Giving Up Freedom and Dignity: The Psychology of Burrhus Frederick Skinner
This column we will be examining a psychological school of thought known as behaviorism. This is the approach that dominates most U.S. university psychology departments. One of its major proponents, theorists, and popularizers is B. F. Skinner, a would-be novelist, turned hard-nosed, scientific experimenter.
The roots of behaviorism go back to V. M. Bechterev and I.P. Pavlov in Russia, plus John B. Watson, Clark Hull, John Dollard, Neil Miller, and—of course—B. F. Skinner in America.
One of behaviorism’s major emphases consists of dealing ONLY with observable, measurable behavior, which is possible to define operationally. Any reference to, or consideration of, inner states, ideas, feelings, is considered totally unnecessary. They assume “no mind as such, but only body; reality is flesh and blood, rather than mind and soul.” (p. 177 Corsini) Some opponents have referred to behaviorism as the “psychology of the empty organism.” (p. 177, Corsini)
Behaviorists emphasize an experimental approach. They gather data, seek to control variables and express results in quantitative terms if possible. The focus is on the responses which an organism makes to stimuli in its external environment. This is referred to as S-R (or stimulus-response) theory. Skinner and others assume ALL behavior is explainable in terms of external stimuli. Skinner has an extremely deterministic view of human behavior. He denies free will. He sees people as controlling their fate ONLY to the extent they manipulate the external stimuli which control them.
Skinner does allow a place for heredity. He sees it as setting limits of possible behaviors. (Here, at least, he goes further than Watson. Watson is often called the father of behaviorism. He led the way in suggesting psychologists study behavior and not mental events. Watson, however, more or less denied the importance of heredity. He declared at one point that if given a healthy infant at birth, he would make that infant ANYTHING—doctor, lawyer, beggar, Indian chief, etc.) Skinner also considers biological urges.
Current behaviorists are more sophisticated than Watson, but Skinner’s goals, just like Watson, are that psychology should predict and control human behavior. So, he and his ilk set up experiments. They provide a stimulus and observe the response. They provide another stimulus, and observe the response, etc. Behaviorists assume ALL behavior is the result of antecedent conditions. The “appearance” of free will, according to them, is merely when we do not know or control all the relevant variables.
Much early behaviorist research was done on lower animals, e.g. rats and pigeons. However, more recently, considerable work has been done with people. The effectiveness of behaviorist principles can hardly be argued. They work often and well. Behavior Mod by Philip J. Hilts gives a popularized summary of some behaviorist “success stories” along with his critique. (I will address philosophical issues later.)
Skinner divides behavior into “respondent” and “operant.” In the first, the environment does something to the organism, which then responds. Pavlov’s pioneering work with dogs used “respondent” behaviors. Dogs salivate when presented with food. That is an “unconditioned” response. (They do not need to learn to do it.) Pavlov rang a bell along with presenting the food. That became a “conditioned stimulus.” The dogs learned to associate the bell with the food. Eventually, they salivated (responded) to the bell alone, even without the food.
Operant behavior is when an individual “operates on” the environment. Reading this article is operant behavior. The vast majority of human behavior is operant rather than respondent.
Skinner’s next step is the principle of reinforcement. As implied, this means strengthening. Reinforcement is defined as strengthening behavior, or the probability of a behavior occurring is increased when reinforcement is provided. Thus, personality is assumed to be acquired and maintained through reinforcement (learning).
Reinforcers may be positive or negative. Positive reinforcement increases the probability of a behavior with its presence. Negative reinforcement strengthens behavior in its absence or removal. Reinforcers can also be primary or conditioned. Primary positive reinforcement includes water, food, sex or stimuli which meet a biological need. Primary negative reinforcers include avoidance of getting too wet, too cold, too hot, etc.
Conditioned reinforcers include money, prizes, etc. Behaviorists also view praise, attention and approval as conditioned reinforcers. Thus, they assume people LEARN to be reinforced by praise, approval, etc. They assume it begins when attention or praise is associated with a primary (biological) reinforcer such as food. Eventually, the individual is reinforced by the praise, attention, approval, etc. without the food. (This seems a questionable assumption. All kinds of stimulation, including attention, praise, seem reinforcing to infants.)
