by B. F. Skinner: In my address to the Japanese Psychological Association on Sunday, I pointed to the importance of cultural practices in bringing out the best of which the individual is
capable. In spite of the extraordinary genetic endowment of the human species,
including the capacity to be changed very quickly by encounters with the environment,
an individual alone, without the help of others, could in one lifetime acquire only a very
small part of the repertoire exhibited by the average person. Exposure to other
members of the species and to practices which have evolved over the centuries in
permitting the individual to profit from what others have already learned makes an
enormous difference. I am aware of one of those differences in discussing my subject
today. Different groups of people have developed different practices – in education,
religion, government, psychotherapy, economics, and daily life. The results have
sometimes led to the notion of national character, as if it were the people who differed in
some genetic way rather than the culture. The behavioral scientist cannot, however,
make much of “character.” The practices are the things we study. Different practices
yield different problems, and I am not at all sure that the subject I am to speak about to
this distinguished audience today will seem to be as important as in America or even as
interesting. I believe it represents, however, a crucial step in the evolution of the species
and of the practices of a culture.

Many things ill the world are called unpleasant or punishing. We avoid or escape from
them if we can. It is part of our genetic endowment that we should do so because they
are biologically harmful things and should have played an important role in natural
selection. Three great historical examples to which the species has been exposed are
starvation, illness, and exhausting labor. The species has made great progress in
dealing with them. Through the discovery of agriculture and ways of storing and
transporting food, mankind has (in part at least) escaped from the suffering of famine.Through medicine and sanitation, it has escaped from many of the sufferings of illness
and early death. Through physical technology it has escaped from the suffering of
exhausting labor. The only sufferings to which many members of the human species are
still exposed are those we inflict upon each other. People threaten or destroy life, liberty,
and property in war, terrorism, and organized crime. Political scientists sometimes
define government simply as the power to punish. The Christian religion threatens an
eternity of hell-fire, and its evangelists continually remind us of this most terrible of all
punishments. Education has a long punitive history. The cane and the taws (a leather
strap, which, like the policeman’s truncheon, leaves fewer permanent marks) are still
used in British schools, and the paddle is once again in use in America. Even without
corporal punishment, teachers are still so punitive that most students simply study to
avoid the consequences of not studying. Industrial incentives are really punitive, We
think of a weekly wage as a kind of reward, but it does not work that way. II establishes
a standard of living from which a worker can be cut off by being discharged, workers do
not work on Monday morning because of the pay they will receive at the end of the
week; they work because a supervisor will discharge them if they do not. Under most
incentive systems, workers do not work for things but to avoid loosing them.

Psychotherapy is not an exception. Psychotics were once put into snake pits (certainly
an aversive measure) and the so-called “behavior therapy” parodied in the movie A
Clockwork Orange, using nausea or electric shock in Pavlovian conditioning, is little
more than a scientific form of punishment. The director of a military hospital in Viet Nam
once told me that he was using operant conditioning with psychotic patients, and I was
pleased until I discovered that he had simply told his patients that if they did not go to
work they would get electric shock therapy. Fortunately much psychotherapy is nonpunitive. But even in our daily Jives, we tend to fall back on mild forms of punishment –
criticizing, complaining, nagging, if not the physical measures which result in the
battered wife or child.

When treatment is too severe, people escape from punishment – from governments by
defecting to other governments, from religions by becoming apostates, from schools
and universities by becoming truants or dropouts, from industry by absenteeism orquitting work, from the family by divorce or running away from home. If those who have
been treated punitively have the power, they may counterattack – as by overthrowing a
government, reforming a religion, vandalizing schools and attacking teachers, striking
against or boycotting industries, and engaging in violent social action.
When those who use punishment are strong enough to prevent the escape and
counterattack of those they punish, the effect may be a kind of sullen inaction, a
numbness, a complete waste of potential.

