Should we Enhance Self-Esteem? more

Philosophica 79 (2007) pp.71-91 SHOULD WE ENHANCE SELF-ESTEEM? Rebecca Roache ABSTRACT The conviction that high self-esteem is beneficial both to the individual and to society in general has been pervasive both in academia and in popular culture. If it is in deed beneficial, it is a prim e candidate for pharmacological enhancem ent. There is evidence to suggest, however , that the b enefits of high self-esteem to the individual have b een exaggerated; and that there are few - if any - socia l benefits. W ith this evid ence in m ind, I consid er in wha t ways high self-esteem is valuable, and suggest how enh ancement could play a role in maximizing its valuable aspects. 1. Introduction The successful use of drugs to improve mood is not particularly novel. Consider the case of Tess, a patient of the American psychiatrist Peter Kramer, whose transformation under Prozac is described in his 1993 book, Listening to Prozac. Tess suffered a deprived childhood plagued with physical and sexual abuse, and was left to raise her nine younger siblings after her alcoholic father died and her mother became clinically depressed. As an adult, her marriage failed and she embarked on a string of affairs with abusive, married men. Her success in her career failed to compensate for the shortcomings of her personal life. Kramer summarises her psychological state as follows: I ran down the list of signs and symptoms [of depression], and she had them all: tears and sadness, absence of hope, inability to experience pleasure, feelings of worthlessness, loss of sleep and appetite, guilty ruminations, poor memory and concentration. W ere 72 REBECCA ROACHE it not for her many obligations, she would have preferred to end her life. (Kramer, 1993: 3) Prescribing Prozac for Tess, however, had startling results: Two weeks after starting Prozac, Tess appeared at the office to say she was no longer feeling weary. In retrospect, she said, she had been depleted of energy for as long as she cou ld remember, had almost not kno wn what it was to feel rested and hopeful. She had been depressed , it now seemed to her, her whole life. She was astonished at the sensation of being free of depression. She looked different, at once more relaxed and energetic – more availab le – than I had seen her, as if the person hinted at in her eyes had taken o ver. Sh e laughed more frequently, and the qua lity of her laughter was different, no longer measured but lively, even teasing. W ith this new demeanour came a new social life, one that did not unfold slowly, as a result of a struggle to integrate disparate parts of the self, but seemed, rather, to reappear instantly and full-blown. (Kramer, 1993: 7) The power of Prozac to effect remarkable improvements in mood is not restricted to people suffering from depression. Kramer reports that, even after his patients had recovered from their mental illness and withdrawn from Prozac, they would sometimes wish to continue using Prozac because, whilst their original symptoms had not returned, Prozac helped them to feel ‘better than well’. As the range of drugs with the power to alter mood in healthy subjects continues to grow, we are likely to see an increasing demand for so-called mood enhancers. What should be our attitude to mood enhancement, both as individuals and as a society? Let us define enhancement as the use of technology (including medicine, techniques, and tools) to improve human capacities or traits above a level that most of us would consider normal. Currently, there exists – or may soon exist – technology to enhance a variety of physical capacities (such as stamina, speed, coordination), cognitive capacities (such as memory, concentration, alertness), and mood states (such as social comfort, happiness, assertiveness). Over the last ten years or so, academic debate about whether or not enhancement is ethically acceptable has burgeoned. Concerns raised about enhancement have included safety, fairness, the possibility of adverse consequences, and the SHOULD WE ENHA NCE SELF -ESTEEM ? 73 question of whether there are acceptable and unacceptable ways to bring about an improvement in our capacities. I shall not discuss any of these issues here. My concern in this paper relates to less frequently debated questions of which capacities it might be worth improving, and what would count as an improvement in certain capacities. Claims about what sort of changes in our capacities would constitute improvements are not always controversial. In the case of changes in some capacities, it is relatively easy to say antecedently whether or not they would be desirable. Broadly speaking, for the subject, increased intelligence is better than decreased intelligence, an increased ability to remember and recall information is better than a decreased ability to remember and recall information, and increased stamina is better than decreased stamina.1 In the case of other capacities, however, deciding whether or not a given change constitutes an improvement is more difficult. Many changes relating to mood fall into this latter category. For example, there is no obvious sense in which – setting aside extreme cases – it is better to be shy rather than confident, or better to be sensitive rather than insensitive to criticism, or better to be introverted rather than extroverted. The difficulties deciding what would count as an improvement in certain dimensions, especially mood states, does not entail that people do not strongly desire to use drugs to effect certain changes in these areas. On the contrary, the key attraction for many of alcohol, tobacco, illegal drugs, and certain prescription drugs is their capacity to affect mood. That people are likely to want to use mood-altering drugs as they become available makes it highly pertinent to address philosophical questions relating to the desirability of the effects such drugs will have, both for the subject and for society. If there are likely to be common misconceptions about which mood states it would be beneficial to target with drugs, it is better that philosophers and scientists uncover them and educate people before the drugs become available, than that people discover to their 1 Som e qua lification is needed here. An inc reased ability to remembe r and recall information is desirable only if the information is valuable; and it could prove unde sirable if one’s ability to recall valuable information were obstructed by memories of irrelevant information. This does not undermine the claim that more memory is better than less; but it does highlight that, in mak ing such a claim, it is tempting to focus selective ly on those aspects of memo ry that we value, and to ignore those aspects that are less valuable. 