he has 5 reasons. i'm actually only going to discuss the reasons that spoke to me.
3) rewards ignore reasons.
they encourage the rewarder to take the easy route instead of addressing underlying issues or working harder to make things more interesting or more in line with the nature of the child. maybe the work is too hard or too boring. maybe a different approach would be better.
4) rewards discourage risk taking
i am finding this to be something i hadn't thought about, and quite intriguing. what happens is, the person does exactly what is necessary to get the reard and no more. it puts students into thinking inside a box, rather than thinking outside the box. the objective is not to suceed at the task, the objective is to get the reward. he cites studies that preschoolers, grade schoolers, and adults working for rewards try to avoid anything challenging.
the lesson is that school is not about playing with ideas or taking intellectual risks; it is about doing what is necessary, and only what is necessary, to snag a better letter or number. most students will quickly accommodate us, choosing "to do that which will maximize the grade and not attempting tasks in which they might fail, even though they would choose to challenge themselves to a greater degree under other circumstances."
if anything deserves to be called natural, it is the tendency to seek optimal challenge, to struggle to make sense of the world, to fool around with unfamiliar ideas. human beings are inclined to push themselves to succeed at something (moderately) difficult.
and 5) perhaps the most deadly and i'm actually noticing that it is true: rewards cut the interest rate. he quotes studies that people's interest in what they are doing typically declines when they are rewarded for doing it. he has an entire chapter devoted to this. extremely convincing. if there is request, i will sum up some of his arguments.