No one likes any of these consequences of punishment, and certainly no one likes to be
punished. Why then does punishment continue to be such an important instrument of
social control? Are we perhaps genetically inclined to be aggressive toward each other?
Certainly it is easy to point to reasons why we should be. Those members of the
species who were most strongly inclined to defend themselves and their property by
physical force, to act aggressively as predators, and to compete aggressively in sexual
competition should have been most likely to survive and transmit their tendencies.
But we also learn to punish. Quite apart from any genetic inclination, the things we do
which harm others usually have reinforcing consequences for us. We learn to use
aversive measures; we also learn to accept the aversive practices of the culture of
which we are a part. And here we see a possible clue to the answer to our question. For
those who are powerful enough to use it, punishment has rewarding consequences. The
people we punish behave as we dictate, and the things we take away from them in the
name of punishment are the things we ourselves get. The unwanted consequences I
have mentioned are all deferred. Unfortunately, we are much more likely to be affected
by things that happen quickly. The immediate rewards of using punishment are much
more powerful than the deferred disadvantages and losses.

Is there a technology comparable to agriculture, medicine, and engineering to which we
may turn to find alternatives to this last great source of human suffering? I believe there
is. To many people it is known as “behavior modification,” but the term has been widely
misunderstood. I do not mean the modification of behavior with drugs or implanted
electrodes, or Pavlovian conditioning with electric shock or nausea-producing drugs.
The term was invented to refer to behavior changed through what the layman calls “reward” or what, in the experimental analysis of behavior, we call “positive
reinforcement.” Behavior modification in the exact sense of the application of an
experimental analysis of behavior is, I believe, the first organized effort to develop
alternatives to punitive practices. Many humane people have, of course, devised
alternatives to punishment, but the fact remains that the principles they have employed
have not prevailed in the world at large. And just as we can explain the widespread use
of punishment by pointing out that the gains are immediate and the losses deferred, so
the failure to use positive alternatives may be clue to the fact that losses are immediate
and gains deferred. When we reward another person, we must give up something we
possess or perform some service. It is only in the future that the person behaves in
ways which are rewarding to us. Behavior modification is at last making inroads into
cultural practices as an alternative to punishment because these various consequences
of both reward and punishment have been clarified by a scientific analysis and by the
emergence of a technology which will in the long run, I believe, be comparable in its
effect upon human life to agriculture, medicine, and physical technology in eliminating
this last great source of human suffering.

Rather than go into the details of the scientific study of positive reinforcement, I shall
simply describe some examples. The classroom is a good place to start. The ordinary
teacher, probably burdened by too many students and poorly designed instructional
materials, is likely to fall back on punishment – on criticism or ridicule, if not a more
violent corporal form. They are all the more likely to do so because, especially when
busy, we all tend to deal only with those things which are brought to our attention.
Students are always reminding the teacher that it is time to criticize or complain but
seldom that it is time to praise or commend. Misbehavior is the signal for punishment.
When students are behaving well, the teacher is tempted to “let well enough alone.” But
“letting well enough alone” is a fatal principle, Students should be given attention when
they are behaving well, not when they are behaving badly. A great change usually takes
place in the classroom when teachers learn to look for chances to use positive

Teachers may also contrive special reinforcing contingencies. They may createreinforcers in the form of credits or tokens exchangeable for some of the natural
reinforcers in the life of the student. A teacher in a sixth-grade class in America used
token reinforcers and gave them special strength when she reinforced her students’
behavior on the schedule which is responsible for the extraordinary power of gambling
systems, as in lotteries or casinos. This particular teacher had had no special training as
a behavior modifier, but she had read of the possibilities and decided to try an
experiment. The school was in a lower economic neighborhood, and she was having
some difficult problems. Families did not insist that their children do their homework, and
the children worked only inefficiently in class. On a Monday morning the teacher put a
small transistor radio on her desk. She told the class: that on Friday afternoon, there
would be a drawing, and that the student who drew the lucky ticket would win the radio.
The students were intrigued. How were they to get tickets? It would be quite simple,
said the teacher. Whenever students brought in their completed homework, they were to
write their names on small cards and drop them in a jar. When they completed a
classroom assignment, they were to do the same thing. On Friday the jar would be
shaken and a card would be drawn. The teacher reported an immediate change in the
behavior of her students. They all did their homework and their assignments. The
teacher’s task was greatly simplified, and she was only too willing to spend the money
needed for a different prize each week. Meanwhile, of course, the students were
learning a great deal because they were doing their work.