74 REBECCA ROACHE personal cost, and to society’s cost, that certain drugs do not bring the unmitigated benefits hoped of them. One likely target for mood enhancement, and one that is especially in need of philosophical analysis, is self-esteem. Self-esteem is, roughly, one’s global estimation of one’s own worth.2 Having high self-esteem is pleasant to experience, it is often believed to benefit the subject, and it is often believed to benefit others. If these beliefs are correct, pharmacological enhancement of self-esteem could make people happier and more successful and society more harmonious. The belief that improving people’s self-esteem is desirable is pervasive in both popular and academic circles. In the 1980s, this belief led California Governor George Deukmejian to set up a task force to promote self-esteem. The Task Force had high hopes for self-esteem, as it reported in its 1990 review of the topic: Self-esteem is the likeliest candidate for a social vaccine, something that empow ers us to live responsibly and that inoculates us against the lures of crime, violence, substance abuse, teen pregnancy, child abuse, chronic welfare dependency, and educational failure. The lack of self-esteem is central to most personal and social ills plaguing our state and nation. (California Task Force to Promote Self-Esteem and Personal and Social Responsibility, 1990: 4) So strong was the conviction of the Task Force that self-esteem is beneficial, that its failure to uncover sufficient scientific evidence in support of this assumption led not to its questioning this assumption, but to its recommending funding for further research in this area! This conviction is also strongly held by many people and organisations today. The Task Force disbanded in 1995, but efforts to promote self-esteem continue. The National Association for Self-Esteem, a US non-profit organisation, defines its purpose as ‘to fully integrate self-esteem into the fabric of American society so that every individual, 2 A disc ussion of exactly what self-esteem is, what it has been taken to mean by different people in the past, and how these interpretations relate to various popular and academic views about it, is a topic for another paper. I will not attempt to address it here, and hope that the reader does not find this too frustrating. SHOULD WE ENHA NCE SELF -ESTEEM ? 75 no matter what their age or background, experiences personal worth and happiness’3 , and lists some benefits of nurturing self-esteem as feeling good, increased willingness to take risks, improved relationships with others, and improved academic performance. Nathaniel Branden, a psychologist with a prominent role in the self-esteem movement, has claimed that he ‘cannot think of a single psychological problem – from anxiety and depression, to fear of intimacy or of success, to spouse battery or child molestation – that is not traceable to the problem of low self-esteem’ (Branden, 1994: 12). The conviction that promoting selfesteem is desirable pervades English-speaking Western culture: a Google search on the term ‘self-esteem’ on 25th October 2007 returned 4,080,000 results; and on the first page alone, eight out of the ten results contained – either in their titles or their two-line descriptions – positive evaluations of the importance of self-esteem, and offered techniques to improve one’s own self-esteem or that of one’s children, self-esteem assessment tests, or information that otherwise emphasised the positive value of selfesteem.4 Six out of the eight sponsored links on the first page had titles that either offered techniques for improving self-esteem, or self-esteem assessment tests.5 It was necessary to look at seven pages of results before finding one – the sixty-ninth – which questioned the importance of self-esteem.6 On the same day, a search on the term ‘self-esteem’ within the book titles of the British arm of the online shop, Amazon, returned 6,205 results. All twelve titles on the first page of results (sorted by relevance, the default setting) promised techniques to help raise one’s own self-esteem or that of one’s children. A similar search using the American arm of Amazon returned 93,490 results; and again, all twelve titles on the first page promised to enhance self-esteem. The academic literature reflects this positive view of self-esteem. In 1996, Baumeister, Smart et al. surveyed literature that either associates low self-esteem with violence, or makes the stronger claim that low selfesteem causes violence. They found that low self-esteem has been linked 3 4 See http://www.self-esteem-nase.org/ The other two results offered information about self-esteem, but did not feature any evaluative com ments abo ut it in their titles or two-line descriptions. 5 The other two sponsored links advertised hypnotherapy services and did not mention self-esteem. 6 The result in question was a link to Mathews 2006. 