There is a glowing review in The Economist of Michael Scammell’s biography of Koestler. I do recall that the manner of Koestler’s death caused a great deal of controversy at the time. My knowledge of Koestler’s work is confined to his Darkness at Noon – given to me my an anti-Stalinist (and committed socialist) friend of mine. The other work I know of his is the very different The Ghost in the Machine (though thoroughly unfashionable in mind circles, it has had some popularization through a well-known beat combo).

The title is of course taken from Ryle’s wonderfully colourful term in The Concept of Mind. Ryle expounds what he takes to be the implications of the Cartesian project: I reconstruct Ryle’s critique as follows:

(i) Ryle rejects the idea that it seems that all intelligent performance involves ‘conscious’ thought, which typically involves the observance of rules, the consideration of propositions and the application of criteria. Were this the case, to do something, would always be to do two things – to first think about rules, propositions and criteria, and then to put into practice what they enjoin.

(ii) This gives the impression that the mind is a storehouse of representations – the “intellectualist legend.”

(iii) The combination of the assumptions that theorizing is the pre-eminent activity of minds and that it is a private operation, amounts to the postulation of a shadowy additional metaphysical entity – the dogma of the “ghost in the machine.”

(iv) This has the further consequence in that it requires the positing of a “central theatre,” some central place in the brain where something like an “I” or the self attends to and witnesses consciousness.

(v) The positing of some central authority or homunculus gives rise to “Ryle’s regress”: an observing self must necessarily contain another observing self, and so on ad infinitum.

(vi) Tempted by our language, a folk anatomy posits the mind as an object made of an immaterial substance (because predications of substance are not meaningful for a collection of dispositions). Thus Descartes allocates concepts to logical types to which they do not belong – hence the “category mistake.”

Ryle’s corrective is summarized as follows:

(i) Ascription of intelligence is to describe behavior, not to name an entity.

(ii) Intelligent conduct of serial operations does not entail that the agent is throughout the progress of the operation conscious both with what he has completed and with what remains to do. The careful driver does not plan for all possible contingencies. His readiness to cope would reveal itself were an emergency to arise but it is latently there even when nothing critical is happening.

(iii) Misunderstanding is a by-product of knowledge-how – mistakes are exercises of competences.