Intuition

Understanding that arrives without conscious inference - and the question of whether it is a source of knowledge or a fast summary of experience.

Theme: Knowledge, truth and reason
Tradition: Cross-tradition
Term fixed: 1641

In plain terms

Sometimes you know something without being able to say how. The answer is there before any reasoning, and reconstructing the steps afterwards produces something that was not what actually happened.

That covers two very different things. One is the mathematician's or philosopher's sense: grasping directly that nothing can be red and green all over, or that if A is bigger than B and B than C then A is bigger than C. Nothing is inferred; the truth is simply seen.

The other is the expert's sense: the practitioner who knows something is wrong before they can say what. That kind is not mysterious at all. It is pattern recognition built from thousands of cases, running faster than description - which makes it excellent where the patterns are real and dangerous where they are not.

As a philosopher would put it

Rationalists treat intuition as a genuine source: Descartes's intuitus is the immediate grasp of a simple truth, and it grounds the deductions built on it. Kant uses Anschauung differently again, for the immediate presentation of a particular in sensibility, which is why translations of the first Critique are a minefield.

In contemporary analytic philosophy intuitions are treated as evidence - the reactions elicited by cases, from Gettier onward, are what theories are tested against. Experimental philosophy has attacked that practice by showing that intuitions vary with framing, order and demographics, which does not refute their evidential use but does complicate it.

The empirical literature divides sharply on when the expert kind is trustworthy. Klein's work on naturalistic decision-making found highly reliable rapid judgement among firefighters and nurses; Kahneman's found systematic failure among clinicians and stock pickers. Their joint paper concluded that the difference is the environment: intuition is reliable where the domain is regular and feedback is prompt, and unreliable where either condition fails.

Where you meet it

A lieutenant leads his crew into a house fire, in a kitchen at the back. They put water on it and the fire does not respond as it should. The room is hotter than a kitchen fire warrants, and it is quiet - far quieter than a fire that size.

He orders everyone out, without being able to say why. Moments later the floor collapses into the basement, where the actual fire had been burning underneath them the whole time.

Gary Klein recorded this in his research on how experienced commanders decide under time pressure, and the officer's own account was that he had a sixth sense. What the debriefing established was that two cues had not fitted - the heat and the quiet - and that neither had reached awareness as a separate observation. The knowledge was real, the inference was real, and the reasoning was unavailable to the person doing it, which is what makes intuition both indispensable and impossible to audit.

What is actually disputed

Whether intuitions are evidence.

In philosophy the practice of testing theories against intuitive responses to cases has been the standard method for fifty years, and the experimentalist critique - that responses vary with irrelevant factors - is either a serious problem or a call for better elicitation, depending on who is arguing. Deflationists reply that what philosophers call intuitions are simply judgements, and that treating them as a special faculty overstates a perfectly ordinary thing.

In the practical domain the disagreement is settled enough to be actionable and is routinely ignored. Where a domain is regular and feedback is fast, expert intuition is excellent and often better than a checklist. Where outcomes are delayed, noisy or rare - clinical prognosis, hiring, forecasting - it performs poorly and confidence rises anyway, which is the worst possible combination.

Defined against

Inference Intuition is defined by the absence of traceable steps, which is what makes it hard to assess and easy to trust

Philosophers

René Descartes Intuition as the direct grasp on which deduction is built
Baruch Spinoza Intuitive knowledge as the highest kind, above reason
Immanuel Kant Anschauung as the immediate presentation of a particular
Henri-Louis Bergson Intuition as the method for grasping duration, which analysis cannot reach
Thomas Reid First principles known without argument, and not therefore groundless

Works

Meditations on First Philosophy Where intuition grounds the method
Ethics The three kinds of knowledge, with intuition at the top
Critique of Pure Reason Intuition as sensible presentation rather than intellectual insight
Creative Evolution Intuition against analysis
An Inquiry into the Human Mind First principles as legitimate starting points

Schools

Cartesianism Where intuition is foundational
Spinozism Where it is the highest form of knowing
German Idealism Where the term shifts meaning entirely
Scottish Common Sense School Where first principles are defended
Lebensphilosophie Where intuition is set against analysis

Related concepts

Reason c. 500 BC · Knowledge c. 369 BC · Inference c. 330 BC · Experience c. 330 BC · Wisdom c. 340 BC

Descartes's intuitus is the mind's clear and direct grasp of a simple truth, and it is the foundation his whole method rests on; the modern psychological sense is quite different and much later.