76 REBECCA ROACHE with domestic violence (Renzetti, 1992; Walker, 1979), terrorism (Long, 1990), armed robbery (McDonald, 1975), sibling violence (Wiehe, 1991), murder by adoptees of their adopted fathers (Kirschner, 1992), hate crimes (Levin and McDevitt, 1993), and child abuse (Oates and Forrest, 1985); and that it has been claimed to cause violence among youth gangs (Anderson, 1994; Jankowski, 1991), male violence (Toch, 1969), and increase crime rates among American black people (Schoenfeld, 1988). In a 2003 paper, Baumeister, Campbell et al. conducted a comprehensive review of literature relating to self-esteem, in which high self-esteem has been associated with improved academic performance, job performance, personal relationships, happiness, and health. Are these views about the positive value of self-esteem correct, and what can an assessment of them tell us about the desirability of enhancing self-esteem? I shall address this question by splitting it into three smaller questions. First, is raising self-esteem likely to lead to the benefits that have been claimed for it? Second, if not, what should be our attitude towards raising self-esteem? And third, what are the implications of this for questions about how self-esteem might be enhanced? Let us turn now to the first of these questions. 2. Is Raising Self-Esteem Likely to be Beneficial? Examining the evidence for claims that high self-esteem is beneficial points to the conclusion that such claims are exaggerated. Baumeister, Smart et al. find little to support the traditional, pervasive view that low self-esteem is a cause of violence and aggression. Of the literature they surveyed, they observe: Most studies failed to find any support for [this claim], and many provided clear and direct contradictory findings. Aggressors seem to believe that they are sup erior, ca pab le beings. Signs of low selfesteem, such as self-deprecation, humility, mo desty, and selfeffacing mannerisms, seem to be rare (underrepresented) among violent criminals and other aggressors. (Baumeister et al, 1996: 26) In contrast, Baumeister, Smart et al. argue that, in so far as violence is linked to self-esteem at all, it is most commonly associated with a subset SHOULD WE ENHA NCE SELF -ESTEEM ? 77 of people who hold unrealistically high opinions of themselves, and who respond to challenges to these views (in the form of feedback from others) with violence. Amongst the evidence they surveyed, ... [f]or huge nationalities, medium and small groups, and lone individuals, the same pattern was found: Violence resulted most com mon ly from feeling that one’s superiority was somehow being undermined, jeopardised, or contradicted b y current circumstances. (Baumeister et al, 1996: 26) Baumeister, Smart et al. speculate that two factors may account for the persistence of the view that low self-esteem causes violence. First, it is due to a running-together of aggressors’ self-appraisals and external appraisals: we falsely suppose that violent people internalise negative appraisals from others. In reality, however, ... self-appraisals are only weakly related to external appraisals and ... in many cases people overtly resist revising their self-appraisals in the face of external feedback. (Baumeister et al, 1996: 28)7 Second, Baumeister, Smart et al. suggest that ... a broad reaction against blaming the victim ... may have encouraged writers to phrase the causes of violence in terms of self-evaluation rather than in terms of provocative, evaluative acts by the future victim. (Baumeister et al, 1996: 28) Baumeister, Campbell et al. find that claims that high self-esteem makes life go better in a variety of ways are largely unsubstantiated. They survey a vast body of research on self-esteem, which investigates its links to academic and professional performance, personal relationships, happiness, and health; and conclude that self-esteem is ‘not a major predictor or cause of almost anything’ (Baumeister et al, 2003: 37). Whilst high self-esteem does correlate with certain desirable outcomes – such as improved academic performance – there is no convincing evidence to establish the direction of causation. Therefore, high self7 In sup por t of this claim, they cite Crocker and Major 1 989, Shrauger and Schoeneman 1979, and Swann and Hill 1982. 78 REBECCA ROACHE esteem could well be a result, rather than a cause, of improved performance; or both high self-esteem and improved performance could both result from a third factor. Further, ‘the effects of self-esteem become weaker as the criteria for evidence become more objective’ (Baumeister et al, 2003: 37): whilst those with high self-esteem report that they are better liked and more popular than others amongst their peers, reports from their peers generally fail to confirm this, suggesting that some of the apparent benefits of high self-esteem are illusory. On the whole, Baumeister, Campbell et al. find that the benefits of high selfesteem are limited to increased initiative and – unsurprisingly – pleasant feelings. These benefits to the subject of high self-esteem can be weighed against the costs borne by others: Peo ple high in self-esteem or narcissism are p rone to bully others, to retaliate aggressively, and to be prejudiced against out-group memb ers. Self-enhancers [those who rate themselves more favourably than others rate them, or who rate themselves on average more favourably than others rate themselves] are sometimes annoying or obnoxious to others. They may be willing to cheat and perform other antisocial, self-serving acts (Baumeister et al, 2003: 37). As a result, whilst it may be pleasant to experience high self-esteem, other benefits for either the subject or others are few, and having to live or work with someone with obviously high self-esteem can be difficult: [i]ndeed, the socially disrup tive consequences of ego tism may explain why people tend to be modest and self-effacing when interacting with friends (Tice et al, 1995) or when living in cultures characterised by stable, interdependent relationships (H eine et al, 1999)’ (B aumeister et al, 2003 : 37). With these observations in mind, Baumeister, Campbell et al. advise that self-esteem – especially children’s self-esteem – should not be boosted indiscriminately; rather, boosts to self-esteem should be used as a reward for good behaviour or worthwhile achievements. However, another view sees tying self-esteem boosts to performance in certain areas as counterproductive and personally costly. Crocker and Knight argue that ‘[b]asing self-esteem on external factors SHOULD WE ENHA NCE SELF -ESTEEM ? 