Why Did Science Find That Narcissists Don't Secretly Hate Themselves?

For years, the common-sense explanation for narcissism was that, deep down, every narcissist hates themselves. The research Kristin Neff reviews in Self-Compassion points in another direction, although the full picture is more nuanced than it’s usually made out to be.

Neff, a psychologist at the University of Texas at Austin, devotes several pages of her book (2011) to an uncomfortable turn in self-esteem research: the data not only question whether it’s worth chasing high self-esteem at all costs, they also cast doubt on one of the most repeated explanations for why narcissism exists.

The mask hypothesis

The idea that the narcissist hides a core of insecurity has clinical roots: psychoanalytic writers such as Otto Kernberg and Heinz Kohut described grandiosity as a defense against a fragile self-image. In social psychology it’s known as the mask model.

Its prediction is concrete and testable: someone with many narcissistic traits should show very high explicit self-esteem (the kind they report) and low implicit self-esteem (the kind that shows up in automatic reactions).

The instrument that put the myth to the test

Testing this required measuring something the person couldn’t simply state out loud. Enter the Implicit Association Test (IAT), developed by Greenwald, McGhee and Schwartz (1998): a computer task that measures how quickly someone associates the label “me” with positive words (“wonderful”) versus negative ones (“horrible”).

• A fast association between “me” and positive words, and a slow one between “me” and negative words, indicates high implicit self-esteem. The reverse pattern indicates low implicit self-esteem.
• As Neff reports, people who score high on the Narcissistic Personality Inventory (NPI, with statements such as “I think I am a special person” or “I like to look at myself in the mirror”) score high both explicitly and implicitly.

The most cited study in this line is by Campbell, Bosson, Goheen, Lakey and Kernis (2007), published in Psychological Science under a title that is itself a question: “Do narcissists dislike themselves ‘deep down inside’?” Their answer was no, with an important nuance: at the implicit level, narcissistic people associated themselves with agentic traits (status, dominance, intelligence), but not with communal traits (kindness, morality, warmth). Rather than self-esteem hidden at rock bottom, what they describe is a lopsided self: highly impressed with its own competence and not very interested in its own goodness.

What the popular summary tends to leave out

This is where the story stops being as tidy as the headline suggests.

• Jordan, Spencer, Zanna, Hoshino-Browne and Correll (2003) found exactly the mask pattern in one of their studies: people who combined high explicit and low implicit self-esteem were the ones who showed the most narcissism, and in other experiments that combination predicted more defensive behavior.
• The review by Bosson and colleagues (2008), co-authored in part by researchers from both camps, concludes that the evidence is mixed and offers three reasons: the low reliability of implicit measures, the difference between agentic and communal self-views, and the conflation of two distinct forms of narcissism.
• A later replication of the 2007 study, by a team at the University of Potsdam, also found no low implicit self-esteem, although it did find a more neutral agentic self-evaluation than had been published.

Our reading of this debate: what the research has dismantled is the universal version of the myth (“every narcissist hates themselves inside”), not the possibility that in some people grandiosity is defensive.

Grandiose narcissism and vulnerable narcissism

The piece that resolves much of the contradiction is the distinction between two profiles, reviewed by Pincus and Lukowitsky (2010) and tested by Miller and colleagues (2011) with 858 participants and three different scales:

• Grandiose narcissism is associated with extraversion, dominance and high, stable explicit self-esteem. It’s what the NPI mainly measures and what shows up in the IAT studies.
• Vulnerable narcissism is associated with neuroticism, hypersensitivity to criticism, shame and low or highly fluctuating self-esteem. Here, grandiosity coexists with feelings of inadequacy.

In other words, the popular intuition wasn’t entirely wrong, but it described the vulnerable form better than the grandiose one, which is what most people picture when they think of a narcissist.

Why the myth stayed so appealing

The book cites an example of how this idea circulated in public conversation: a television commentator, talking about young celebrities, claimed that inside them “a little inner voice” kept telling them they weren’t good enough.

Neff quotes a statement Paris Hilton gave to the Sunday Times in 2006 (“There is nobody in the world like me… right now, I am that icon”) as an example of the kind of talk that tends to be read as a front, and contrasts it with what the IAT data show. The myth is comfortable because it turns irritating behavior into something understandable, even deserving of sympathy, without any need for evidence.

What happens when the ego feels threatened

If the grandiose narcissist’s high self-esteem isn’t a cover, where does their intense reaction to criticism come from? Neff describes a classic experiment, by Bushman and Baumeister (1998), designed to answer exactly that.

1. Participants wrote an essay that would supposedly be evaluated by a partner in another room, who didn’t actually exist.
2. The “evaluation” they received was, at random, either praise (“no suggestions, great essay”) or an insult (“one of the worst essays I have read”).
3. Then, in a task presented as competitive, participants could set the volume and duration of an unpleasant noise aimed at that same “partner.”
4. Those who scored high on narcissism and had received the criticism delivered the loudest and longest noise blasts. Self-esteem on its own did not predict that aggression.

Neff’s reading, which the authors share: the aggression doesn’t appear because the narcissist feels bad about themselves underneath, but as a response to a specific threat against a self-image that, until that moment, had held up without cracks. Baumeister and his team called it threatened egotism.

