Courage

What it is

Aristotle's account is narrower than the one attached to his name, and the narrowing is what makes it useful. Courage is the first of the moral virtues he treats in Book III of the Nicomachean Ethics. That ordering is frequently reported as a ranking, and it is not one. He does not say that courage makes the other virtues possible. That formulation is modern, and it attaches to his name because it sounds like the sort of thing he would say.

What he does say is more demanding. Courage is a mean between two failures, cowardice on one side and rashness on the other, and it is concerned specifically with fear and confidence rather than with difficulty in general. Its fullest expression, on his account, is facing death in battle. The consequence that matters is that the virtue only shows itself when the stakes are real. A person who is unafraid because nothing is at risk demonstrates nothing whatever, which rules out most of the situations in which anybody is called brave.

That condition is also, awkwardly, a specification for an experiment that cannot be run. To measure courage as Aristotle defines it, you would have to put a person in genuine danger, having measured something about them beforehand, and record what they did. No ethics committee is going to approve it and no researcher should want to. So the field has done what fields do when the central case is out of reach: it has measured fear instead, which is easy, and self-reported courage, which is easier still, and the behavioural work has been done at the smallest scale the condition permits.

The gap this leaves is not academic. Fifty years of bystander research established that people help less as bystanders are added, and the field data reversed the conclusion, with intervention occurring in nine out of ten of the 219 real conflicts Philpot and colleagues coded on CCTV. That result also cannot explain itself, because the footage records outcomes and nothing about the people. Something separates the person who moves from the person standing beside them who does not. The oldest candidate for that something is on this page, and it has been characterised behaviourally in one small pilot study. See The Bystander Effect for the full shape of that gap.

The unreviewed status here is honest rather than dismissive. Courage is not a construct that failed replication. It is a construct that has barely been tested under the conditions its own best definition requires.

In effect

Research: thirty-two people and a dead spider (Norton and Weiss)

Peter Norton and Brandon Weiss published the most-cited behavioural work on courage in the Journal of Anxiety Disorders in 2009. Thirty-two participants with elevated spider fear completed a self-report courage measure, then faced a display of four taxidermied tarantulas with an instruction to bring a hand as close as they were willing. The measure of behaviour was approach distance. Courage scores predicted how close people got, even after controlling for how frightened of spiders they were, which is the finding that matters: it is not simply a fear measure wearing a different label.

It is a real result and the authors are unusually careful with it. Their own title calls it a proof-of-concept pilot study, and the phrase should travel wherever the finding does. Thirty-two people and a spider that cannot move is the strongest behavioural evidence the field has that self-reported courage predicts what a person actually does.

Everyday life: the meeting nobody speaks up in

The everyday case has the structure Aristotle specifies, with the stakes scaled down and made professional rather than physical. Something is said in a room that several people privately think is wrong, and speaking carries a visible cost while silence carries none that anybody will see. Asch's line-judgment work found roughly a third of responses going along with a confidently wrong group, which is the same transaction with the fear removed and only the social price left. See Conformity. What separates the person who speaks from the person who does not is precisely the disposition this page names, and it is precisely the thing nobody has measured under those conditions.

Commerce: values statements written on an unmeasured trait

Speak-up culture, whistleblower policies, bystander intervention training, upward feedback, psychological safety programmes: a large amount of organisational practice is built on the assumption that a disposition to act at personal cost can be identified, encouraged and trained. The training is a real industry with real budgets. The instrument underneath it is a self-report validated once, on thirty-two people, against approach distance to a stuffed tarantula. That is not an argument for abandoning the practice. It is an argument for describing it accurately, as an intervention on something nobody has characterised under the conditions that matter.

What it does not say

Aristotle does not say courage makes the other virtues possible. He treats it first among the moral virtues in Book III, and the ordering is not a ranking. The famous formulation is modern and should not be quoted as his.

Norton and Weiss do not claim to have measured courage in general. Their title says proof-of-concept pilot study, and a sample of thirty-two people facing an immobile spider is a long way from Aristotle's condition that something be genuinely at stake.

Self-reported courage is not courage. What the 2009 study establishes is that a questionnaire score predicts one behaviour in one low-stakes setting after controlling for fear. Whether it predicts anything when the cost is real is the open question, and it is the whole question.

Nothing here identifies who is brave. There is no validated instrument for that, no normative data for one, and no study following people from a measurement to a genuine test. Any product, assessment or training programme claiming otherwise is claiming more than the literature supports.

Who developed it

Aristotle gives the account this page works from, in Book III of the Nicomachean Ethics, written in the fourth century BCE. Peter Norton, with Brandon Weiss, produced the 2009 behavioural study; Norton is first author. There is no modern originator of the construct in the way there is for most pages on this site, which is part of the problem it describes.

---