Why Do We Punish Children?

January 2013

 Why Do We Punish Children?

By Alfie Kohn

Adapted from a four-part public exchange with Deborah Meier, beginning in January 2013, that appeared on Education Week‘s “Bridging Differences” blog.

Whenever it appears that I disagree with someone, I like to begin by figuring out where we’ve parted company. Is our disagreement substantive or just a function of how we’re defining our terms? Is it about description (the way we think things are) or prescription (the way we think they should be)? Is it about ends or means? With respect to the issues you’ve raised this week, Deb, I’d add a more specific question: Do you think punishment is sometimes appropriate and beneficial, or do you agree with me that it isn’t but contend that it sometimes must be used as a transitional measure, the idea being that it should be faded out once the students’ (or their parents’) trust in the educators has been established and a caring school community has been constructed?

I’m hoping you’ll say the latter, because then we can just argue about whether other stopgap measures might be used instead — or, if students are going to be punished, we can quibble about how often and for how long. I’d certainly agree that it’s difficult to quit cold turkey, but I think it’s important to invite educators and parents who may have been raised with punishment to question its value. (A small study published in Contemporary Educational Psychology found that preservice teachers were more likely to endorse punitive strategies with students if they themselves had been punished when they were children.)

By definition, to punish is to deliberately make someone suffer — either because a primitive version of justice seems to demand it (If you do something bad, then something bad must be done to you) or because it’s assumed that punishment will teach you a lesson. The premise here is that when we make you unhappy by forcing you to do something you find aversive, or by preventing you from doing something you enjoy, you’ll become a better person.

What punishments — even if they’re euphemistically called “consequences” (so we can feel better about making a child feel bad) — really do is make the child angry, teach him that you get your way in life by using your power over those who are weaker, and make it less likely that he’ll focus on how his actions affect others. Punishment undermines moral development by leading people to ask, “What do they want me to do, and what happens to me if I don’t do it” and actively discouraging them from asking, “What kind of person do I want to be?” (I’ve laid out these arguments in more detail in my books Unconditional Parenting and Beyond Discipline, and I’ve just posted the relevant section of the former book on-line for those who are interested.)

It’s crucial to question not only the effectiveness of punishment — in fact, it can never buy us anything more than temporary compliance, and it does that at a disturbing cost — but the beliefs that often underlie it: that kids are basically bad and will do terrible things without the threat of punishment hanging over them, that punishment is the best (or even only) way to socialize children, that the only alternative to punishment is permissiveness, that it’s an appropriate way to express love and care, and so on. As you know, many kids, too, have internalized some of these myths, which may be even sadder than encountering them in adults.

I don’t want to be a purist here and demand that we (as educators) impose on parents our humanistic ideals that may sound unfamiliar and suspicious, particularly regarding what we do with their children, without explaining what we’re up to, respectfully sharing our intentions, reassuring them about how our long-term objectives overlap with their own, and inviting their responses. But neither would I want us to shrug and say, “Well, they expect (or even demand) that kids will be punished when they misbehave, so we’d better do that.” You wouldn’t paddle a child just because a parent urged you to do so, right? Well, I draw the line at punishment itself, not just corporal punishment. I do so partly because of my bedrock values about how people of any age should be treated, and partly because the empirical case against punishment — its destructive effects and its lack of benefits — is so powerful.

Many people like to pretend it’s not really punishment if they can portray what they’re doing as “logical” or “natural”; you acknowledge candidly that at its core it’s really about power. And that in itself is a good reason to avoid it — for reasons based on what you have written about so eloquently elsewhere. Kids need to be able to develop trusting relationships with adults. You’ve pointed out many times that top-down, test-driven “school reform” teaches kids that their teachers aren’t trustworthy. But power-based interactions between teachers and kids — such as punishment — fundamentally disrupt that trust and any sort of caring alliance. The tougher the kid, the more critical it is to establish that alliance — and thus, paradoxically, the more important it is not to punish when the kid does something wrong.