79 [those that require validation by others] such as appearance, others’ approval, or academic achievement has more negative consequences than basing it on internal factors such as virtue of God’s love’ (Crocker and Knight, 2005: 201). Pursuing self-esteem in external domains brings short-term emotional benefits when successful, but leaves unsatisfied ‘fundamental human needs for learning, relatedness, and autonomy’. Allowing the pursuit of self-esteem to shape behaviour can become addictive, with greater successes required in order to avoid feelings of worthlessness, and it ‘has costs for learning, relatedness, autonomy, selfregulation, and, over time, physical and mental health’. This is because it leads to a focus on those activities that are likely to bring success at the expense of those activities that carry a higher risk of failure but which may be otherwise beneficial, thus impeding learning and reducing autonomy (Deci et al, 1994); it encourages one to focus on one’s own needs rather than those of others, thus undermining relationships (Crocker and Park 2004); and it is associated with depression (Crocker et al, 2003) and self-destructive behaviour (Crocker, 2002). Because boosts to self-esteem confer short-term emotional benefits but can be costly in the long term, Crocker and Knight view them as ‘analogous to sugar: tasty but not nutritious’ (Crocker and Knight, 2005: 201). They advise against allowing the pursuit of self-esteem to determine one’s goals, and instead advise ‘adopting goals that are good for others as well as the self’ (Crocker and Knight, 2005: 202). Boosting self-esteem, then, is not likely to help realize the benefits often associated with high self-esteem, and could bring unwelcome sideeffects. Given this, there is a need to reassess the value we place on selfesteem. 3. What Should be our Attitude Towards High Self-Esteem? To begin to answer this question, it is important to note – as recognised by Baumeister, Smart et al. and by Baumeister, Campbell et al.– that the category of high self-esteem is heterogeneous. It encompasses not only those people who have what we might call ‘healthy’ high self-esteem – that is, a favourable view of their own worth that accurately reflects their true level of worth, as determined by their character, behaviour, 80 REBECCA ROACHE achievements, and so on8 – but also those who have an unrealistically high estimation of their worth. The latter group includes pathological individuals, such as narcissists and sociopaths. Many of the problems we have considered in association with high self-esteem relate not to healthy high self-esteem, but to this latter category of unrealistically high selfesteem. No doubt few people, even amongst those who enthuse about the benefits of high self-esteem, would welcome the prospect of holding deluded high opinions of their own worth. High self-esteem may feel good, but as Robert Nozick’s ‘experience machine’ thought experiment (Nozick, 1974) and the soma users of Aldous Huxley’s Brave New World (Huxley, 1932) demonstrate, what matters in life is not simply enjoying pleasant experiences. We also want our experiences to bear the right sort of relation to reality. We want to engage with the world, and we care whether things are going well or badly for us. As a result, we want to be able to recognise when life is going well or badly, and to respond accordingly. By the same token, we do not want simply to have favourable opinions of ourselves; we want to have such opinions because they reflect the way we really are. That it is important that one’s opinion of oneself should not remain constantly high, but should track reality, is supported by some of our common attitudes. There are, we believe, circumstances in which high self-esteem is inappropriate. For example, it may be quite proper that someone who betrays a friend should feel bad about himself afterwards. Moreover, there are acceptable and unacceptable ways for such a person to attempt to restore his self-esteem. We may approve of his doing so by making amends to the betrayed friend (providing that this is not merely a means to making himself feel better); but we disapprove of attempts to 8 It is an interesting to consider what might count as an appropriate level of selfesteem given one’s character, behaviour, and achievements, and how one might properly calculate this; or whether it is even co herent to supp ose tha t there co uld be such a metho d of calculation. I do no t have sp ace to give full co nsideration to these questions here. For the purposes of this paper, I rely on the assumption that we have intuitions ab out what counts as an ap propriate level of respect, admiration, esteem, disdain, loathing (and so on) to accord a person based on their character, behaviour, and achievements; and about when such levels are too high or too low. I assume that it is appropriate for us to evaluate ourselves using the same standards that we ap ply to others. SHOULD WE ENHA NCE SELF -ESTEEM ? 81 escape his unpleasant feelings that do not also address the source of these feelings, or that attempt to avoid confronting them, such as drinking large quantities of alcohol. The importance of anchoring self-esteem in reality, however, is often not reflected in popular culture, from which a confused picture of self-esteem emerges. The National Association for Self-Esteem, whilst noting on its website that ‘self-esteem is tied to reality, not to faking reality’9 , nevertheless recommends strategies for raising self-esteem that focus not on making the right sorts of achievements, having a strong moral character, or improving oneself in other ways that might plausibly make a positive impact upon one’s perception of one’s worth; but on learning to perceive and interpret one’s experiences in a certain way, developing certain communication skills, and behaving in a way likely to command respect from others. As such, despite its claims to the contrary, it is promoting a notion of self-esteem that is concerned more with how one perceives oneself and one’s efforts than with developing the right sort of character or making the right sorts of efforts. Other attempts to promote self-esteem by focusing on perception rather than reality are not difficult to find: the famous L’Oreal advertisement campaign assures us unconditionally that we’re ‘worth it’, and book searches on the term ‘self-esteem’ returned, as the most relevant title on both the British and American Amazon websites, books offering cognitive techniques to improve self-esteem.10 There is, it seems, a pervasive belief in British and American culture that everyone deserves, and can attain, high selfesteem.11 We have seen that one’s self-esteem should be grounded in an accurate evaluation of oneself. What, then, explains the widespread emphasis on unconditionally boosting self-esteem? The common but largely mistaken assumption, discussed earlier, that high self-esteem is beneficial plausibly goes some way towards explaining this, particularly when coupled with the common but largely mistaken assumption that low 9 See http://www.self-esteem-nase.org/self-esteem-questions-answers.shtml. 