Alfie Kohn's newest book, Punished by Rewards: The Trouble with Gold Starts, Incentive Plans, A's, Praise, and Other Bribes, details the destructive effects of rewards and questions many of the most common assumptions of teachers, parents, and employers about motivation. Kohn has also written No Contest: The Case Against Competition and The Brighter Side of Human Nature: Altruism and Empathy in Everyday Life. He writes frequently on human behavior and education and gives lectures and workshops for teachers. He was interviewed for the Harvard Education Letter by Sara-Ellen Amster.

HEL: Rewards and praise—how big a problem are they, really?

Kohn: If I had to choose between punishing and rewarding, I would pick rewarding. But many of the people who are already persuaded that punishment is not a good idea nevertheless believe that rewards represent an enlightened alternative. The truth is that rewards and punishments are not opposites—simply two sides of the same coin.

Rewards, as one pair of researchers has aptly put it, are just control by seduction. It's important to reveal controlling techniques for what they are and to oppose them, even if we're talking about what might seem to be the slightly more innocuous of two version s of control.

HEL: What does the research tell us about the relation between rewards and creativity?

Kohn: Rewards kill creativity. Some twenty studies have shown that people do inferior work when they are expecting to get a reward for doing it, as compared with people doing the same task without any expectation of a reward. That effect is most pronounced when creativity is involved in the task.

Rewards undermine risk-taking. When I have been led to think of the "A" or the sticker or the dollar that I'm going to get, I do as little as I have to, using the most formulaic means at my disposal, to get through the task so I can snag the goody. I don't play with possibilities. I don't play hunches that might not pay off. I don't attend to incidental stimuli that might or might not turn out to be relevant. I just go for the gold. Studies show that people who are rewarded tend to pick the easiest possible task. When the rewards are removed, we tend to prefer more challenging things to do. Everyone has seen students cut corners and ask: "Do we have to know this? Is this going to be on the test?"

But we have not all sat back to reflect on why this happens. It's not laziness. It's not human nature. It's because of rewards. If the question is "Do rewards motivate students? The answer is "Absolutely. They motivate students to get rewards." And that's typically at the expense of creativity.

Rewards undermine intrinsic motivation. At least seventy studies have shown that people are less likely to continue working at something once the reward is no longer available, compared with people who were never promised rewards in the first place. The more I reward a child with grades, for example, the less appeal those subjects will have to the child. It is one of the most thoroughly researched findings in social psychology, yet it is virtually unknown among educational psychologists, much less teachers and parents.

HEL: Is there something children can say to themselves to counter this effect, such as, "this reward is good, but it's not the reason I'm doing the work"?

Kohn: There has been some research on trying to immunize children against the destructive effects of rewards by modeling just this kind of response. "It's nice to get the "A" or the sticker, but it's really fun to solve a problem or write a story." There have been mixed results.

I think it's a mistake to place the burden on students. What has to change is the structure. If a teacher must give grades, the number one obligation of that teacher is to do everything possible to help students forget that grades exist. The last thing we as educators ought to be saying is, "C'mon Sheila, you're so close to that A," or "Only a C this time—what happened, Benjamin?" or "Do you want a zero in here, young lady?" In fact, I think that individual assignments ought never to carry a letter or number grade, even if the teacher is required, unhappily, to give grades at the end of the term.

HEL: What about giving kids grades for effort?

Kohn: Never grade for effort—especially with a separate mark. Grades and other rewards and punishments help to explain why a lot of kids have given up trying. To think that we can solve the problem by the same technique that caused it is naïve, to say the least.

Many students perceive a poor grade for effort as a message that they are failures even at trying. And a high effort grade coupled with a low achievement grade sends the message that there's no point in putting out any effort because they're not going to be able to do better. Finally, giving effort grades removes responsibility from teachers to help students engage with what they're learning. Instead, it blames the child for not wanting to do whatever the teacher has demanded.

HEL: Is there anything parents can do to counteract the negative effects of grading?

Kohn: parents ought to join in making grades as unsalient and invisible as possible, in order to give kids a chance to find the material intrinsically interesting. Unfortunately, many parents—well-meaning but appallingly misguided—actually pay children for good grades. This is a reward for a reward! If some perverse foundation had hired me to try to devise a plan that would destroy what is left of students' interest in learning, it would look very much like this.

There's a nationwide program sponsored by Pizza Hut called "Book it!" that consists of bribing kids with pizzas for reading books. John Nicholls at the University of Illinois says that the predictable effect of such a program is a lot of fat kids who don't like to read.

