Self-Compassion
Self-compassion is treating your own failures the way you would treat a friend's. Kristin Neff defined it in 2003 as three things: kindness instead of judgment, recognising that failure is common to everyone rather than particular to you, and holding the difficulty in view without being swallowed by it. The evidence runs against the intuition that harshness motivates. Pooled across 79 samples and more than 16,000 people, a caring self-attitude tracks higher wellbeing, and shame after a lapse predicts more of the behaviour, not less.
What it is
Most people carry an unexamined theory that being hard on yourself is what keeps you in line, and that easing off is the first step towards letting yourself go entirely. Neff's construct is built on the observation that this theory does not survive contact with the data. Her 2003 paper in Self and Identity, "Self-Compassion: An Alternative Conceptualization of a Healthy Attitude Toward Oneself", set out the alternative in three components, and the three-part structure is the reason the idea became measurable rather than merely appealing.
The first component is self-kindness against self-judgment: responding to your own failure with warmth rather than criticism. The second is common humanity against isolation: recognising that failing is a condition of being a person, not a private defect that marks you out. This is the component people miss, and it is the one that does most of the work, because the sting of a lapse usually comes less from the lapse than from the belief that other people would not have lapsed. The third is mindfulness against over-identification: holding the painful thing in awareness without fusing with it, so that "I did something badly" does not become "I am someone who does things badly".
The evidence base is unusually good for a construct of this kind. Zessin, Dickhäuser and Garbade's 2015 meta-analysis in Applied Psychology: Health and Well-Being pooled 79 samples covering 16,416 people and found an overall relationship between self-compassion and wellbeing of r = .47, with a confidence interval from .44 to .49. The relationship was strongest for psychological wellbeing at r = .62 and weakest for positive affect at r = .39. A trim-and-fill analysis found no evidence of publication bias, which matters more than the headline figure does, because publication bias is what inflated most of the famous self-control findings of the same era. A small experimental subsample within the meta-analysis, 394 participants, supports reading at least part of the relationship as causal rather than purely correlational.
Where the construct earns its place on this site is as a replacement, not an addition. Ego depletion, the theory that self-control draws on a finite fuel tank, did not survive testing at anything like its original size. What survived the wreckage is closer to a claim about belief: people who think of themselves as having no self-control behave accordingly. Nir Eyal makes the connection explicit in Indistractable, where self-compassion is offered as the operational alternative. If believing yourself powerless is what produces powerlessness, then the intervention is on the belief rather than on the supposed reservoir, and self-compassion is the practised form of that intervention.
The practice Neff's work implies is unglamorous by design. Talk to yourself in the second person, the way a friend would, about a setback you have just had. "This is what it is like to get better at something" is a sentence you would say to someone else without effort, and a sentence most people will not say to themselves. The mechanism it interrupts is rumination, the passive replaying of a failure against a standard you did not meet, which is the process by which one bad evening becomes a bad month.