Skip to content
ascd logo

Log in to Witsby: ASCD’s Next-Generation Professional Learning and Credentialing Platform
September 1, 1995
Vol. 53
No. 1

Punished by Rewards? A Conversation with Alfie Kohn

    In classrooms where students can make choices about learning and have tasks of worth to explore, the need for punishments or rewards declines sharply.

      Both rewards and punishments, says Punished by Rewards author Alfie Kohn, are ways of manipulating behavior that destroy the potential for real learning. Instead, he advocates providing an engaging curriculum and a caring atmosphere “so kids can act on their natural desire to find out.”
      The following interview took place at ASCD's Annual Conference on March 27 in San Francisco.
      Alfie, we educators use punishment quite a lot, but we've come to understand that it's not a very effective motivation. We've been convinced that it's much better to use rewards instead. But now you come along and say that's wrong, too. Why?
      First, let's make sure we agree on your first premise, which is that punishment is destructive. A number of people seem to think if we call it “consequences” or insert the modifier “logical,” then it's okay. “Logical consequences” is an example of what I call “punishment lite,” a kinder, gentler way of doing things to children instead of working with them.
      Having said that, I'll move on to rewards. Rewards and punishments are both ways of manipulating behavior. They are two forms of doing things to students. And to that extent, all of the research that says it's counterproductive to say to students, “Do this or here is what I'm going to do to you,” also applies to saying, “Do this and you'll get that.” Ed Deci and Rich Ryan at the University of Rochester are right when they call rewards “control through seduction.”
      And you're saying rewards are just as undesirable as punishment.
      By virtue of being controlling, they're likely to be experienced as aversive in the long run. The reason is that while students would certainly like to have the goody itself—the pizza or money or gold star—none of us enjoys having the very things we desire used as levers to control our behavior. So it's the contingency of the goody—“Do this and you'll get that”—that accounts for its punitive status over the long haul.
      You're saying that's the case even for kids who find a certain task rewarding for its own sake?
      Rewards are most damaging to interest when the task is already intrinsically motivating. That may be simply because there is that much more interest to lose when extrinsics are introduced; if you're doing something boring, your interest level may already be at rock bottom.
      However, that doesn't give us license to treat kids like pets when the task is uninteresting. Instead, we need to examine the task itself, the content of the curriculum, to see how it can be made more engaging. Regardless of what we do about it, though, one of the most thoroughly researched findings in social psychology is that the more you reward someone for doing something, the less interest that person will tend to have in whatever he or she was rewarded to do.
      In Punished by Rewards, you cite a lot of research on points like that. You're saying this is not just your opinion.
      That's right. There are at least 70 studies showing that extrinsic motivators—including A's, sometimes praise, and other rewards—are not merely ineffective over the long haul but counterproductive with respect to the things that concern us most: desire to learn, commitment to good values, and so on. Another group of studies shows that when people are offered a reward for doing a task that involves some degree of problem solving or creativity—or for doing it well—they will tend to do lower quality work than those offered no reward.
      That seems so contrary to our everyday experience. Everybody is used to getting rewards and giving them. As educators we think it's only right to give rewards; kids who do good things deserve rewards.
      What kids deserve is an engaging curriculum and a caring atmosphere so they can act on their natural desire to find out about stuff. No kid deserves to be manipulated with extrinsics so as to comply with what others want.
      It's remarkable how often educators use the word motivation when what they mean is compliance. Indeed, one of the fundamental myths in this area is that it's possible to motivate somebody else. Whenever you see an article or a seminar called “How to Motivate Your Students,” I recommend that you ignore it. You can't motivate another person, so framing the issue that way virtually guarantees the use of controlling devices.
      Moreover, motivation is something that kids start out with. You don't have to bribe a young child to show you how she can count to a thousand million or decode signs on the highway. But research shows that by the middle—or certainly by the end—of elementary school, this intrinsic motivation starts to tail off sharply—by an extraordinary coincidence, around the time that grades have started to kick in.
      Surely it's unrealistic to expect that all kids will find all the curriculum intrinsically motivating. There are some things that kids just have to slog through, aren't there?