10 The ‘most relevant’ title on the American site was McKay and Fanning 2003. On the British site, it was Fennell 1999. 11 I am po orly qualified to comment on the extent to which this belief pervades other cultures; however, conversations with colleagues from other countries have suggested that it is present throughout much of Europe. 82 REBECCA ROACHE self-esteem is at the root of a variety of personal and social ills. Perhaps the anticipated benefits of raising everyone’s self-esteem eclipse the importance of grounding it in reality. On the other hand, perhaps it is uncharitable to view the popular attitudes to self-esteem discussed above as advocating unconditional boosting of self-esteem. It may be that those who believe that everybody’s self-esteem should be raised do acknowledge the importance of anchoring self-esteem in reality, but also believe that people’s selfesteem is, in general, too low given their character, achievements, and other relevant factors. I have not encountered any reliable data to support the latter belief, but it may have its roots in various cultural phenomena. For example, we might speculate that the trend of turning away from schooling methods that emphasized the importance of ‘building character’ – popular half a century ago – has resulted in a belief that children are owed a debt of nurturing, consideration, and gentleness, all of which might involve self-esteem-boosting assurances that the child is worthy of certain good things. The increasing secularisation of (parts of) society might also play a role: people are turning away from traditional Christian values such as modesty and self-sacrifice, and embracing sexual freedom and – particularly in certain professional spheres – selfpromotion. Changing attitudes towards women may also be relevant: over the past half-century, the expectation that women place their own needs behind those of their family has become outdated; and it is now acceptable, or even – as the advertisement campaigns of myriad beauty products and indulgent foods remind us – desirable, for women to spoil themselves. These and other factors may embody and promote the belief that the self-esteem of certain groups of people is too low and ought to be raised; however, they do not constitute evidence for that belief. 4. Could Self-Esteem be Enhanced? Much of popular culture, with its conviction that everybody should have high self-esteem, points to the conclusion that we could enhance selfesteem merely by raising it. We have seen, however, that indiscriminately boosting self-esteem is not desirable. In this case, what – if anything – would constitute an enhancement of self-esteem? In order to answer this question, it is necessary to consider what we want from self- SHOULD WE ENHA NCE SELF -ESTEEM ? 83 esteem: what role does it play in our lives? One plausible answer is that we want pleasant feelings: high self-esteem feels good, low self-esteem feels bad, and so we prefer to have higher rather than lower self-esteem. However, given that it is also important that one’s self-esteem should reflect the way one really is, a genuine enhancement of self-esteem could not be achieved merely by simulating the pleasant feelings associated with high self-esteem. The pleasant feelings should arise as a side-effect of high self-esteem, which should itself arise only if one warrants favourable self-evaluations. And, whether favourable self-evaluations are warranted depends on one’s character, behaviour, and so on. Given this, is it even possible to enhance self-esteem with drugs? That it may be possible becomes evident when we consider the relation between self-esteem and self-evaluations. There are a variety of ways in which self-evaluations can shape our behaviour, and some ways are preferable to others. Negative self-evaluations can challenge selfesteem.12 This can provoke a variety of responses, but – ideally for one’s personal growth – it should motivate one to attempt to improve oneself in the relevant respects. That this is desirable is suggested by our attitudes towards the relation between people’s self-evaluations and their resulting behaviour: we approve of those who take honest and accurate negative self-evaluations (and honest and accurate criticism in general) ‘on the chin’, and who attempt to improve in light of them; whereas we disapprove of, or pity, those who make no constructive use of criticism, refuse to accept it, allow it to wear them down, or even – as in those cases discussed by Baumeister, Smart et al. – respond with aggression. Positive self-evaluations, on the other hand, can boost self-esteem, which leads to pleasant feelings that reward and incentivise improvement; but as Crocker and her colleagues point out, it is not desirable that such feelings should become addictive or corrupt one’s motivation by leading 12 Only certain types of negative self-evaluations challenge self-esteem, as James noted: ‘I, who for the time have staked my all on being a psychologist, am mortified if others know much more psychology than I. B ut I am contented to wallow in the gro ssest igno rance of Greek. My deficiencies there give me no sense of personal humiliation at all. Had I “pretensions” to be a linguist, it would have been just the reverse’ (James 1890, 311). The reverse holds true for positive self-evalua tions: they can bo ost self-estee m, but only if they relate to certain valued areas. Crocker’s work, as w e have seen, also notes that selfesteem can be contingent on one’s pe rformance in certain areas. 