Were you really able to get away with saying “we’ll see,” rather than laying out a list of penalties as most schools do, when students wanted to know how they’d be punished for an offense? Most observers would be appalled by the prospect of inconsistency and unpredictability. My own view is that a rigid, legalistic system, with the same penalties applied to every offender, does offer some reassurance that adults can’t play favorites — but it does so only in the context of a punitive system. You don’t need to worry about applying the same penalty to every offender when you’re not using penalties — that is, when you’ve replaced a “doing to” approach with a “working with” approach.

The latter, which I’ve described elsewhere (as have many others, of course), isn’t just about reacting differently to what kids do but seeing it differently, so a troubling action is construed as a problem to be solved rather than an infraction to be punished. And that’s the model I think we try to communicate to parents. When we’re asked, “What are you going to do to the kid who did this to my kid?”, we might reply, “Well, that depends on our goal. If we’re looking for revenge, I guess we could punish him. But if the goal is for your kid — and all the kids — to be safe here, then punishing that other student is the last thing we’d want to do. Here’s why…”

I’d like to say two things about what I’m calling a “working with” model. The first is that a lot of schools try to do something along those lines — building community, working on conflict resolution, implementing peer mediation programs, and so on — but they’re still “doing to” kids at the same time, relying on punitive interventions like time-outs, detentions, suspensions, etc. What they’re doing, I fear, is taking away with one hand what they’re giving with the other.

My other observation is that rewards (or “positive reinforcement”) isn’t an alternative to punishment; it’s just the other side of the same doing-to coin. That includes the reward of adult approval. Trying to win the favor of a teacher is still an extrinsic reason for treating other people decently or exploring ideas, and it’s no more likely than other rewards or punishments to foster a genuine commitment to the value or an enjoyment of learning.

All of this leaves me musing about the related question you raised, which has to do with the nature of adult authority. Like you, I believe there’s an important role for adults to play in kids’ education, and it’s not a passive one by any means. (Educational authoritarians and libertarians — Our Lady of the Fiercely Snapping Ruler and “free” schools like Sudbury Valley — paradoxically share the assumption that adult authority is based on power; they just disagree about whether that’s a good thing.) But it’s not just a question of whether teachers or administrators have authority; the question is what kind of authority we’re talking about. Is it based on the adult’s position and relative power? Or is it authority that emerges from the quality of the adult’s thinking, her wisdom, her deep concern for students’ well-being?

Wouldn’t you agree that we want kids to respect adults because their judgment is good rather than because they have the power to hurt kids who don’t obey them? It’s sort of like wanting kids to look up at adults metaphorically, not just literally — because the grown-ups are impressive and not because they’re tall. The more we want the former kind of authority, the more we need to steer clear of punishment and the more we should, as a teacher once explained his approach to me, “be in control of putting the kids in control.” . . . .

I share your rage at — and an impulse to punish — the bankers who created so much misery [in 2008-09]. But even if such punishment really did have a constructive effect (other than our own temporary feeling of grim satisfaction), I hope we also share the realization that we’re left none the wiser about how to respond to children who do something wrong — which presumably involves something closer to disrupting than to bankrupting. When I was in the classroom, some of my students managed to enrage me from time to time, but I knew my job wasn’t to get even or make myself feel better at their expense. (The adults who worry me most are those who use their power over kids to work out their own psychological issues and then pretend that what they’ve done is in the kids’ interest: “They need to learn that they can’t…”)

Sometimes — often — we’ll need to intervene, as you say, to make sure that one child isn’t making another miserable. But we have to keep reminding ourselves (and our kids, and their parents) that intervention doesn’t have to mean punishment. In fact, punishment is a particularly noxious and counterproductive form of intervention. You say that your “in-between position…creates a conversation about who got hurt, who wasted whose time, and what provoked it.” Such conversations are vital — not just for resolving a given incident but, if handled skillfully, for promoting children’s social and moral development. However, we don’t need punishment for that!

