, and promote growth. If that's the worst thing you can say about what I do, I guess I'm pretty happy.

*A very long time—Apparently cows never hurry home. Why, I'm not sure.

-December 2005