84 REBECCA ROACHE one to pursue only those activities likely to result in success. There are, then, various ways in which one can respond to challenges or boosts to self-esteem. Best for one’s personal growth and well-being is responding in such a way that uses self-esteem-challenging negative evaluations as a springboard to improvement, and enjoys the pleasant feelings associated with self-esteem-boosting positive evaluations without becoming addicted to them. When one responds as such, let us say that one is benefiting from the motivational role of self-evaluations. Unfortunately, few people benefit from the motivational role of self-evaluations. In order to make constructive use of successes and failures, one must recognise them as such: one must internalise negative and positive feedback, and one must be able to evaluate oneself honestly and accurately. People generally lack this ability, however. Let me explain. William James, who is credited with coining the term ‘selfesteem’, defined it as success divided by pretensions (James, 1890). On this view, one’s self-esteem is completely determined by how one evaluates one’s individual achievements. This ‘bottom-up’ view is pervasive in modern and failures, and psychology13 ; however, it has recently been disputed. We have seen that people often resist revising their level of self-esteem in the light of external feedback14 ; moreover, Brown, Dutton et al. find that people’s level of self-esteem remains relatively stable in the face of fluctuating successes and failures, and that it biases one’s judgments about those successes and failures (Brown et al, 2001). In essence, those with high self-esteem are reluctant to revise their self-esteem downwards, and those with low self-esteem are reluctant to revise their self-esteem upwards. Feedback that confirms one’s existing self-esteem level is accepted, whilst feedback that threatens to undermine it is rejected or rationalised so as not to challenge one’s existing selfesteem level. Those with high self-esteem are likely to accept responsibility for success yet refuse to accept responsibility for failure (usually by attributing their performance to lack of effort or to factors 13 For example, Pelham and S wann describe individual self-evaluations as ‘the building blocks of self-esteem’ (Pelham and Swann 1989, 673; cited in Brown et al 2001). 14 See the references cited in footnote 7. SHOULD WE ENHA NCE SELF -ESTEEM ? 85 beyond their control); and the reverse holds true for low self-esteem. As such, self-evaluations do not always play their motivational role. With this in mind, perhaps improving our ability to evaluate ourselves fairly and accurately, and to recognise successes and failures as such, would constitute an enhancement of self-esteem. At the very least, this ability seems to be necessary for achieving well-functioning selfesteem, which involves the motivational role of self-evaluations. However, there is some evidence to suggest that mentally healthy individuals hold a variety of overly positive illusions about themselves, and that their mental health is tied to their holding such illusions; whereas mentally unhealthy people – including people with low selfesteem – perceive themselves, the world, and their future more accurately (Taylor and Brown, 1988). This research has been disputed15 , but if there is a grain of truth in the claim that certain illusions about oneself and the world promote mental health, there are implications for our treatment of self-esteem. Specifically, attempting to enhance self-esteem by improving our ability to evaluate ourselves fairly and accurately could be counter-productive. We may be forced to choose between two highly valuable mental capacities: happiness and accurately perceiving reality. Perhaps there is a way out of this dilemma. It seems plausible that the reason many of us avoid accurate self-evaluations is that we are unable to take them ‘on the chin’: people are reluctant to revise their level of self-esteem upwards or downwards, and so we avoid internalising feedback that challenges our current level of self-esteem. It is, then, unsettling to confront information about ourselves that conflicts with our self-esteem level. If, however, we were able to improve our ability to tolerate accurate feedback – including honest, accurate selfevaluations – regardless of its impact on our level of self-esteem, and without becoming overly disturbed by it, this could facilitate the motivational role of self-evaluations. As such, it could constitute an enhancement of self-esteem. There is some suggestive evidence that such an enhancement would be possible. The psychoanalyst Elizabeth Zetzel, drawing on her work with the British Emergency Medical Services during the second World War, discussed the importance of ‘affect tolerance’ (Zetzel, 1970). She noted that, for soldiers, the capacity to experience anxiety is useful, 15 See, for example, Colvin and Block 1994. 