Withholding reinforcement can weaken or alter behavior. “Extinction” is the term for eliminating a previously conditioned behavior by removing the reinforcement. E.g. A child gets parental attention by throwing temper tantrums. The parents begin ignoring the tantrums. Eventually, the tantrums cease.
This is a good example to illustrate some criticisms of the techniques of behaviorism. A conscientious parent would first ascertain that there is no physical problem or reason for the tantrums. This is not always done. Symptoms may be removed which should have been recognized as a warning signal. Even if no physical difficulty exists, there may be ramifications to the tantrums. Perhaps the child is not just seeking attention. Perhaps there is a fear component (of the dark, of being alone, etc.) Perhaps anger is a part of the issue. Etc. As critics have pointed out, behaviorists sometimes treat the symptom and miss underlying issues. So a new, perhaps more serious, symptom arises. If they continue to treat the symptoms, the end result may not be helpful.
There is yet another issue inherent in this example. Who decides to get rid of the temper tantrums? The parents do. I would be the first to admit to not wanting a screaming, kicking child on my hands, but the issue of power is an important, ethical question. Behavior modification does work. Its principles have been shown effective in many settings. A lot of the human settings have included school (especially for “problem” children, i.e. handicapped, retarded, delinquent, etc.), prisons, and mental institutions. In early years, such places provided convenient populations (captive audiences). Those in power were willing to have behavior-shaping principles tried out on the powerless. The results have been largely successful.
Children have been conditioned to attentive, quiet behavior in school. Patients have learned to make their beds, clean up their living areas, stand neatly in line to the dining areas, etc. Some homosexuals have been “cured” of their sexual preference through electric shock or aversion therapy using drugs with unpleasant side effects. Prisoners have been frightened from crime by being strapped down and injected with a drug which makes them feel like they are dying; they cannot breathe. Government (public) money has trained porpoises to kill frogmen and private soldiers to kill less reluctantly. (See Philip Hilts’ Behavior Mod.)
Does a quiet child who is looking at a book learn more? Are we training passive, compliant children who automatically APPEAR attentive, even if learning nothing? Are neat, orderly patients more able to cope with the outside world and graduate from the hospital? Or do they merely ease the custodial burden on the staff? Are frightened criminals less crime-prone? Or will the utter terror they underwent elicit tremendous rage and frustration and greater violence? These and other questions remain. Carl Rogers has said: “Behavioral scientists, holding their present attitudes, will be in the position of the German rocket scientists specializing in guided missiles. ... If behavioral scientists are concerned solely with advancing their science, it seems most probable that they will serve the purpose of whatever group has the power.” (p. 148, Hilts)
Skinner was, and is, an idealist. In 1948, he published Walden Two, a utopian novel about the positive potentials of a community based on behavioral principles. But our society is not a utopia. Even the community (Twin Oaks) founded in Virginia in an attempt to build a Walden Two has been able to incorporate a very limited number of Skinnerian ideas.
By ignoring what behavior MEANS to an individual, behaviorists risk being only symptom-alterers. By using the definitions of the powers-that-be in terms of behaviors to be changed, behaviorists risk at best keeping the status quo, and at worst entrenching tyranny and the abuse of power.
To return to our discussion: behaviorists tell us personality develops through generalization, discrimination and modeling. Generalization is learning about classes, groups of similar items, e.g. “These are all cats.” Discrimination is learning to make distinctions (through differential reinforcement). “This is OUR cat; that is their cat.” It is assumed children (and adults) generalize first, then learn to discriminate. Modeling means learning by watching and imitating others. (With practice, our imitation gets closer and closer to the examples we observe.)
