by Alfie Kohn
SELF-CONTROLLED TO A FAULT
Much might be made of the mainstream media’s many misinterpretations of Mischel’s marshmallows. Those misreadings are likely related to the value the writer places on having children defer gratification, which, in turn, reflects widely shared and largely uncritical support for the idea of self-discipline in general. So let’s turn our attention to that concept.
The party line—from therapists, journalists, educators, and parenting advisers—is that discipline should give way to self-discipline, that control from within is better than control from without, that kids should internalize a desire to succeed and (what we regard as) good values. But there’s much more to the story here, psychologically, philosophically, and even politically. If we’re interested in what’s healthiest for kids rather than just in getting them to follow directions, then we will likely see the situation as more complicated. Is it useful to be able to persevere at worthwhile tasks? Undoubtedly. Do some children seem less able than others to do so? Again, yes. But it’s not at all clear that self-discipline should enjoy a privileged status compared to other attributes. In some contexts, it may not be desirable at all.
If that proposition seems surprising, it may be because self-discipline (or deferral of gratification) is usually described as an “ability” or “skill.” The question is how proficient you are at making yourself get to work, or at resisting temptation. Some abilities and skills are more useful than others, but we don’t tend to think of any of them as bad things to have, or something of which you could have too much.
Several decades ago, however, the late Jack Block, an eminent research psychologist, offered a different perspective. Self-regulation may not be a skill, like having a good sense of direction, so much as “an orientation toward motivational expression”—a psychological tendency, if you will—like being introverted or extroverted. He proposed that people can be described in terms of their level of “ego control,” which means the extent to which impulses and feelings are either expressed or suppressed. Those who are undercontrolled are impulsive and distractible, and children who fit that profile are somewhat more likely to suffer from health problems and financial difficulties when they grow up, as well as to have committed a crime.19 But other individuals are overcontrolled. Their actions seem to be compelled rather than freely chosen, and they often seem joyless.
Block made two complementary points. First, a lack of self-control isn’t always a bad thing because it may “provide the basis for spontaneity, flexibility, expressions of interpersonal warmth, openness to experience, and creativ[ity].” Second, too much self-control is as worrisome as too little, even though parents and teachers tend to be more irritated by undercontrol in children and thus more likely to define the latter as a problem. “The idea of self-control is generally praised,” Block observed, but we should be careful not to endorse “the replacement of unbridled impulsivity with categorical, pervasive, rigid impulse control.” As long as you get your work done and don’t make trouble, people in positions of authority don’t care if you’re “rigid, unexpressive, routinized, and flattened in affect.” But that’s just not an ideal way to live.20
When Block said that self-control is “generally praised” in our society, he was putting it mildly. Those who write on the subject rarely if ever consider the possibility that it can be overdone. In addition to Lehrer’s article about Mischel’s experiments, we might add a popular book called How Children Succeed by journalist Paul Tough (who argued in an earlier essay that adults should organize and plan children’s play for them in an effort to help them acquire more self-control). Two neuroscientists, meanwhile, declared in the New York Times, “All [children] can benefit from building self-control.” And a typical discussion of the topic in an education journal included the assertion that “the promotion of self-discipline is an important goal for all schools.”21 Given this consensus, we shouldn’t be shocked to learn that there is now a self-help program called Self-Discipline in 10 Days for those who lack the, um, self-discipline to develop this attribute on their own (or the ability to defer gratification for more than a week and a half).22
A dose of skepticism would seem to be called for here, if only because this enthusiasm seems every bit as intemperate as those declarations a couple of decades ago that self-esteem can serve as a social vaccine. (Of course the latter claims were scorned by some of the same people who now gush about the wonders of self-control.) But the problem isn’t just a one-sided picture of self-discipline—it’s a failure to appreciate the relevance of the circumstances in which we find ourselves. “Disciplined and directed behavior, which can be advantageous in some situations,” said Block and his colleagues, “is likely to be detrimental” in others.23
That’s similar to a point Mischel made about delaying gratification, years after his experiments were concluded. We’ve already heard him say that whether it’s smart to wait for a larger reward depends on the context. But he also pointed out that some people tend to be overly inclined to wait. While the inability to delay may be a problem, he said, “the other extreme—excessive delay of gratification—also has its personal costs and can be disadvantageous. . . . Whether one should or should not delay gratification or ‘exercise the will’ in any particular choice is often anything but self-evident.”24
This may sound obvious: Naturally one can go too far in either direction—with deferral of gratification or self-discipline in general. Yet some social scientists have explicitly disputed this claim. Martin Seligman and a colleague wrote, “Our belief [is] that there is no true disadvantage of having too much self-control.”25 And a group of researchers that included Roy Baumeister professed to have data demonstrating that self-control is “beneficial and adaptive in a linear fashion. We found no evidence that any psychological problems are linked to high self-control.”26
That assertion turns out to be rather misleading for two reasons. First, it’s based on the researchers’ having found an inverse relationship between self-control and negative emotions (among undergraduates who filled out a questionnaire). Other research, however, has found a similar inverse relationship between self-control and positive emotions.27 Even if highly self-controlled people aren’t always unhappy, they’re also not particularly happy; their emotional life tends to be muted.