Such an experiment is often criticized. It is said that the children are being “bribed” to do
their work. This is not exactly true. A bribe is something paid to induce someone to do
something illegal or wrong. Those who call positive reinforcement bribery are confessing
to a very low opinion of school work. One could also argue that students might better
work for positive reinforcers of that kind than to escape punishment. Of course we do
not want students who will continue to study only when they get lottery tickets. The
behaviors they acquire in school should be those which will eventually be reinforced by
the natural contingencies of daily life. The natural contingencies cannot be brought into
the classroom for instructional use. That was the great misunderstanding of the
philosophy of education of John Dewey. We should educate for real life, but we cannotuse real life effectively in the school. Classroom contingencies must be to some extent
contrived, but if contrived effectively, they will produce behavior which will work to the
advantage of everyone in the natural contingencies to which the student is later

Even in the school, conspicuous reinforcers such as tokens or credit points are needed
only if the classroom is badly out of control. There is a natural reinforcer available in the
classroom. An important genetic feature of the human species – possibly of all species –
is that being successful is itself reinforcing. One pushes, and the pushing is reinforced
when the object moves. Finding the right answer to a question can be a highly
reinforcing event. In traditional instructional material. the student is not often right. One
of the essential points of programmed instruction is to increase the chances that the
student will be successful. This is achieved by breaking material into many small steps
so designed that each can be taken readily and successfully.

Another feature of a good program is that the student’s progress is obvious. The student
moves into material which a short time before he could not have dealt with properly.
There arc learning centers in American schools which teach children to read even
though they come from illiterate or non-English speaking families. Each student works at
his own pace, listening to a tape recording and responding by marking chemicallytreated worksheets on which the student’s responses are immediately shown to be right
or wrong. Children like these centers: they do not vandalize them; they do not try to
escape from them. Success and progress are highly reinforcing. They are always
available as an alternative to the punitive practices of the classroom.
Another field in which behavior modification (or the application of a behavioral analysis)
has been effective is industry. The industrial revolution made a great change in the
incentives of the worker. It destroyed many natural reinforcing contingencies. In the long
run, the old craftsman was perhaps working for money or for other goods, but every
step of what he did was reinforced by certain immediate consequences. When, in the
industrial revolution, his work was broken up into small pieces and single pieces
assigned to separate workers, there was nothing left by way of a reinforcer except
money. The natural consequences of the behavior had been destroyed. That is whatMarx called the alienation of the worker from the product of his work. In addition. the
system became primarily aversive. The worker did not work, as I have said before, for a
wage, but to avoid discharge and the loss of a standard of living maintained by a wage.
Workers work under supervision, and supervisors, like teachers, tend to respond only to
opportunities to criticize or complain. When they are taught to look for chances to
commend rather than criticize, workers’ behavior improves, and workers report that they
like their jobs better. Good industrial engineering also attempts to make clear the
relation between the work and the ultimate product. Problems of absenteeism and
changing jobs have been solved in some cases by adding a scheduled reinforcer similar
to that employed by the teacher I mentioned. The employee who turns up for work each
day receives a lottery ticket; the employee who stays home may miss his big chance.
In America there have been some violent objections to the use of behavior modification
in industry on the grounds that it is designed simply to get more work out of the worker
and to increase the profits of management. That may often be true, and in the long run it
could be self-defeating. On the other hand, most countries in the world today are
suffering from a declining productivity of the worker. It is said to be one of the principal
causes of inflation. Any method of control needs to be properly contained, and the
exploitation of the worker should certainly be prevented, but if changes in industrial
incentives will make it possible for workers to work more productively and carefully and
at the same time to enjoy their work, then everyone, and particularly the workers
themselves, will benefit.