Skinner has emphasized that punishment is less effective than positive reinforcement. If it is continuous, punishment will often suppress the undesired behavior. As soon as the punishment is removed, however, the undesired behavior is often back at full strength. Conversely, intermittent positive reinforcement (e.g. rewards which are NOT continuous) tends to be highly effective. Punishment has the additional disadvantages in that it may disrupt social relationships, create difficulties for the punisher, or the aggression felt by the punished may be displaced and directed against people who were not responsible for the punishment.
In behavioral terms, motivation is either deprivation or satiation of the organism in regard to a reinforcer. E.g., food has little reinforcement value to a full person, but high reinforcement potential to a starving one. Chemical control can also be used to affect behavior (through drugs, surgery, etc.) Simple fatigue can alter the likelihood of certain responses. Behaviorists are strongly involved in current psychosurgery research. There are many ethical questions about who is subjected to mind and behavior-altering drugs and surgeries, and why. Who decides what is appropriate? (Not so many years ago, frontal lobotomies were hailed as a wonderful breakthrough in treating insanity!)
In summary, behavioral techniques have often been successful. I have no quarrel with the efficacy of the techniques. “Behavior Mod” can indeed help people stop smoking, lose weight, and a host of other wonderful results. I have strong questions with the use of some behavioral techniques. I have philosophical differences with Skinner and his ilk. Skinner is a die-hard determinist. He asserts that people are not autonomous, but “always under the control of variables outside” themselves. (p. 193, Corsini)
One of the difficulties with disagreeing with Skinner on his home ground is the chicken-and-egg problem. Behaviorists can always say that what APPEARS to be free will is merely their insufficient data about antecedent decisions. I am sure Skinner would tell me my defense of free will is predictable on the basis of my upbringing and early conditioning.
Logically, one of the few places available to question behaviorist philosophy, then, is at the very start of life. Behaviorism does assume tabula rasa—the blank slate at birth (within parameters set by heredity). Observers of newborns could certainly disagree with that. Differences appear from the start. Perhaps behaviorists would extend the environment to the intra-uterine and explain differences at birth to physiology not just of the baby, but also of the mother, hormonal influences before birth, etc. It rapidly becomes difficult to argue if one sticks with their definitions. I will, however, mention a few concepts that, while they cannot “prove” anything in an empirical sense, cast some doubts in my mind on the validity of the behavioristic PHILOSOPHY.
(1) Human behavior is highly complex, and individualized. Yet some people exhibit great skill in complicated talents very young (E.g. Mozart). How could they learn so quickly?
(2) We are realizing more and more that the brain is a FILTER. We SELECT, out of a multitude of perceptions and sensations, which ones to attend. If we were all blank slates to begin with, how did we start filtering? On what basis did we choose to attend? The behaviorist would say we attended to what was reinforced. Does that really cover the complexity of our countless perceptual decisions to be made daily? Do not some people seem to “choose” to attend to the unexpected. (Examples include the exceptions, e.g. the ghetto child that does NOT turn to the life of crime or victimization, but escapes. Usually such a person will cite some inspirational role model, e.g. a teacher, leader, parent, etc. They say, “Despite all the negativity I saw every day, almost all the time, I listened to this person, and believed I could do it, so I did.” Is that child really explained in terms of “conditioning” alone?!)
(3) Stanley Schachter and others have demonstrated that many emotional responses are identical physiologically, but given different labels (fear, sorrow, excitement, etc.) by individuals, depending on the social context and individual interpretations. Knowing this, we can CHOOSE to respond one way or another. (Behaviorists deny personal choice.) Does anyone believe such a choice would be determined ONLY by his/her past conditioning and present pressures in the environment?
(4) Behaviorists seem to ignore the future except in avoidance due to anxiety. Future plans, hopes, wishes are motivators for me, and certainly “inner states” to be considered!
(5) Behaviorists ignore all information from psychic research, or assume they will eventually find a material cause for ESP. They also ignore studies supporting reincarnation (which would blow away the blank slate assumption).