The underlying problem: what self-esteem depends on

Neff widens the lens beyond narcissism. She cites the term psychologists use for a much more widespread pattern: contingent self-esteem, the kind that depends on success, approval or comparison in a specific area, whether physical appearance, job performance or the perceived approval of strangers.

• The more self-esteem depends on a specific area, the worse the person feels when they fail precisely there. The result resembles, in the author’s words, an emotional roller coaster.
• Neff uses the term “hedonic treadmill” (Brickman and Campbell, 1971) to describe the need for an ever larger dose of approval or success to sustain the same level of well-being.
• As historical context, she mentions the report commissioned by the state of California in the late 1980s to promote self-esteem in schools, which failed to find most of the expected effects. Years later, the review by Baumeister, Campbell, Krueger and Vohs (2003) reached a similar conclusion with far more evidence: high self-esteem is associated with feeling better and with more initiative, but it barely predicts academic or job performance, and to a large extent it’s a consequence of success rather than its cause.

If you’d like to explore what your own self-esteem rests on, the self-esteem test focuses on precisely that foundation.

Mistaking the map for the territory

One of the book’s central metaphors compares each person’s concept of themselves to a painting: a representation, not reality itself. Neff likens it to mistaking the fruit in a still life for real fruit, and getting frustrated when it turns out you can’t eat it.

The self-concept, she argues, is always a simplified version of something far more changeable: thoughts, emotions and behaviors that vary with the moment and the context. Clinging to a fixed, positive self-image works, in her image, like an inflatable raft with a hole in it: it keeps you afloat for a while, but demands constant effort. The alternative she proposes isn’t to replace a positive label with a negative one, but to stop needing a fixed label of “good” or “bad” at all.

The alternative Neff has researched

Neff’s argument doesn’t stop at critiquing self-esteem. She proposes self-compassion: treating yourself with the same kindness you’d show a friend, recognizing failure as part of shared human experience, and observing discomfort without dramatizing it.

Her study with Roos Vonk (Neff and Vonk, 2009), with more than 2,000 participants, compared the two variables. Self-compassion was associated with a sense of worth that was more stable over time and less dependent on outside approval; and, unlike self-esteem, it was not related to narcissism. These are correlational data, largely from the author of the construct herself, so they’re best read as a promising line of research rather than a settled conclusion.

This article summarizes research on narcissism and self-esteem for informational purposes. It is not a diagnostic criterion for narcissism or for any other personality trait or disorder.

This mechanism, by which self-esteem comes to depend on comparing and competing, ties in with what we explain about the brain chemistry of self-criticism and with the Big Five personality traits, where neuroticism and agreeableness relate to very different ways of handling social comparison.

It also connects with how inner security takes shape from childhood, covered in the article on secure attachment.

Sources

• Neff, K. (2011). Self-Compassion: The Proven Power of Being Kind to Yourself. William Morrow.
• Neff, K. D., and Vonk, R. (2009). Self-compassion versus global self-esteem: Two different ways of relating to oneself. Journal of Personality, 77(1), 23-50.
• Greenwald, A. G., McGhee, D. E., and Schwartz, J. L. K. (1998). Measuring individual differences in implicit cognition: The Implicit Association Test. Journal of Personality and Social Psychology, 74(6), 1464-1480.
• Campbell, W. K., Bosson, J. K., Goheen, T. W., Lakey, C. E., and Kernis, M. H. (2007). Do narcissists dislike themselves “deep down inside”? Psychological Science, 18(3), 227-229.
• Jordan, C. H., Spencer, S. J., Zanna, M. P., Hoshino-Browne, E., and Correll, J. (2003). Secure and defensive high self-esteem. Journal of Personality and Social Psychology, 85(5), 969-978. doi:10.1037/0022-3514.85.5.969
• Bosson, J. K., Lakey, C. E., Campbell, W. K., Zeigler-Hill, V., Jordan, C. H., and Kernis, M. H. (2008). Untangling the links between narcissism and self-esteem: A theoretical and empirical review. Social and Personality Psychology Compass, 2(3), 1415-1439.
• Pincus, A. L., and Lukowitsky, M. R. (2010). Pathological narcissism and narcissistic personality disorder. Annual Review of Clinical Psychology, 6, 421-446.
• Miller, J. D., Hoffman, B. J., Gaughan, E. T., Gentile, B., Maples, J., and Campbell, W. K. (2011). Grandiose and vulnerable narcissism: A nomological network analysis. Journal of Personality, 79(5), 1013-1042. doi:10.1111/j.1467-6494.2010.00711.x
• Bushman, B. J., and Baumeister, R. F. (1998). Threatened egotism, narcissism, self-esteem, and direct and displaced aggression: Does self-love or self-hate lead to violence? Journal of Personality and Social Psychology, 75(1), 219-229.
• Baumeister, R. F., Campbell, J. D., Krueger, J. I., and Vohs, K. D. (2003). Does high self-esteem cause better performance, interpersonal success, happiness, or healthier lifestyles? Psychological Science in the Public Interest, 4(1), 1-44.
• Brickman, P., and Campbell, D. T. (1971). Hedonic relativism and planning the good society. In M. H. Appley (Ed.), Adaptation-Level Theory: A Symposium. Academic Press.