We think we reason our way to our moral positions. Haidt's research suggests we have it exactly backwards — and that this mistake is responsible for most of what makes political and cultural conversation so consistently fruitless.

There is a model of persuasion that most analytical and professional practice implicitly relies on: present good evidence, construct a clear argument, address objections, change minds. It is a model with considerable intuitive appeal and a poor track record. Jonathan Haidt's book offers a precise account of why. It’s central claim is that moral judgment works in the opposite direction from the one the rational model assumes. We do not reason our way to a position and then feel the appropriate emotion about it. We feel first — an immediate intuitive response that is faster and more powerful than conscious deliberation — and we reason afterward, largely to justify and defend a conclusion that was already reached. The reasoning, on this account, is less a path to truth than a performance of coherence: the brain's press secretary, generating plausible narratives to defend decisions the president made without consultation.

The contribution Haidt makes beyond the intuition-first framework — which has antecedents in cognitive science — is to give that framework specific moral content. He identifies six foundations on which moral intuitions are built: care and harm, fairness and cheating, loyalty and betrayal, authority and subversion, sanctity and degradation, liberty and oppression. These categories correspond to recurring adaptive challenges in human social life, and they appear across cultures. What varies — and varies enormously — is the weight different groups place on each foundation.

In the American political context Haidt uses as his primary case, the asymmetry is striking: progressive voters draw predominantly on care, fairness, and liberty; conservative voters draw on all six, including loyalty, authority, and sanctity. The result is not disagreement within a shared moral language, but mutual incomprehension — each side reasoning fluently from intuitions the other side does not share and cannot easily access.

The methodological implication is direct. If you want to understand why a person or group holds the positions they hold, engaging with their stated reasoning will only take you so far. The reasoning is post-hoc, constructed to defend an intuition that preceded it. To reach the intuition, you need different tools: not debate but proximity, not argument but sustained attention to what produces the immediate emotional response, the visceral reaction that happens before the justification is assembled.

Moral intuitions are not formed in isolation. They are socially reinforced, shaped by the groups we belong to, and maintained by the reputational pressures Erving Goffman mapped with such precision: the constant management of how we appear to those whose regard matters to us. A moral foundation that is self-evident within a group — that feels not like an opinion but like simple reality — becomes literally difficult to perceive from outside that group, and vice versa. The progressive who cannot understand how anyone could weight loyalty over care, the conservative who cannot understand how anyone could dismiss sanctity as irrelevant — both are experiencing the limits of their own intuitive formation.

The intuitions that feel self-evident to an analyst are partly a function of the groups they have inhabited, the moral foundations those groups weighted, and the social reinforcement that made certain things feel obvious and others feel strange. Recognising that is the precondition for the kind of attention that produces genuine understanding rather than the sophisticated restatement of prior assumptions.