ETHISMOS · HABITUATION
Nobody thinks you can read your way to a physical skill. The conviction evaporates the moment the skill is interior.
Give somebody a document about staying calm under pressure and they will learn the model, be able to describe it accurately, and behave exactly as before.
Nobody would teach a physical skill this way. You would demonstrate it, have them do it badly, correct it, and have them do it a few thousand more times. The conviction that this is how skills are built disappears the instant the skill is something interior — composure, attention, not saying the thing.
Aristotle’s position is that there is no difference. The mechanism for both is ethismos: doing the thing, repeatedly, before you are any good at it.
One failure never gets to the reps. The other does reps that entrench the error.
Understanding the model feels like acquiring the skill, and the feeling is strong enough to end the effort. The person can now explain the thing they still cannot do, which is worse than not knowing about it, because the question feels closed.
Reps are necessary and not sufficient. Practice without feedback consolidates whatever you are doing, including the wrong thing, and twenty years of it produces twenty years of the same error rather than expertise.
There is no hedge in it and no intermediate step. Not by understanding justice, not by valuing it, not by intending it.
The sentence is unglamorous enough that it gets read as a platitude, which is a shame, because it is a specific empirical claim about how interior capacities are acquired and it is mostly right.
His examples are crafts, and he chooses them because nobody argues about how they work. You become a builder by building, a harpist by playing the harp. There is no version where you study building and emerge a builder.
Picture the first day of it. The apprentice is handed a tool and does something with it that an experienced person can see is wrong from across the yard. The thing produced is bad. It is bad the next day too, and for a long time after that, and at no point does anybody conclude that the apprentice lacks the capacity or should have read more first.
That patience is completely ordinary when the work is visible and almost unavailable when it is interior. Aristotle expects the early attempts to be bad. He does not treat it as a problem, or as a sign, or as something to be got past quickly. It is what the early attempts are.
His answer to the obvious objection turns on this. If we become just by doing just acts, must we not already be just to do them? No — the early acts are done on instruction, awkwardly, without the state behind them. That is not a failed version of the real thing. It is the only available route to it, and it is what apprenticeship is.
Which means the standard for a first attempt is that it happened, not that it went well. A great deal of interior work never starts because the first attempt is measured against the settled version.
Peter Gollwitzer’s implementation intentions are sometimes described as habituation with a specification attached, which overstates the relationship.4 Specifying in advance when, where and how — an if-then plan tied to a concrete cue — substantially outperforms a goal intention alone, and meta-analysis puts the effect in the medium range.
The size has come down somewhat as the literature has been examined for publication bias, and the effect is larger for starting behaviours than for stopping them. Both caveats are worth having. It remains one of the better-supported findings in applied psychology.
The mechanism matters, and so does its scope. An if-then plan works by handing control of a behaviour to a situational cue rather than to intention in the moment. That is a mechanism for getting a specific action to happen reliably. It is not evidence that character is built by repetition, which is a much larger claim this research does not address and could not settle. What it supports is the narrow and useful thing: specification beats resolve, for a single behaviour, at the moment of acting.
The first is about scope; the second is about what drilling does to someone.
The case against
Habituation works well for capacities that recur in recognizable form: attention, composure, following through. It works poorly for one-off judgments, for grief, and for anything whose difficulty is that the situation has never occurred before.
The more serious objection is that a person can be habituated into compliance and the result is indistinguishable, from outside, from virtue. Aristotle’s defence against this is thinner than the problem requires: he largely assumes a settled society whose habits are broadly sound, which is exactly what cannot be assumed. Anyone who has watched an institution produce reliable, well-drilled, ethically inert people has seen the failure mode.
He does have an answer, and essay six is about it: phronesis, the perception of what this situation wants, is ranked above the drilled response, so someone who has it can recognize when the habit is wrong. Whether that is enough is a fair question. My view is that it is not, because acquiring phronesis on his account depends on having been habituated well already, which is the thing in question.
Pick one. Composure, noticing when you have stopped listening, resetting after a mistake, saying the difficult sentence. One, not three.
Write the if-then. Not “I will be calmer” but “when X happens, I will do Y”, with X being something you will actually notice.
Then get reps in low stakes. The mistake everyone makes is first attempting the interior skill at the moment it matters most, which is like debuting a technique in competition.
And arrange feedback. Repetition without correction entrenches whatever you are doing. For an interior skill the feedback is usually a person, because you cannot see your own performance of it.
The constraint is the point: one skill, and no requirement that any rep go well.
Nobody would hand somebody a document about a physical skill and consider the teaching done. The same conviction vanishes the moment the skill is interior, and then everyone is surprised that reading about attention did not produce any.
The correction is small and it is mostly about the standard you apply. Demonstration, repetition, correction, and a long period of doing it badly.
And the limit is real: this builds capacities that recur, and it is also how institutions produce well-drilled people who have rarely been asked to judge anything. The next essays are about the part that resists drilling.
There is no route to an interior skill that does not involve doing it badly a great many times. The standard for a first attempt is that it happened.
But habituation is also how you produce reliable, well-drilled, ethically inert people. His safeguard — that practical wisdom can recognize when the habit is wrong — leans on having been habituated well already.
Quotations are from the Nicomachean Ethics unless noted, cited by book, chapter and Bekker number; renderings follow W. D. Ross (rev. Urmson), Terence Irwin or Roger Crisp, and a condensed line is marked “after Aristotle”. Modern research is cited by author and publication.