86 REBECCA ROACHE since anxiety produces a state of alertness necessary for self-preservation in a dangerous environment. To fulfil this function, however, one must be able not only to experience anxiety, but also to tolerate it: to experience it without either ‘blocking’ it or becoming overwhelmed by it. More generally, Zetzel believed that the capacity to experience and tolerate states like anxiety and depression is beneficial, since it is this that gives rise to our emotional growth. Richard Schwartz later took up the idea of affect tolerance and its benefits, and argued that, by preventing patients from experiencing states like depression (therefore removing the need to tolerate them), antidepressant drugs stunt emotional growth (Schwartz, 1991, cited in Kramer, 1993). The effect of at least some antidepressant medicines is more complicated than Schwartz supposes, however. Kramer tells us that, far from emotionally deadening patients by removing their depression, [i]n some patients, Prozac quite directly incre ases the ability to bear troubling emotions. On Prozac, Pau l, the Renaissance history teacher, no longer just imagined but, for the first time, felt his childhood mem ories o f trauma. His em otional palette expanded quite directly because of medication. In many patients, Prozac lets feelings emerge in new settings. Allison, the fashion designer, was able on Prozac to display her gentle and concerned side in the office, because she felt less anxious. Certainly people who become less obsessional on the drug are thereby made mo re open to emotion. No t only does P rozac increase resilience, in so me p eop le it increases the profund ity of emo tion available to them as well. (Kramer, 1993: 259) Indeed, so startling are Prozac’s powers to increase affect tolerance in some cases that, Kramer goes on to remark, ‘Prozac raises the opposite issue: how comfortable are we with a pill that increases affect tolerance?’ What, exactly, is affect tolerance: what is involved in becoming more tolerant of certain emotional states? Kramer suggests two possibilities: first, increasing one’s affect tolerance might involve allowing one to experience a greater variety or intensity of disturbing states; and second, it might involve making one less likely to experience ‘disruptive emotion’ in response to certain types of events (Kramer, SHOULD WE ENHA NCE SELF -ESTEEM ? 87 1993: 258). Kramer describes Prozac as having had each of these effects in certain patients. Given the success of Prozac in increasing affect tolerance, we might be optimistic about the possibility of a drug that would improve affect tolerance in ways relevant to self-esteem. First, it might be possible to create a drug that would enable people to feel a greater variety and intensity of emotion in response to accurate feedback. Second, such a drug might be able to make people less likely to experience ‘disturbing emotion’ in response to such feedback, thus enabling the motivational role of self-evaluations described above. Were such a drug available, would it enhance self-esteem? That is, would it increase our capacity to evaluate ourselves honesty and accurately, and to respond in the right sort of way to those evaluations? The answer to this question depends in large part on what exactly would be involved in increasing our affect tolerance, and what states would be targeted. Not all disturbing states would be good candidates. Whilst some emotional states like anxiety and depression are – according to Zetzel – beneficial to experience, meaning that it is better to tolerate them than to block them, there seems little value in tolerating certain other states. For example, whilst it can be beneficial to be able to experience acute pain, since it generally alerts us to the fact that something is wrong with the body and needs to be put right; there is little to be gained from being able to tolerate chronic, pathological pain, such as the pain that amputees can feel in their phantom limbs. In the case of the latter, it is usually better where possible to block the pain than to tolerate the pain. The acutechronic model is instructive in thinking about emotional states, too16 : whilst states like anxiety and depression may be appropriate and valuable in some cases, such as in response to the right sorts of stimuli, they can be disruptive and disabling if they continue indefinitely and bear no relation to the state of the subject or the world around her. Whilst there may be a case for increasing tolerance of some of the former, ‘acute’, types of state; the latter, ‘chronic’, types of state serve no useful function and are better blocked than tolerated. However, it is not always easy to tell which types of state are functional and appropriate and which are not, and our beliefs about them – and about whether they should be tolerated 16 I am grateful to Dirk de Ridder for drawing my attention to the relevance of this distinction to other types of state. 88 REBECCA ROACHE – may be influenced by cultural factors. Consider bereavement. In the US, it is common to prescribe antidepressants for the bereaved if their grief continues for longer than about a year. On the other hand, in rural Greece, it is expected that a mother’s grief for a lost child, or a wife’s grief for her dead husband, will last for five years (Schwartz, 1991, cited in Kramer, 1993). Adherents to each of these cultures, then, are likely to disagree about what constitutes an appropriate duration of grief. In addition, there are some unpleasant emotional states that should be neither tolerated nor blocked. It is disturbing to read of the effects of the fictional soma in Brave New World, which renders its users apathetic about the dystopia in which they live; or of the effects of ‘mother’s little helpers’ like Valium and Librium, which enabled housewives in the 1950s and 60s to content themselves with monotonous, undemanding lifestyles. It seems plausible that such cases are disturbing because we believe that it is quite proper that people in these situations should feel frustrated and oppressed; and that their frustration should not be blocked or tolerated, but should spur them to remove themselves from the frustrating situation and live fulfilling lives elsewhere. Still other states may serve useful functions that depend on our having a finely-tuned level of tolerance for them. For example, guilt is unpleasant to experience, and the anticipation of feeling it can inhibit our behaviour, such that we avoid behaving in a way likely to result in guilt. Since guilt, when functioning properly, is felt in response to acting immorally, our avoidance of acting in a way likely to result in guilt helps to prevent us from acting immorally. Except, perhaps, in cases of people whose guilt may be deemed pathological17 , increasing tolerance for guilt (or blocking it) may turn out to be disastrous, since it may result in an increase in immoral behaviour. Whilst, then, the possibility of using drugs to increase affect tolerance suggests that it may be possible, by increasing our tolerance for truthful feedback about ourselves, to facilitate the motivational role of self-esteem; more research is needed to determine whether doing so would constitute an enhancement of self-esteem. Empirical research is required to elucidate further the interaction between self-evaluations and self-esteem levels; in particular, the extent to which increasing tolerance 17 For example, some people feel guilt in response to eating. See, for example, Frank 1991. SHOULD WE ENHA NCE SELF -ESTEEM ? 