Evidence? We Don’t Need No Stinkin’ Evidence!

April 7, 2015

Evidence? We Don’t Need No Stinkin’ Evidence!

By Alfie Kohn

Have you ever suspected that much of what you do for a living is an extended exercise in missing the point?

I’ve spent many years challenging claims about the benefits of rewarding or praising children (when they act the way we want) and punishing them (when they don’t). An impressive body of research shows that neither carrots nor sticks are actually beneficial, particularly in the long run. In fact, both are likely to do more harm than good. Moreover, there’s no evidence to show that the failure to reward kids when they’re successful, or to punish them when they misbehave, produces the dire outcomes we’re often warned about: laziness, aggression, entitlement, and so on.

But here’s the thing: Many people who argue strenuously for dangling rewards in front of children or for threatening them with punishments — and who have harsh words for parents and teachers who deliberately avoid these tactics — are not really making the kind of argument that can be challenged with evidence. Their claims may seem to be descriptive (“Rewards motivate people”) or predictive (“If kids aren’t punished when they act out, their behavior will get worse”), but when you challenge those claims, they just pivot to a declaration that’s purely prescriptive.

Take punishment. When we make children suffer for having acted badly, are we “teaching them a lesson”? Yes. We’re teaching them that people with more power can compel someone with less power to do whatever they demand. We’re teaching that reasons and motives are irrelevant; only observable behaviors matter. We’re teaching that self-interest — the consequence to you, the actor — is the only relevant consideration in deciding how to live your life: Avoid doing stuff that authority figures don’t like — that is, if you think you’re likely to be caught.

I have filled books and articles with evidence showing that punishment isn’t just ineffective; it’s counterproductive. But the reality is that many defenders of get-tough tactics (or get-tough-but-pretend-you’re-being-reasonable tactics, which go by euphemisms like “logical consequences”) don’t seem to care that these tactics don’t work. The point isn’t really to bring about a better outcome, to produce constructive effects on children’s values or future actions, so it doesn’t matter if someone shows that’s unlikely to happen. Rather, the point is to make sure that kids don’t “get away with” something. Children, like adults, must be “held accountable.”

That’s not a practical justification; it’s a moral imperative. So when I stand there clutching my pile of studies about the destructive effects of punishment, I feel a little like the policeman in Arlo Guthrie’s ballad “Alice’s Restaurant” — you remember, the one who showed up in court with “twenty-seven eight-by-ten color glossy photographs with circles and arrows and a paragraph on the back of each one…to be used as evidence against us” — and then watched, aghast, as the judge walked into the courtroom with a seeing-eye dog.

You can point out that punishment never produces anything beyond temporary (resentful) compliance, and it elicits even that at a huge cost. You can explain that the alternative is active problem solving, not permissiveness; the opposite of “doing to” is “working with,” not doing nothing. But you’ll be talking to yourself because, for many people, the real point of imposing what’s euphemistically called a “consequence” is retribution, a primitive version of justice whose premise is If you do something bad, something bad must be done to you. So you might as well put away all your longitudinal studies proving that punishment just makes things worse.


A group of Canadian and American psychologists recently conducted some research that dealt with . . . . the irrelevance of research. They were interested in the philosopher Karl Popper’s distinction between beliefs that are falsifiable (that is, capable of being tested and shown to be incorrect) and those that are unfalsifiable. In the latter case, no matter how much evidence is presented to challenge an assumption, the assumer always find some way to dismiss it so the claim can never be debunked. That’s why, during an argument, I often ask — and am prepared to answer — this question: What would have to be true in order for you to change your mind?

In a series of four experiments reported in the March 2015 issue of the Journal of Personality and Social Psychology, Justin Friesen, Troy Campbell, and Aaron Kay found that a lot of people, upon encountering facts that contradict a political or religious belief they hold, don’t “revise [their] belief to be more in line with the new information” but instead flick away the facts by reframing the issue as a moral one. And the more that people’s convictions are threatened, the more likely they are to rely on unfalsifiable beliefs.

I spent much of my most recent book, The Myth of the Spoiled Child, grappling with the implications of this phenomenon for common convictions about children. For example, scathing attacks on “helicopter parenting” — particularly of teenagers and young adults — are typically framed in empirical terms: “Parents who keep running interference may think they’re helping, but they’re actually making things worse because their kids never learn to solve their own problems.”

The available evidence, however, fails to show that overparenting is particularly common, or, when it does occur, that it has the effect of making children less capable or more dependent. Indeed, we now have data to suggest that close contact with (and even intervention by) parents may be positively beneficial for young-adult children in many cases. In my experience, however, this revelation fails to give critics pause. Their outrage, I’ve come to suspect, may have been more prescriptive than predictive from the beginning: Never mind what the studies find. Parents ought to push kids to become independent as soon as possible. When it comes down to it, these critics are just uncomfortable with, or personally offended by, what they think is excessive parental intervention, and data won’t make those feelings go away.

Exactly the same is true of the furious opposition one encounters to those who object to activities like dodge ball (in which children are turned into human targets), awards assemblies (in which kids are pitted against their peers for artificially scarce recognition), or the practice of giving zeroes to students (which pulls down their final grade disproportionately and irreversibly). The same is true when everyone on a playing field is given a “thanks for playing” trophy. Competition must not be diluted! A sharp line must be drawn between winners and losers, and the latter must not receive something that even looks like an award. Why? Because “life” is a series of contests and children had better get used to the misery of losing right now.

Again, I try to marshal the evidence that these premises are mistaken. Competition isn’t necessary to promote excellence and often holds people back from doing their best. Exposing children to unpleasant experiences — rubbing their noses in their public failure — is not a constructive way to get them ready for the possibility that they’ll encounter more unpleasant experiences when they’re older. And there is no evidence whatsoever to suggest that children who receive recognition just for playing, or support just for trying, will develop unrealistic expectations, a sense of entitlement, or a sudden lack of interest in doing well.

And, again, none of these findings seems to matter. “Losers shouldn’t get trophies! For Pete’s sake, they lost! They’re supposed to go home empty-handed!” Empirical findings are beside the point. It took me years to figure out that this isn’t about psychology; it’s about ideology. Hence the sneering sarcasm about “precious snowflakes” and “tender self-esteem,” the white-hot anger over the possibility that kids will get off too easy or feel good about themselves without having earned that right.

The economist Paul Krugman once pointed out that “the great divide in our politics isn’t really about pragmatic issues, about which policies work best”; it’s about differences in conceptions of morality and justice. So, too, evidently, for disputes involving children.

To be notified whenever a new article or blog is posted on this site, please enter your e-mail address at www.alfiekohn.org/contact-us.