Second, Block and his associates noticed that there was something fishy about the questionnaire used by this group of researchers. It included questions “reflective of an appropriate level of control and [of] undercontrol, but not overcontrol. It is therefore not surprising that the correlates of the scale do not indicate maladaptive consequences associated with very high levels of control.” In other words, the clean bill of health they awarded to self-control was virtually predetermined by the design of their study.28
Overall, the data do indeed support Block’s balanced view that “although it may be psychologically undesirable to be extremely impulsive, it is also psychologically undesirable to be extremely controlled”:29
•A high degree of self-control tends to be associated with less spontaneity and a blander emotional life.
•Preschool children who seemed overcontrolled were likely to be conventional, moralistic, and uncomfortable with uncertainty when they were young adults.
•Highly self-controlled teenagers were likely never to have used drugs, but they were also less well-adjusted overall than those who had “lower ego control and may have experimented briefly with drugs.”
•“A tendency toward overcontrol puts young women (but not young men) at risk for the development of depression.”
•A preoccupation with self-control is a key feature of anorexia.
•When there’s plenty of time, impulsive people don’t perform as well as self-controlled people on certain tasks, but those results are reversed when decisions must be made quickly.30
The more you explore the dynamics of self-discipline, the more you come to understand why it isn’t always productive or healthy. Consider a stud
ent who always starts her homework the moment she walks in the door. Is this an admirable display of “effortful self-control”31—of what can be achieved by force of will given that there are other things this girl would rather be doing? Or might it reflect her acute discomfort with having anything unfinished? It’s possible that she wants—or, more accurately, needs—to get the assignment out of the way in order to stave off anxiety.32
Self-discipline can be less a sign of health than of vulnerability. It may suggest a fear of being overwhelmed by external forces, or by one’s own desires, that must be suppressed through continual effort. One might say that such individuals suffer from a fear of being out of control. Half a century ago, the psychologist David Shapiro described how someone can function as “his own overseer, issuing commands, directives, reminders, warnings, and admonitions concerning not only what is to be done and what is not to be done, but also what is to be wanted, felt, and even thought.” Secure, healthy people can be playful, flexible, open to new experiences and self-discovery, deriving satisfaction from the process rather than always being focused on the product. An extremely self-disciplined individual, by contrast, may not “feel comfortable with any activity that lacks an aim or a purpose beyond its own pleasure, and usually . . . [does] not recognize the possibility of finding life satisfying without a continuous sense of purpose and effort.”33
A couple of interesting paradoxes follow from this analysis. One is that while self-discipline implies an exercise of the will, and therefore a free choice, many such people are actually not free at all, psychologically speaking. It’s not that they’ve decided to discipline themselves so much as that they can’t allow themselves to be undisciplined. The same is true of deferral of gratification: When Block and a colleague conducted an experiment in which teenagers had to choose between getting paid a certain amount now or more money later, those who waited “were not just ‘better’ at self-control, but in a sense . . . seemed to be unable to avoid it.”34
A second paradox is that impressive self-discipline may contain the seeds of its own undoing: an explosive failure of control, which psychologists call “disinhibition.” From one unhealthy extreme (even if it’s not always recognized as such), people may suddenly find themselves at the other: The compliant citizen abruptly acts out in appalling fashion; the pious teetotaler goes on a dangerous drinking binge; absolute abstinence gives way to reckless, unprotected sex.35 Moreover, making an effort to inhibit potentially undesirable behaviors can have other negative effects. A detailed review of research concerning all sorts of attempts to suppress one’s feelings and behaviors concludes that the results of self-control often include “discomfort or distress” and “cognitive disruption (including distractibility and intrusive, obsessive thoughts about the proscribed behavior).”36
In short, we shouldn’t always be reassured to learn that children are remarkably self-disciplined. The same caution is appropriate regarding those who are inclined to delay gratification: Delayers “tend to be somewhat overcontrolled and unnecessarily inhibited.”37 Likewise for those who always persist at a task, even when they’re unsuccessful. (I’ll have more to say about the last of these tendencies, commonly romanticized as tenacity or “grit.”) On the other hand, self-discipline obviously isn’t always pathological. So what distinguishes the healthy and adaptive kind? Moderation, perhaps, but also flexibility. What counts is the capacity to decide whether and when to persevere—or to control oneself, delay pleasure, or follow the rules—rather than the simple tendency to do these things. It’s this capacity, rather than self-discipline or self-control itself, that children would benefit from acquiring.38
CONTROLLED FROM WITHIN
What passes for discussions of self-discipline tends to resemble unreflective cheerleading for the concept. The result is that we may not even stop to consider that too much of it can be as unhealthy as too little. But I’d like to push beyond this point, and in the process provide a fuller answer to that question about distinguishing the good kind from the bad. What we really need to ask about self-discipline, just as with love or self-esteem, is not merely “How much?” but “What kind?”
