Criticizing (Common Criticisms of) Praise (##)

February 3, 2012

Criticizing (Common Criticisms of) Praise

By Alfie Kohn

Over the last few years I’ve had the odd experience of seeing my work cited with approval by people whose views on the issue in question are diametrically opposed to my own. The issue I have in mind is praise. I’m troubled by it, as are the people who quote me, but for very different reasons. So I thought I’d try to set the record straight even if the result is that I antagonize some folks who seem to regard me as an ally.

I explained my concerns about praise — and outlined alternatives to it — in two books (Punished by Rewards and Unconditional Parenting) and in an article called “Five Reasons to Stop Saying ‘Good Job!'”, so I’ll quickly summarize my arguments here rather than trying to lay them out in sufficient detail to convince a skeptic.

Praise is a verbal reward, often doled out in an effort to change someone’s behavior, typically someone with less power. More to the point, it’s likely to be experienced as controlling regardless of the praiser’s intention. Praise is a pat on the head, “pat” being short for “patronizing,” that’s offered when the child (or student or employee) impresses or pleases the parent (or teacher or manager). Like other forms of reward (or punishment), it’s a way of “doing to,” rather than “working with,” people. My value judgment is that the latter is more respectful and therefore preferable to the former.

Value judgments aside, though, praise has very real and unfortunate effects — again, just like other types of rewards. It tends to reduce the recipient’s interest in the task, or commitment to the action, that elicited the praise. Often it also reduces the quality of whatever was done. The effect of a “Good job!” is to devalue the activity itself — reading, drawing, helping — which comes to be seen as a mere means to an end, the end being to receive that expression of approval. If approval isn’t forthcoming next time, the desire to read, draw, or help is likely to diminish. Praise isn’t feedback (which is purely informational); it’s a judgment — and positive judgments are ultimately no more constructive than negative ones.

Some years after laying out these concerns, I came to realize that praise was troubling in yet another way: It signals conditional acceptance. Children learn that they’re valued — and, by implication, valuable — only when they live up to the standards of a powerful other. Attention, acknowledgment, and approval must be earned by doing a “job” that someone else decides is “good.” Thus, positive reinforcement is not only different from, but antithetical to, the unconditional care that children need: to be loved just for who they are, not for what they do. It’s no surprise that this strategy was designed to elicit certain behaviors rather than to promote children’s psychological health.

That’s the basic critique. Now allow me to point out what it isn’t.

1. It’s not an argument for praising less frequently. The problem isn’t with how often it’s done but with the nature of a verbal reward — how it’s intended and especially how it’s construed.

2. It’s not an argument for offering more meaningful praise — as distinguished from the “empty” kind. Quite the opposite, in fact. Yes, some teachers and parents reflexively hand out the equivalent of a doggie biscuit every few minutes, the result being that kids habituate to it and it has no impact. If so, good! It may be a waste of breath, but at least it’s not doing too much damage. The kind of praise that’s rationed and carefully timed for maximum impact is more manipulative and more harmful.

3. It’s not an argument for praising people’s effort rather than their ability. That distinction, which has attracted considerable attention over the last few years, is derived from the work of Carol Dweck. I have been greatly impressed and influenced by Dweck’s broader argument, which spells out the negative effects of leading people to attribute success (or failure) to their intelligence (or its absence). Intelligence, like other abilities, is often regarded as innate and fixed: You either got it, or you ain’t.

But the critical distinction between effort and ability doesn’t map neatly onto the question of praise. First of all, while it’s impossible to dispute Dweck’s well-substantiated contention that praising kids for being smart is counterproductive, praising them for the effort they’ve made can also backfire: It may communicate that they’re really not very capable and therefore unlikely to succeed at future tasks. (If you’re complimenting me just for trying hard, it must be because I’m a loser.) At least three studies have supported exactly this concern.

Second, the more attention we give to the problems of ability-focused praise in particular, the more we’re creating the misleading impression that praise in general is harmless or even desirable. Of the various problems I’ve laid out — its status as an extrinsic inducement and a mechanism of control, its message of conditional acceptance, its detrimental effects on intrinsic motivation and achievement — none is limited to the times when we praise someone’s ability. In fact, I’m not convinced that this type is any worse than other praise with respect to these deeper issues.

Third, to the extent that we want to teach the importance of making an effort — the point being that people have some control over their future accomplishments — praise really isn’t required at all. (Dweck readily conceded this in a conversation we had some years ago. Indeed, she didn’t seem particularly attached to praise as a strategy and she willingly acknowledged its potential pitfalls.) It would be a useful exercise, for an individual teacher or as a staff development activity, to figure out how we might be leading students to conclude that failing at a task means they just don’t have what it takes. What policies, and what approaches to assessment in particular, might incline someone to think that ability, as opposed to effort, makes the difference?

4. Most of all, this is not an argument that praise is objectionable because we’re spoiling our kids, overcelebrating their accomplishments and convincing them that they’re more talented than they really are. If you have read any article critical of praise in the last two decades, it has probably proceeded from this premise, which represents a form of social conservatism widely shared even by political liberals. Here, praise is seen as just one more symptom of a culture of overindulgence, right alongside grade inflation, helicopter parenting, excessive focus on self-esteem, and the practice of handing out trophies to all the participants.

Microsoft Word lacks a font sufficiently bold to emphasize how starkly this sensibility — and this reason for opposing praise — differs from my own. In fact, I’m so troubled by the values underlying this critique and its mistaken empirical assumptions (about child development, learning, and the psychology of motivation) that I may write have written a book on the subject. You can imagine my reaction, then, when people who think along these lines invoke something I’ve written about praise to help make their case.

Some of these people wax indignant that children are praised — and consequently come to expect praise — for doing things that they ought to do just because they’ve been told to do them. This old-school argument for unquestioning (and unrewarded) obedience contrasts sharply with my claim that praise is more likely to function as a tool for imposing our will and eliciting compliance. Like much of what is called “overparenting,” praise doesn’t signify permissiveness or excessive encouragement; to the contrary, it is an exercise in (sugar-coated) control. It is an extension of the old-school model of families, schools, and workplaces — yet, remarkably, most of the criticisms of praise you’re likely to read assume that it’s a departure from the old school, and that that’s a bad thing.

Praise is typically faulted for being given out too readily (see point #2, above), with the bar having been set too low. We’re told that kids should do more to deserve each “Good job!” they get — which is a way of saying it should be more conditional. Again, this is exactly the opposite of my objection to the conditionality inherent in rewards. The problem isn’t that kids expect praise for everything they do. The problem is with our need for control, our penchant for placing conditions on our love, and our continued reliance on the long-discredited premises of behaviorism.

You may not be persuaded by my critical analysis of praise and the assumptions that underlie its use. Just don’t confuse it with criticisms that reflect an entirely different set of values.

Back to Blogs