One of the first fields in which the analysis of behavior was applied was the institutional
care of psychotic and retarded people. Here, again, the standard practice encouraged
punitive measures. Attendants who are charged with watching rooms full of psychotics,
most of them sitting around doing nothing, are likely to respond only to misbehavior. As
a result, misbehavior is reinforced by the attention and eventually calls for more punitive
measures. When attendants are taught to look for behavior to commend, there is a
great change. Psychotic and retarded people, because of their detects, are not sensitive
to the normal reinforcing contingencies of daily life. They need a “prosthetic”
environment. Eyeglasses, hearing aids, crutches, and wheelchairs are prosthetic devices, which enable people to behave effectively although handicapped in one way or
another. A prosthetic environment is an environment in which those who are insensitive
to standard contingencies of reinforcement may nevertheless behave in productive and
dignified ways. A token economy is a prosthetic measure which may permit psychotic
and retarded people to lead reasonably dignified lives in spite of their disadvantages.
Prisons and schools for juvenile delinquents are other places in which behavior
modification has replaced punitive measures. Not only are these institutions designed to
punish people for misbehavior in the past, they are punitive during incarceration in the
sense that prisoners tend to receive attention from prison authorities mainly when they
have misbehaved. There are few incentives in a prison for behaving well. This need not
be the case. In one experiment in a school for juvenile delinquents in America, the boys
were given a choice. They could, if they liked, do nothing during the day. They could sit
on a bench, eat nutritious if not very interesting meals, sleep in a dormitory at night. If
they earned points, however, they could get more interesting food, have access to
billiard tables and television sets, rent a private room, or even buy some time away from
the institution. They earned points in part by performing janitorial services, but mostly by
studying under programmed instruction. Many of these hoys had been abandoned by
the educational systems to which they were exposed, and they now discovered that
they were able to learn to read and write and do simple arithmetic.

The boys participated in the experiment for only a few months, but the recidivism rate
was greatly changed. At the end of one year after release, 25 percent of them were
again in trouble, but the figure would otherwise have been 85 percent. At the end of the
second year, 50 percent were in trouble, and after three years, there was little evidence
of the effectiveness of the program. The boys had gone back to a world in which too
many wrong contingencies prevailed. Even so, the experiment had more than paid for
itself, so far as the state was concerned.

A serious question has also been raised about programs of this sort – and strangely
enough in the name of civil rights. Do psychotic or retarded persons and prisoners not
have a right to food, clothing, privacy, and a reasonable chance to enjoy life? Can these
things properly be taken away so that they can be used as reinforcers? In some states,laws have been passed to restrict the use of behavior modification on just those
grounds, but the argument rests on a serious misunderstanding. What are the rights of a
prisoner, for example? A person who has been incarcerated and then given the things
he needs to survive is being denied a very basic right. He is being destroyed as a
person by having his reinforcing contingencies stripped away. The same thing happens
to those on welfare. A humane society will, of course, help those who need help and
cannot help themselves, but it is a great mistake to help those who can help
themselves. Psychotic or retarded people who in essence earn their own living would be
happier and more dignified than those who receive their living free and are then treated
punitively because in the absence of reinforcing consequences they behave badly.
Those who claim to be defending human rights are overlooking the greatest right of all:
the right to reinforcement.