One of the most fruitful ways of thinking about this issue emerges from the work of psychologists Edward Deci and Richard Ryan. I’ve already drawn from their theory and research in reconsidering the casual way people tend to treat motivation as if it were a single substance that’s possessed in a certain quantity. We want kids to have more of it, so we try to “motivate” them. In fact, though, as I explained in Chapter 4, we’d do better to think in terms of different types of motivation—intrinsic and extrinsic. And the type matters more than the amount. Intrinsic motivation, you’ll recall, means wanting to do something for its own sake, whereas extrinsic motivation concerns a reward or punishment that a behavior may elicit. Furthermore, these two varieties aren’t just different; they’re often inversely related. The more that people are led to focus on an extrinsic reason for doing something, the more likely that their interest in the task itself will diminish.
Here’s the puzzle, though: Children do some things that aren’t intrinsically appealing, and they sometimes do them in the absence of rewards. Why? We might say they’ve internalized a commitment to doing them. This, of course, brings us right back to the idea of self-discipline, which is where many parents and educators have placed their bets. As I noted at the beginning of this chapter, they want kids to comply with their expectations without having to stand next to them, carrots and sticks at the ready. They want kids to follow the straight and narrow, even when no one is watching.
Deci and Ryan, however, are not finished complicating our lives. Having shown that there are different kinds of motivation (which are not equally desirable), they go on to suggest that there are also different kinds of internalization (where exactly the same thing may be true). This is a possibility that few people, including researchers, seem to have considered. Even someone who’s aware of the difference between intrinsic and extrinsic may just insist that children should be helped to “internalize good values (or behaviors)” and leave it at that.
But what exactly does that internalization look like? On the one hand, someone else’s rule or standard can be swallowed whole, or “introjected,” so it controls children from the inside: “Behaviors are performed because one ‘should’ do them, or because not doing so might engender anxiety, guilt, or loss of esteem.” On the other hand, internalization can take place more authentically, so the behavior is experienced as “volitional or self-determined.” It has been accepted—fully integrated into one’s value structure—and feels chosen.39 A child, for example, may study because she knows she’s supposed to do so and will feel lousy about herself if she doesn’t. Or she may understand the benefits of doing so and want to follow through, even if it’s not always pleasurable.
Controlling adults end up promoting the former (introjected) approach, and that often results in a style of learning that’s rigid, superficial, narrow, and ultimately less successful (if the assignment involves anything more ambitious than rote memorization).40 This same basic distinction between types of internalization has proved relevant not only to academics, but to sports, romantic love, generosity, political involvement, and religion—with research in each case, much of it done by Deci, Ryan, and their associates, demonstrating that the integrated kind of internalization (in terms of the way one pursues any of those activities) leads to better outcomes than the introjected kind.41
The upshot is that just because motivation is internal doesn’t mean it’s intrinsic or integrated or ideal. If a child feels controlled, he’s still at the mercy of rewards and punishments. It’s just that now they live inside him. To do one thing is to be rewarded (temporarily) with good feelings about himself; to do another thing is to feel lousy. What drives this form of internalization, in other words, is a concept we’ve already encountered: conditional self-esteem. And the result is that such children are likely to be conflicted, unhappy, and typically l
ess likely to succeed—at least by meaningful criteria—at whatever they’re doing.42 Their reason for acting is inside them, and yet it doesn’t feel as though it’s coming from them. They may be suffering from what the psychoanalyst Karen Horney called the “tyranny of the should”—to the point that they no longer even know what they really want or who they really are.
Many older children have internalized just such a compulsion to do well in school. They look like every parent’s dream of a dedicated student, but in reality they may have mortgaged their present lives to the future: noses to the grindstone, perseverant to a fault, stressed to the max. High school is just preparation for college, and college consists of collecting credentials for whatever comes next—year after year of holding out for the possibility of more marshmallows. Nothing right now has any value, or provides any gratification, in itself. These students may be skilled test-takers and grade grubbers and gratification delayers, but they are often motivated by a perpetual need to feel better about themselves rather than by anything resembling curiosity.
By the same token, children who have introjected commands to be polite or dutiful or helpful are not really moral agents in any meaningful sense. They haven’t chosen to do good because they don’t experience themselves as choosing. After all, ensuring that children internalize our values isn’t the same thing as helping them to develop their own. Moreover, that sense of having to act in a certain way plays havoc with their emotions, which they either can’t control (the result being that feelings bubble up dangerously of their own accord) or feel they must suppress.43 Again, notice how this brings us a step beyond Block’s work: It’s not just that some kids regulate themselves too much; it’s that the way they’ve been taught to regulate themselves is unhealthy. It has negative effects on their intellectual, moral, and emotional development.