What kinds of books are they likely to pick in such a program? Short books. Easy books. The whole point is just to get through this tedious chore so they can snag a slice of pepperoni.

When anyone says rewards work, I always respond with two questions: Work to do what? And at what cost?

HEL: Are rewards ever good?

Kohn: If you understand the distinction between something pleasant and using that something as a reward, then I'm not sure we needs rewards. I want to make sure I'm not misunderstood here. There's nothing wrong with having a popcorn party for a class on Friday. There is something wrong with saying, "If you're good this week we'll have a popcorn party."

There's nothing wrong with giving a kid an allowance. There is something wrong with saying, "I'll pay you if you help out around here." There's nothing wrong, heaven knows, with love and affection and approval. There is something dreadfully wrong with using those things as M&Ms to reward a child for compliance.

What's wrong, in a word, is contingency. If we define rewards in terms of that contingency, then I scratch my head to find any examples where it's necessary or desirable.

HEL: The most surprising thing in your book is what you say about praise. Many teachers feel it's their duty to praise children, particularly those with low self-esteem, who come from families where they rarely get praise.

Kohn: Praise is a different matter from tangible rewards in some respects. I don't argue that all positive comments are bad. It depends on why we're giving them, how we're giving them, and what the effects of giving them are.

But even to take a moderate position strikes many people as heretical because teachers are taught to slather it on, to praise every child every day, with nary a thought that there could be something damaging about it.

Children from loveless homes need unconditional support and warmth and encouragement. Praise is the very opposite of that. Praise by definition is conditional: Do this and you'll get that. Jump through my hoops and I'll tell you how proud I am of you. More conditionality, more strings attached, more control is really not what students need. It does not build self-esteem to be given verbal doggie biscuits for complying with someone else's standards.

What builds self-esteem is the chance to feel self-determining, to help to formulate the standards for one's own behavior and then be able to judge one's own conduct or works in accordance with those standards.

HEL: You caution teachers to avoid praise that sets up a competition. What do you mean?

Kohn: It's never a good idea to praise a student by comparing her to someone else: "You're the best in the class," for example. The research is quite clear that such comments undermine intrinsic motivation, but their most pernicious effects are subtler. They encourage a view of others as rivals rather than as potential collaborators. What's more, they lead people to see their own worth in terms of whether they have beaten everyone else—a recipe for perpetual insecurity.

HEL: Is there research evidence of the long-term effects of this kind of praise?

Kohn: No one has ever done a controlled, longitudinal study. We know about short-term effects of praise, and we can extrapolate. It is possible that children who are marinated in rewards, including verbal rewards, become less interested in the things for which they are rewarded.

How many adults are serious readers, for example, if it's not part of their profession? And how many adults do we know who continue to be desperate for someone else's approval, incapable of feeling satisfied with their work, such that their moods rise or fall depending on other people's thumbs up or thumbs down? It seems plausible to me that a childhood full of being praised for meeting someone else's standards is related.

HEL: What are some of the good ways to praise—or, if you don't want to use that word—to encourage?

Kohn: We could improve our responses immeasurably by saying less and asking more. It is very useful and affirming when we ask questions about what a student has done and why he or she has done it. That opens up a conversation where the student comes to reflect on his or her work and also feels a sense of support from the teacher.

The classic example is of a preschooler who has drawn something and present sit to a parent or teacher who might respond by saying, "I notice you have the sun over here next to the mountain instead of behind it. That was an interesting choice. Why did you do that? Or, "You use a lot of orange. Is that your favorite color? Or "What are you going to draw next?"

HEL: What would you like educators to walk away saying after they read Punished by Rewards?

Kohn: The first thing that occurs to many people who read the book or hear me speak is "oh my God, I have everything wrong. Does this mean I'm a bad teacher?" If you respond that way you're probably a very good teacher, because the willingness to question one's basic assumptions and practices stands right at the top of my list of qualities of a good teacher. The people I worry about are those who dismiss this out of hand and say, "It's unrealistic," or "Rewards work, and I don't want to hear about your studies."

If I had to summarize the book in one sentence, it would be this: We can never meet our long-term goals by doing things to students, only by working with them. Rewards, like punishments, are ways of doing things to people. And to that extent they can never help them to take responsibility for their own behavior, to develop a sense of themselves as caring people, to work as creatively as they can or become excited learners for the rest of their lives. Rewards, like punishments, actively undermine those goals.