Face-to-face psychotherapy is another field in which behavior modification is used, and
it is particularly significant because that kind of therapy is usually concerned precisely
with the effects of punishment. Psychoanalysis can be regarded as a systematic
reversal of the effects of the punishment one has received at some earlier time, and
psychotherapeutic counseling is largely a matter of finding a way of life – a new place to
live, a new job, new friends–in which the client’s behavior will be positively reinforced.
I wish I could say that government is another field in which there is an interest in
abandoning punitive measures, but proposals to work through positive reinforcers in
government are usually viewed as surprising or amusing. Jonathan Swift,
in his great book, Gulliver’s Travels described a state in which good behavior was
reinforced and bad behavior not punished, but it was offered as a satire. And many
people find amusing a small experiment in an American city in which motorists began to
receive postcards saying, “You were observed to come to a full stop at the intersection
of such-and-such a street at such-and-such a time. Congratulations.” No doubt serious
violators were not affected, but I am sure that many drivers receiving those cards came
to a full stop at intersections for some time thereafter. The problem of punishment in
dealing with criminal behavior is of long standing, and it will not easily be solved,
because those who are harmed by crime tend to be vicious in suppressing it. InAmerica, the death penalty is being reinstated, and in a recent television presentation
called “Scared Straight,” potential young offenders were taken to a prison to hear
prisoners describe their lives in brutal terms. The program was favorably received by
critics. It was said that the potential offenders then “went straight,” but the figures have
been disputed. In any case, the intensification of punishment is no solution. In England
in the eighteenth century, 200 crimes were punishable by death. One of them was
stealing silk handkerchiefs, but the crowds who assembled to watch the executions
were so intensely interested in what was happening that handkerchief thieves had a fine
opportunity. Obviously they were not deterred by the spectacle of their colleagues’ being
hanged. In the long run, the solution to the problem of crime is not punishment but the
elimination of the conditions under which people commit crimes. For example, there
would be far less crime if everyone had a job.

There are those who object in a much more general way to behavior modification or
the application of a behavioral analysis on the grounds that it is not right for one person
to control another. We have had so much experience with punitive control that we
conclude that all control is wrong. That conclusion is perhaps one of the greatest
wrongs worked by punishment. We are all engaged in controlling behavior all the time.
As parents we control the behavior of our children and (if less obviously) as children, the
behavior of our parents. As teachers we control the behavior of our students, and as
students, the behavior of our teachers. As employers we control the behavior of
employees and as employees, the behavior of our employers. As governors we control
the behavior of those we govern and as the governed, the behavior of governors. As
acquaintances, friends, and lovers, we control the behavior of each other. We may not
may not know we are doing so; few, if any, of us are aware of all the ways in which
behavior is controlled. We may or may not control deliberately – that is, because of any
particular consequence for us. Nevertheless, we do control. The fact is that too often we
do it badly, and badly because, most often, punitively. The more we know about control,
the more rapidly we shall move toward acceptable methods.

When we look at the world today with its war, terrorism, and violence in so many
places, a non-punitive society seems “utopian” in the sense of impossible. And, indeed,we are not likely to arrive at a peaceful world in the immediate future by applying the
experimental analysis of behavior to international diplomacy. In any case, peace in the
simple sense of the absence of violence is no solution to the problem. Like the
permissiveness which some countries have recently explored, it offers no effective
alternative to punitive measures. Perhaps our best opportunity will be to start below the
level of international affairs. If, because of positive consequences alone, people can
acquire knowledge and skills, work productively, treat each other well, and enjoy their
lives, those who deal with international affairs may be able to use non-punitive
measures more effectively. It is the unhappy and the frightened who resort to war.
International negotiations among happy nations should be more successful.
In a sense the search for a non-punitive society is nothing more than the traditional
search for happiness. The experimental analysis of behavior helps in that search by
identifying the essential conditions of happiness. When we act to avoid or escape from
punishment, we say that we do what we have to do, what we need to do, and what we
must do. We are then seldom happy. When we act because the consequences have
been positively reinforcing, we say that we do what we like to do, what we want to do.
And we feel happy. Happiness does not lie in the possession of positive reinforcers; it
lies in behaving because positive reinforcers have then followed. The rich soon discover
that an abundance of good things makes them happy only if it enables them to behave
in ways which are positively reinforced by other good things.

We cannot arrive at a happy world simply by foregoing punitive measures. We must
solve other problems. Overpopulation, the ultimate exhaustion of the world’s resources,
and the pollution of the environment are the natural punitive consequences of the
reckless behavior we now exhibit. If we fail to solve them we shall all be punished by the
most terrible miscarriage of the evolution of the human species: a nuclear holocaust.
We cannot solve these problems through any aversive means. A cooperative rather
than a competitive solution is needed. Whatever the final form of that solution, we can
all move toward it by turning as often as possible to positively reinforcing measures in
our schools, our industries, our governments, our families, and our daily lives.