Punished by Rewards: The
Trouble with Gold Stars, Incentive Plans, A’s, Praise and Other Bribes by Alfie Kohn
Alfie Kohn argues the more you
reward, the more the person rewarded loses interest in what we reward for. He claims, not only will they be less likely to perform
an act, but they will not be a caring person. Intrinsic motivation
matters. Learning succeeds, he argues, when constructed around "concepts"--whatever that means. Kohn thinks facts impair construction. He makes the good point that self-worth based
on beating others stinks.
Interest, obviously, also matters for success and
memory. If what we read interests us, we are many times more likely
to remember it.
The extrinsic and intrinsic motivation
debate, however, misses the point that the more important considerations for
performing actions are not hedonic but what reasoning takes place about
relevant standards and consequences. And the results of actions. Intrinsic motivation is jargonese for instantly fun, and instantly fun is not the primary criterion for judging an
education or any important endeavor.
Kohn's works spend much time
redefining other individuals' terms in a bad light, but his
definitions are inadequate.
Kohn fails to distinguish between reward meaning
merited desert and reward meaning a carrot-and-stick used to motivate. A
skeptic might claim this book is a course in behavior modification using
lofty rhetoric, rhetorical definitions, shaming attempts, questionable research
and straw person attacks as stimuli. Responses and incentives are
sometimes intertwined in ways that cannot or should not be separated. If worker
A works harder and is more productive than worker B and worker B refuses to be
more productive, they should not be paid the same salary. The other solutions obscure the facts from worker A or "bribe" worker A with
faulty equality rhetoric.
Kohn argues that choice, content
(personal interest and value) and collaboration should be the three important C's of
motivation--a splendid three C's--yet more exists to the story.
Many work places possess the three C's in spades, yet little
good work gets done. This work seems oblivious to defection incentives and
other incentives that may not show up in contrived, short-term experiments but
show up in the real world. Over time, individuals who realize they can not be fired
or punished gradually perform worse, becoming more callous and thoughtless. Kohn uses the
"studies show" phrase often, but he does not provide helpful details.
Elsewhere Pierce, Cameron and Eisenberger call the
punished by rewards theory "cognitive evaluation theory." Cognitive
evaluation theory claims rewards always or almost always reduce intrinsic
motivation. In addition, rewards always or almost always reduce perceived autonomy and
willingness to engage in an action in the future. Cognitive evaluation theory
concludes that rewards should almost never be used.
Pierce, Cameron and Eisenberger counter
with the "general interest theory." Human interests, they claim, vary
for an enormous variety of reasons, including novelty, masochism, fantasies, self-talks,
comfortable habits, feeling efficient, feeling in control, genetic influences,
identification with others, identification with objects, and so on over time
and with situations. Merit pay and unexpected gifts, they argue, increase intrinsic
motivation. Merit pay increases the feeling of self-determination.
Pierce and friends note
that the bad effects of rewards depend on the situation and can be avoided.
When we provide rewards merely for engaging in a task, intrinsic motivation declines and feelings related to lack of control and learned helplessness result. Rewards for completing a task or meeting a standard can increase intrinsic
motivation, however, because self-perceptions of competency and self-efficacy increase.
Rewards for meeting standards do not undermine intrinsic motivation and
autonomy because individuals have the choice of meeting the standard or pursuing
some other activity. Responses to rewards also depend on differences among
recipients (especially beliefs), the details of the reward and the method of administration.
Arguing that bad effects from many
rewards are often temporary, Pierce and friends claim rewards for reading may reduce interest immediately after giving the reward, but they suggest the rewards have no
long-term effects on interest. The individual, for example, may temporarily
tire of an activity. She may lack interest because the activity lacks value.
Negative contrast occurs, a temporary negative reaction to getting no more
reward, which leads to less interest.
Motivation also depends on beliefs. Receiving
half the top merit pay may increase or decrease motivation depending on whether
we get inspired, discouraged, ticked off, or think we got what we deserved.
Pierce and colleagues propose "learned
industriousness theory." Effort is often no fun. If we reward individuals for pouring on the effort, the effort
becomes less unpleasant or even fun. High effort then seems less unpleasant in the future
and becomes habitual. When we reward individuals for low-effort, learned
laziness results. Future work then seems more odious and becomes less likely.
Research suggests rewards decrease the
intrinsic motivation to pursue stupefying, worthless activities, but why should
we care? Stupefying, worthless activities should not be pursued anyway.
Even if negative effects of rewards
were long lasting, the intrinsic motivation crowd gives little consideration to
expected value. Suppose an ill-informed civics teacher uses a no rewards, no punishments
method. Pretend that 90 percent of her students develop interest in civics
as a result. Now suppose a second teacher uses rewards and punishments to rigorously
teach civics and only 60 percent of his students leave class with an interest in
civics. The first teacher may not be better. Students from the first class may
emerge enthusiastic but clueless, with heads full of garbage.
They may be motivated to learn civics as self-help or political
If the second teacher is a rigorous
teacher of important knowledge, all his students received good
ideas. Most will have some wisdom, even students disliking civics. And the motivated 60
percent may pursue a high quality citizenship, and
even some not interested students may pursue civics out of moral duty. The
second teacher would be far superior. And that assumes all the punished by rewards
stuff is true. If Pierce, Cameron and Eisenberger are correct, the intrinsic
motivation advantages enjoyed by the first teacher are bunk. The students from the first teacher
may end up arrogant or unmotivated over the
Kohn's argument implies that starting
anxieties and achievement anxieties are great wrongs. But wholehearted
effort makes the unpleasantness of starting work fade. We may develop
new thoughts and passions we never thought possible. After a while, many
things can become fun. Many important things in life take time and effort before they
become fun. Learning to find gratification in not so easily pleasurable tasks
is worth learning. The ability to delay gratification is worth
Kohn argues that motivation should
almost always be intrinsic and that values should be self-constructed. Unfortunately,
he expects individuals to do their constructing out of almost nothing other than
inaccurate rhetoric. If individuals fail to learn important principles of
moral reasoning, then fail to go through multitudes of arguments and alternatives, disaster
results. Expecting individuals to construct good values out of what
constructivists offer is almost as doomed as expecting students to construct equations of quantum mechanics from the theory of phlogiston.
Kohn blasts grading. Grades have flaws, but perhaps grades are the best of the bad available alternatives. Grades can be useful to indicate academic performance and to show where improvement would help or to suggest a different course of study would be a better use of our energies. It is wrong for someone who works less to obtain the same academic regard as someone who works harder. If a person is indolent and we give them the same credentials as someone performing much better, the underperformer uses those credentials to obtain jobs, a reward--an undeserved reward. Undeserved credentials put individuals in positions where their lack of skill can harm others. Problems with grades may result from the problem of attaching too much value to educational credentials from corrupt institutions, then attaching human worth to educational credentials. Maybe dropping out of school should be called dropping into life. Worth browsing.
398p (H) 1993
—Book review article by
JT Fournier, last updated July 11, 2009