Appearance
9.6 — Self-Esteem, and the Thing That Works Better
In 1986 the state of California funded a task force to promote self-esteem, on the theory that raising it would reduce crime, teenage pregnancy, drug abuse and school failure.
The task force commissioned reviews of the evidence. The reviews did not find what was expected, and one of the academics involved noted publicly that the associations were mostly weak.
The programme went ahead anyway, and a generation of schools spent two decades telling children they were special.
Two later findings settled it.
Self-esteem does not cause achievement; achievement causes self-esteem. Longitudinal studies find that doing well raises how you feel about yourself far more reliably than feeling good about yourself raises how you do.
And high self-esteem is not protective against the things it was supposed to prevent. Roy Baumeister's major review concluded that high self-esteem does not reduce aggression, does not prevent early sexual activity, does not reduce smoking or drinking. What it reliably produces is feeling good and being more willing to speak up.
And there is the finding from 5.6: aggression is associated not with low self-esteem but with high and unstable or inflated self-esteem under threat.
Why the target was wrong
Self-esteem is an evaluation — a judgement of your own worth. And every evaluation has the same two problems.
It requires evidence, so it depends on things going well. Self-esteem built on achievement collapses when you fail, which is precisely when you need something to stand on.
And it is comparative. Worth is largely judged against other people, so someone else's success is a threat, and maintaining it requires either winning or avoiding comparison.
This is 5.6's contingency problem exactly. The question is not how high your self-esteem is; it is what it rests on and who controls that.
Self-compassion
Kristin Neff's alternative, and it has held up considerably better.
Self-compassion is treating yourself as you would treat a good friend in the same situation. It is not an evaluation, so it does not need evidence and does not depend on comparison.
Three components.
Self-kindness rather than self-judgement. Responding to your own failure with warmth rather than attack.
Common humanity rather than isolation. Recognising that failing, struggling and being inadequate are part of being a person, rather than evidence that you specifically are defective. This component does a surprising amount of the work, because the loneliest part of failure is usually the belief that everyone else manages.
Mindfulness rather than over-identification. Holding the difficulty in awareness without being consumed by it — 7.5's observing position.
The objection, answered
"Won't that make me lazy and complacent?"
This is the first thing almost everybody says, and it is the most-tested question in the literature. The evidence goes the other way, consistently.
Self-compassion is associated with greater personal initiative and motivation to improve, not less. In studies where people are induced to be self-compassionate after a failure, they subsequently spend more time preparing for a retest and show more motivation to change, compared with people given a self-esteem boost or nothing.
Self-compassionate people take more responsibility for their mistakes, not less, and are more willing to look at them — because looking is survivable when the response is not attack.
And self-criticism predicts the opposite: more avoidance, more procrastination (8.5), more giving up, more depression and anxiety.
The mechanism is 4.8. Self-criticism produces shame, shame is about the self, the self cannot be repaired, so the response is to hide. Self-compassion allows the error to be about an act, and acts can be fixed.
The honest limits
Self-compassion research has the same methodological issues as most of positive psychology — a lot of cross-sectional self-report, much of it using a scale developed by the person proposing the construct. There is a technical argument about whether the scale's negative items are really measuring self-compassion or simply the absence of self-criticism.
The effects on wellbeing are consistent and moderate, not transformative.
And it is not a treatment for depression on its own, though compassion-focused approaches have a growing evidence base as part of therapy.
What to take: the direction is well supported and the size is modest. Self-criticism does not work as a motivational strategy, and that specific claim is solid.
How to actually do it
The instruction "be kind to yourself" is useless. These are concrete.
The friend test
The single most useful item in this Part. After a failure, write down what your inner voice actually said, word for word (2.9). Then ask: would I say those exact words, in that tone, to a friend who had done the same thing?
Almost nobody can answer yes. Then write what you would say to them, and read it to yourself.
And notice that the friend version is usually more accurate, not merely kinder. It is proportionate, it distinguishes the act from the person, and it contains a next step. The critic's version does none of those things.
The three-line break
For the moment itself (7.6):
This is hard right now.Other people feel like this too. This is part of being a person.May I be kind to myself in this.
Twenty seconds. Usable anywhere.
The name substitution
Talk to yourself in the second person or by name (2.9). You made a mess of that, and it is fixable. The distance changes the tone automatically.
Separate the act from the self, every time
From 4.8. That was careless rather than I am careless. This one substitution does most of the work of the whole page.
And write the letter
Once, for something you have been carrying: write to yourself from the perspective of someone who loves you unconditionally and knows everything about the situation. This is a standard exercise in the research literature and it produces measurable effects on mood and self-criticism, and it is uncomfortable enough that most people avoid it, which is a reason to do it.
Self-concept and the stories
One further layer.
Self-concept is the set of beliefs about who you are — the narrative self from 2.7. It is built from experience, from what people told you, and from repeated self-description.
Two things about it are worth knowing.
It is revised every time it is retold (2.4). The story you have about your childhood, your marriage, your career is the current edition, not the record.
And the negative parts are frequently inherited verbatim. I'm the difficult one. I'm not academic. I've never been good with people. Many of these are direct quotations from a specific person at a specific age, carried for decades without ever being examined.
The exercise: write down three fixed beliefs you hold about yourself. For each, ask where it came from, whose voice it is in, and what evidence there is against it from the last five years. Most people find at least one that does not survive the question.
What to do with this
Stop trying to raise your self-esteem. It is an outcome, not a lever, and the achievement produces it rather than the other way round.
Do the friend test the next time you fail at something. Write the critic's exact words, then write what you would say to someone you care about. This is the whole page in five minutes.
Substitute act for self, permanently. That was rather than I am.
Learn the three-line break and use it this week.
And examine one inherited belief. Where it came from, whose voice, and what the last five years actually show. You are unlikely to argue yourself out of it in one sitting, and identifying it as a quotation rather than a fact is where the change begins.
Next: 9.7 turns from how you treat yourself to what you are actually aiming at — finding your real values, as distinct from the ones you would say out loud.