From the archives: The original post was posted on 4th October, 1996
I read Punished by Rewards by Alfie Kohn and thought it was very good. Of course I was hearing what I wanted to hear.
Oh dear. I really didn’t like it at all! Yes okay, his conclusion is true. But he seems to me to be using the perceived authority of science to convince the reader, when in fact the ‘evidence’ is more pseudo-science than science. The criticisms I have made in the past of the pseudo-scientific empirical studies that (for instance) ‘show’ television to be harmful apply to this ‘evidence’ too. It seems to me a really bad idea to back up our ideas with ‘evidence’ that just isn’t what it purports to be. It is dishonest.
He’s not completely non-coercive, but he gives the theoretical underpinnings
No, that is precisely what he doesn’t give! He makes assertion after assertion, but appears to have no coherent theory at all. There are lots of headings that look promising — for example, “Four Reasons Rewards Fail,” “The Reason For the Effect,” and “Why Behaviourism Doesn’t Work in the Classroom” — but read the text and all you’ll find is statements like “extrinsic rewards reduce intrinsic motivation” and “rewards reduce performance because they reduce internal interest.” Well yes, fine, we agree, but the question is why? How? He never tells us. He hasn’t got an explanatory theory. In this respect, Kohn’s book reminds me a little of John Holt’s writing. Except that John Holt never purported to be presenting scientific evidence. That is why I recommend John Holt (with some reservations) but I cannot in good conscience recommend this book, despite the fact that many people taking children seriously love it.
Perhaps worse, Kohn seems to accept uncritically the conventional criteria of what it means for an educational technique to “work,” and what “performance” is. That has the tacit implication that if one of these vile techniques ever was “scientifically” shown to “work” by increasing children’s “performance,” he would start to advocate it.
Kohn’s book can be summed up in one sentence, which includes the subtitle: “The trouble with gold stars, incentive plans, A’s, praise, and other bribes… is that Scientific Evidence Shows that they have bad effects on performance.” The problem is, when you take out the pseudo-scientific ‘evidence,’ there is not much left to read (and what there is has been posted by ————————!). I think Kohn has a gut feeling that behaviourist dog training techniques are bad, and he is quite right about that. But he has no explanation of why they are and how they are. All he has is (worthless) ‘evidence’ that they are.
I am saddened to see those whose views I share adopting the same dishonest tactics as those whose purpose is to deceive the reader into thinking that there is scientific evidence for views opposing ours. This is not an innocent mistake. Appealing to the perceived authority of science is a way of avoiding criticism that could in fact address the issue. That is why he never bothers to explain his view or argue the case. Why should he? He has the backing of the Authority of Science. Who would question that?
- Taking Children Seriously: a new view of children (Oxford Karl Popper Society talk)
- It is impossible to control for all the variables in any experiment involving human psychology
- Isn’t taking children seriously a risky experiment with children? Is there any evidence that it works? Has it been studied?
Sarah Fitz-Claridge, 1996, ‘Punished by rewards’, https://takingchildrenseriously.com/punished-by-rewards/