(6) Julian Rotter has experimented with people who have a more internal versus external locus of control. Basically, that is the feeling of: “I’m in charge of my life.” vs. “The universe is in charge.” People with an internal locus of control contradict behaviorism’s “normal” results in terms of reinforcement. Over and over again, behaviorists find intermittent reinforcement more effective than continuous. (When continuous reinforcement is stopped, even for a short time, the behavior stops quickly. When intermittent reinforcement is stopped, the behavior will continue for considerably longer.)
In Rotter’s work, those people having an internal locus of control took much longer to give up their behavior after continuous (100%) reinforcement than intermittent (50%) reinforcement. This is an exact reverse of behaviorists’ “laws of behavior.”
Subjects with an external locus of control (feeling the situation was experimenter-controlled or due to luck or chance) had lower expectations, even after success, then did subjects with an internal locus of control (who felt the situation was due to their skill or ability). Similarly, the subjects feeling personal responsibility had lower expectations after failures than did those ascribing the results to luck, chance or the experimenter. Subjects who perceived the results as due to outside factors were more likely to raise their expectancies after failure and lower them after success! Subjects who felt personally responsible seemed to follow a sort of common-sense logic: “I did it before; I can do it again.” Or, “I didn’t get it before; maybe I don’t have the necessary skills.” This trend carried over to similar tasks as well. That is, those people feeling personally in control expected to do about as well or poorly on a similar task as they did in an original situation. Those feeling the control was outside were less likely to assume any relationship between their first task and subsequent, similar ones (because they felt it was all pretty random).
Furthermore, Rotter tells us: “...the individual who has a strong belief that <s>he can control his <her> own destiny is likely to (a) be more alert to those aspects of the environment which provide useful information for his <her> future behavior, (b) take steps to improve his <her> environmental conditions, (c) place greater value on skill or achievement reinforcements, and be generally more concerned with his <her> ability, particularly his <her> failures and (d) be resistive to subtle attempts to influence him <her>.” (p. 30, Dobyns)
In my reading of Skinner, the concept of an innate drive toward self-actualization (as conceptualized by Maslow and others) would be explained as a “learned” response. (Presumably those fortunate self- actualizing people early on got primary reinforcers linked with attitudes and actions of a self-actualizing nature.) Even if Skinner were right, consider the options.
Self-actualizing people are happier, better adjusted, contribute more to society and their fellow humans. As I have seen too often in astrology, determinists are many times looking for an excuse. They want a reason, an explanation for not having or being or doing all that they want or could. So, such people eagerly attach to “The planets are doing it;” (or “Mother did it,” or “The culture did it.”) or any handy cop-out.
Clearly, nothing is one-sided. I do believe much of reinforcement theory works. I believe societal pressures and cultural discrimination exists. I believe parents influence children AND children influence parents.
I worry about people who are eager to give power and responsibility over their lives to someone or something else, in exchange for security and quick, easy answers. They are sheep and easily led. The more sheep we train, the more precarious our freedoms become.
The data suggests people who believe in their own power are more powerful. Rosenthal’s self-fulfilling prophecy works. In the end, it does not even matter if Skinner is correct. What matters is what we believe. I CHOOSE to believe that I make choices. The data suggests that as long as I operate in that framework, I EXPERIENCE free-will, decision-making, and power. That, for me, is preferable. It works. Skinner is welcome to dis-own his own transcendence, to damn himself as being only a consequence of what he learned from external stimuli. Trusting inner transcendence uplifts our reach, thrust and experience of life. Skinner has mankind on a straight-line extrapolation, with everything neatly tied down. I’ve seen curves, and leaps and gaps in that line. The infinite side to human nature goes far beyond past experiences, stimuli and conditioning, just as a piece of music, performed with dedication and feeling, goes far beyond black notes in bars on white paper.
Corsini, Raymond J., Editor, Current Personality Theories, Itasca, Ill., F.E. Peacock, 1977.
Dobyns, Zipporah Pottenger, Development of Devices for Measuring Selected World Views, THESIS, University of Arizona, 1966.
Hilts, Philip J., Behavior Mod, New York, Bantam Book, 1976.