In fact, conversations play out quite differently when they take place in the shadow of punishment. In that context, the primary message kids hear is, “If you do something to displease the people with the power, they’ll make you miserable.” That makes it extremely unlikely that students will reflect on more important questions: How did my action affect other people? What kind of person do I want to be? What kind of classroom (or school) do we want to have? (Research by developmental psychologists has found that punishment — and, again, it doesn’t matter if we call it “consequences” — tends to impede moral development by arresting children at the level of self-interest.)

You and I have both heard people insist that it’s unrealistic to abandon punishment. While it’s certainly true that not all nonpunitive interventions — call them “working with” strategies — are effective, and that no intervention will solve every problem, let’s not judge the alternatives against an impossible standard. Judge them, rather, against the proven failure of the default response: Punishment can never achieve anything but temporary, resentful compliance — and it generates significant problems in the bargain. It doesn’t make the school safer. The wrongdoer doesn’t become a better person — just cleverer about avoiding detection. In fact, it may even make all the other kids, the ones in whose name we’re allegedly taking disciplinary action, feel uneasy because we’re saying, “All of you are members of this community only conditionally. You, too, could be ejected.”

One of the most striking features of any punishment is the way it creates a vicious cycle. No matter how many times we’ve watched as a punitive intervention failed to bring about any improvement (and, more likely, actually made things worse), we may assume that the only possible response is to punish again – perhaps even upping the ante. Interestingly, research psychologist Martin Hoffman found that the worst effects aren’t due to the adult’s initial intervention but to the use of punishment after the child fails to comply with the first request. It’s the reactive use of punishment, the choice to employ it once we’ve already locked horns with the child, that proves most worrisome. Therefore, it’s most important to refrain from punishing precisely when we’re most angry or frustrated.

Ideally, non-punitive intervention strengthens rather than threatens the relationship between the adult and the student who did something wrong. For example, a teacher or principal can express strong disapproval of what the student did to a peer, but then add, as Lilian Katz suggests, “I would never let anyone do something like that to you.” This accomplishes several things at once: It distinguishes between the act and the actor, leaving no doubt that the student is still cared about and still has rights; and it communicates that the act is unacceptable because of its effect on the victim, not because the teacher happens not to like it or because it breaks a rule.

This returns us to one of your favorite topics, Deb: how kids view grown-ups. You wrote, “Students…are eager for strong and powerful and all-knowing adults.” To which I’d reply: I’m not sure that they’re eager for power, per se, or omniscience, so much as for adults who devote their strength and knowledge to supporting, protecting, guiding, and loving. ….

It goes without saying that we’re not going to convince everyone — and, in this culture, we may struggle to convince even a substantial number — of the merit of any of the positions we support. That doesn’t mean we don’t keep trying, of course. But when you ask “If [our views are] as sensible as you and I think, how come the world is still plagued by its obsession with punishment?”, well, I guess we need to remind ourselves that the unpopularity of a viewpoint doesn’t in itself constitute an argument against it.

At the same time, something we believe is wrong doesn’t become right just because we’ve watered it down a little — such as by offering a few open-response items on a high-stakes test, or making sure that children are punished by nice people whom they’ve trusted. I know we may have to settle for a compromise sometimes, but that doesn’t mean the action isn’t still troubling.

Things get more interesting when the moral questions really are unclear, as they may be when, to take your example, Deb, what’s best for one person isn’t best for the group. That’s much too ambitious an issue for a couple of blog posts; it encompasses the limits of utilitarianism, the struggle between collectivist and individualist political models, and a whole lot more. But where classroom discipline is concerned, I can only repeat that a punitive approach seems to benefit neither the individual nor the group — and the negative effects on both are multiplied over the long run. The rest of the class may feel a twinge of relief when an obnoxious kid is booted out or otherwise made to suffer. But a punitive climate isn’t in anyone’s interest. As I argued earlier, it strains the fabric of the relationship between kids and adults, as well as among kids.