      Well, a given child is likely to be more interested in some things than others, but we're not talking about putting something on the chalkboard and expecting kids to jump up and down and say, “I can't wait to get at this!”
      Skillful teaching involves facilitating the process by which kids come to grapple with complex ideas—and those ideas, as John Dewey has told us, have to emerge organically from the real-life interests and concerns of the kids. “Which is bigger, 5/7 or 9/11?” The correct answer is, “Who cares?” But kids care very much about how fast they are growing. Within that context, the skills necessary to figure it out become interesting to most kids. “What's the difference between a simile and a metaphor?” Same answer; few members of our species would find that distinction intrinsically motivating—but kids are highly interested in writing a story about dinosaurs or how a spaceship carries them away. In the context of a task that matters to students, the specific skills we care about can be taught naturally without sugarcoating, without games, and above all without offering kids little doggie biscuits for doing what we tell them.
      Let me ask about praise, which is particularly tricky, because it's not a tangible reward. If I tell one of my staff members that he or she did a terrific job on something, am I giving a reward at that point?
      That's an interesting question, and I wish more educators would ask it, regardless of what the answer turns out to be.
      Positive feedback that is perceived as information is not in itself destructive and indeed can be quite constructive, educationally speaking. And encouragement—helping people feel acknowledged so that their interest in a task is redoubled—is not a bad thing. But most praise given to children takes the form of a verbal reward, which can have the same destructive impact as other rewards: it feels controlling, it warps the relationship between the adult and the child—and between the child and his or her peers—and it undermines interest in the task itself.
      It's not a coincidence that coercive discipline programs rely to a large extent on getting compliance by slathering on praise. A typical example is the elementary school teacher who is taught to say, “I like the way Cecilia is sitting so nice and quiet and ready to work.” I have multiple objections to this practice.
      First, the teacher hasn't done Cecilia any favors. You can imagine some of the other kids coming up to her after class: “Miss 'nice and quiet' dork!”
      Second, the teacher has just turned a learning experience into a quest for triumph. She has introduced competition into the classroom. It's now a contest to see who is the nicest, quietest child—and the rest of you just lost.
      Third, this is a fundamentally fraudulent interaction. The teacher is pretending to speak to Cecilia, but she's really using Cecilia to manipulate the behavior of the other people in the room—and that's simply not a nice way to deal with human beings.
      Fourth, and possibly most important, I ask you to reflect on what is the most important word in that expression. I believe it's I. Even if such a practice “works,” it has worked only to get Cecilia and the other people watching to become concerned about what I demand, regardless of what reasons I may or may not have for asking her to do something. Cecilia is not helped one iota to reflect on how her experience affects other people in the room or what kind of person she wants to be.
      On that point, I like to think about the questions that kids are encouraged to ask in different kinds of classrooms. In one dominated by consequences, kids are led to think, “What do they want me to do, and what will happen to me if I don't do it?” In a reward-oriented classroom, including one that is characterized by praise, kids are led to ask, “What do they want me to do, and what will I get for doing it?” Notice how fundamentally similar those two questions are and how radically different either one is from the questions, “What kind of person do I want to be?” or “What kind of classroom do we want to have?”
      What about less successful students? A lot of educators feel strongly that they need even more praise than other kids. They need to be praised when they make the slightest bit of progress.
      No research supports the idea that praising children for inching up the adult-constructed ladder helps them develop a sense of competence. Indeed, praise for success at relatively easy tasks sends a message that this child must not be very bright. Moreover, children are not helped to find the material itself important or interesting if they are praised for doing it. In general, the more kids are induced to do something for a reward, whether tangible or verbal, the more you see a diminution of interest the next time they do it. That can be explained partly by the fact that praise, like other rewards, is ultimately an instrument of control, but also by the fact that if I praise or reward a student for doing something, the message the child infers is, “This must be something I wouldn't want to do; otherwise they wouldn't have to bribe me to do it.”
      What you're saying is not going to be readily accepted by most people. It seems to go against our everyday experience.