89 for disturbing feedback about ourselves might increase our willingness to revise our self-esteem. And, philosophical research is required to assess whether tolerating such emotions would be desirable, given our other values. University of Oxford Email: rebecca.roache@philosophy.ox.ac.uk REFERENCES And erson , E (1994 ). The Co de of the Streets, Atlan tic Month ly 273, pp. 81-94. Baumeister, RF, C amp bell, JD , Krueger, JI, Vo hs, KD (2 003 ). Do es High SelfEsteem Cause Better Performance, Interpersonal Success, Happiness, or Healthier Lifestyles? Psych ological Science in the Pu blic Interest 4, pp. 1-44. Baumeister, RF, Smart, L, Boden, JM (1996). R elation of Threatened E gotism to Violence and Aggression: The Dark Side of High Self-Esteem, Psychological Review 103, pp. 5-33. Bra nden, N (1 994 ), The Six Pillars of Self-Esteem. New Y ork: Ba ntam B ooks. Brown, JD, Dutton, KA, Cook, KE (2001). From the Top Down: Self-Esteem and Self-Ev aluation, Cognition and Emotion 15, pp. 615-631. California Task Force to Promote Self-Esteem and Personal and Social Resp onsibility (1990), Toward a State of Self-Esteem. Sacramento: California State Department of Education. Colvin, CR, Block, J (1 994 ). Do Positive Illusions Foster M ental H ealth? An Examination of the T aylor and B rown Form ulation, Psychological Bulletin 116, pp. 3-20. Crocker, J (2002). Contingencies of Self-Worth: Implications for Self-Regulation and Psych ological Vulnerability, Self and Iden tity 1, pp. 143-149. Crocker, J, Ka rpinski, A, Quinn, DM, Chase, S (2003 ). When G rades Determine Self-Worth: Consequences of Contingent Self-Worth for Male and Female Engineering and P sychology M ajors, Journal of Personality and Social Psychology 85, pp. 507-516. Crocker, J, Knight, KM (2005). Contingencies of Self-W orth, Current Directions in Psychological Science 14, pp. 200-203. Crocker, J, Major, B (1989). Social Stigma and Self-Esteem: The Self-Protective Pro perties of Stigm a, Psychological Review 96, pp. 608-630. Crocker, J, Park, LE (2004). The Co stly Pur suit of Self-Esteem , Psychological Bulletin 130, pp. 392-414. 90 Deci, REBECCA ROACHE EL, Eghrari, H, Patrick, BC, Leo ne, DR (199 4). Facilitating Internalization: The Self-D etermination T heory Perspec tive, Journal of Personality 62, pp. 119-141. Fennell, M (1999), Overcoming Low Self-esteem: Self-help Guide Using Cognitive Behavioural Techniques, London: Constable and Ro binson. Frank, ES (1991). Sham e and Guilt in E ating D isorders, American Journal of Ortho psychiatry 61, pp. 303-306. Heine, SH, Lehman, DR, Markus, HR, Kitayama, S (1999). Is There a Universal Need for Po sitive Self-Regard?, Psychological Review 106, pp. 766-794. Huxley, A (1 932 ), Bra ve New W orld, New Yo rk: Harper Perennial. James, W (1890), Principles of Psychology, New Yo rk: Holt. Jankowski, MS (19 91), Islands in the Street: Gangs and American Urban Soc iety, Berkeley: U niversity of California Press. Kirschner, D (1 992 ). Understanding Ado ptees W ho K ill: Dissoc iation, P atricide, and the Psychodynamics o f Ado ption, International Journal of Offender Therapy and Comparative Criminology 36, pp. 323-333. Kra mer, P (1993), Listening to Prozac, 2nd ed., London: Penguin. Levin, J, M cDevitt, J (1993 ), Hate Crimes: The Rising Tide of Bigotry and Bloodshed, New Y ork: Plenum Press. Long, DE (1990), The A natom y of Terrorism , New Y ork: Free P ress. Mathews, J (2006). For M ath Students, Self-Esteem Might Not Equal High Sco res, Wash ington Post, p A02. McDonald, JM (1975), Armed Robbery: Offenders and Their Victims, Springfield, IL: Cha rles C Thomas. McKay, M , Fann ing, P (2 003 ), Self Esteem: A Proven Program of Cognitive Techniques for Assessing, Improving, and Maintaining Your Self-Esteem, MJF B ooks. No zick, R (1974), Anarch y, State and Utopia , New Yo rk: Blackwell. Oates, RK, Forrest, D (1985). Self-Esteem and Early Background o f Abusive Mothers, Child Abuse and Neglect 9, pp. 89-93. Pelham, B W , Swan n, W B, Jr. (1989). From Self-Conceptions to Self-Worth: On the Sources and Structure o f Glob al Self-Esteem, Jou rnal of Personality and Social Psychology 57, pp. 672-680. Renze tti, CM (1992), Violent Betrayal: Partner Abuse in Lesbian Relationships, Newbury Park, CA: Sage. Schoenfeld, CG (1988). Blacks and Violent Crime: A Psychoanalytically Oriented A nalysis, Journal of Psychiatry and Law 16, pp. 269-301. Schwartz, RS (1991): Mo od B righteners, Affect Tolerance, and the Blues, Psych iatry 54, pp. 397-403. Shrauger, JS, Schoenem an, T J (19 79). S ymbo lic Interactionist V iew of S elfConcep t: Through the Looking Glass Darkly, Psychological Bu lletin 86, pp. 549-573. SHOULD WE ENHA NCE SELF -ESTEEM ? 91 Swann, W B, H ill, CA (1 982 ). W hen O ur Identities are M istaken: R eaffirming Self-Conceptions Through Social Intera ction, Journal of Personality and Social Psychology 43, pp. 59-66. Taylor, SE, Brown, JD (1988). Illusion and Well-Being: A Social-Psychological Perspective on Mental H ealth, Psycholo gical Bulletin 103, pp. 193-210. Tice, DM , Butler, JD, Muraven, M B, Stillwell, AB (1995). W hen M ode sty Prevails: Differential Favorab ility of Self-Presentation to Friends and Strangers, Journal of Personality and Social Psychology 69, pp. 11201138. Toch, H (1 969 ), Violent Men: An Inquiry into the Psychology of Violence, W ashington, DC: American Psychological Association. W alker, LE (1979 ), The Battered Woman, New York: Harper & Row. Wiehe, VR (1991), Perilous Rivalry: When Siblings Become Abusive, Lexington, MA: Heath/Lexington B ooks. Zetzel, ER (1970), The Capac ity for Emotiona l Gro wth, Londo n: Hogarth Press.