      It does and it doesn't. For example, parents come up to me and say things like, “You know, it's funny you say this, because just yesterday I asked my kid to clear the table after dinner, and he said, `What are you going to give me for it?'” What I find remarkable about that is not what the child said, but that the parent is asking me to shake my head and commiserate about These Kids Today. What I want to ask is, “Where do you think the kid learned this?” And if I do ask that, with very little prompting, people understand.
      There's even some research in Missouri showing that when undergraduates were asked, “Do you think rewards lead to higher or lower interest in a task?” they guessed wrong. But as soon as the research results were explained, everyone said, “Oh yeah, I knew that.” A lot of people have had the experience of having done something just because they loved it—until they started to get paid for doing it, after which they wouldn't dream of doing it again without getting paid. The phenomenon whereby extrinsic motivators cause intrinsic motivation to evaporate is not on the tips of our tongues, but it's not that far from consciousness, either.
      All the same, it's a different way to think about things. For example, I like it when people recognize me for an accomplishment of some kind.
      Yes, of course. We all want to be appreciated, encouraged, and loved. The question is whether that need must take the form of what often looks like a patronizing pat on the head and saying “Good boy,” to which I believe the most logical response is, “Woof!”
      Now, I know a lot of adults who are praise junkies: sadly unable to think about the worth of their own activities and actions and products, and utterly dependent on someone else to tell them they did a good job. That is the logical conclusion of being marinated in praise for years. But maybe there is a more empowering and respectful way of sharing one's opinions than what amounts to a verbal reward.
      I'm struck by teachers who say over and over to me, “You don't understand the kind of backgrounds and home lives that these kids have; they come from loveless, sometimes brutal places, and you're telling me not to praise them?” My answer is, “Yes.” What these kids need is unconditional support and encouragement and love. Praise is not just different from that; it's the opposite of that. Praise is, “Jump through my hoops, and only then will I tell you what a great job you did and how proud I am of you.” And that can be problematic. Of course, with positive feedback, it's a matter of nuance and emphasis and implementation. That is not the case with gold stars, candy bars, and A's, which I believe are inherently destructive.
      One of the central myths we carry around in our heads is that there is this single entity called “motivation” that one can have more or less of. And of course we want kids to have more of it, so we offer them A's, praise, and pizza. The truth is that there are qualitatively different kinds of motivation. We need to stop asking “How motivated are my students?” and start asking “How are my students motivated?” The kind of motivation elicited by extrinsic inducements isn't just less effective than intrinsic motivation; it threatens to erode that intrinsic motivation, that excitement about what one is doing.
      So what are you suggesting instead?
      I sometimes talk about the three Cs of motivation. The first C is content. Far less interesting to me than whether a student has learned what he was supposed to is the question, “Has the child been given something to do worth learning?” If you ask me what to do about a kid being “off task”—one of our favorite buzzwords—my first response is going to be, “What's the task?” If you're giving them garbage to do, yes, you may have to bribe them to do it. If the kids have to endlessly fill in the blanks on dittos, you're not going to get rid of rewards or threats anytime soon.
      The second C is community: not only cooperative learning but helping kids feel part of a safe environment in which they feel free to ask for help, in which they come to care about one another as opposed to having to be manipulated to share or not be mean. Some of the outstanding work on creating caring communities is being done by the Developmental Studies Center in Oakland, California.
      The third C is choice: making sure that kids are asked to think about what they're doing and how and with whom and why. You know, kids learn to make good choices not by following directions but by making choices.
      You show me a school that really has those three Cs in place—where students are working with one another in a caring environment to engage with interesting tasks that they have some say in choosing—and I'll show you a place where you don't need to use punishments or rewards.
      End Notes

      1 A. Kohn, (1993), Punished by Rewards, (Boston: Houghton Mifflin).

      Education writer and consultant Ron Brandt is the former editor of Educational Leadership and other publications of the Association for Supervision and Curriculum Development (ASCD).

      Learn More

      ASCD is a community dedicated to educators' professional growth and well-being.

      Let us help you put your vision into action.
      From our issue
      Product cover image 195217.jpg
      Strengthening Student Engagement
      Go To Publication