The expression `in my head’ is only a metaphor when used in sentences such as `I heard the tune in my head’ and a deeply misleading one; because in no sense could the tune be found in that location. The metaphor is tempting because of the ways in which it is possible for us to experience an apparently unique phenomenological texture based on a common stimulus: viz. when I hear or derive the impression of a tune from the sound of a train. In addition, sounds I hear may have an external source but I appreciate them via internal effects.
I am able to exclude external visual stimuli by closing my eyes; this is ineffectual in the case of mental imagery. Therefore we are tempted to employ the metaphor for the latter. This then tends to produce the error of reification of the location of mental imagery by analogy with the physical location of external seen objects.
If intelligent performance is a single operation, as opposed to dual physical/mental operations, then a way of distinguishing intelligent from unintelligent performance must be found which does not rely on physical distinctions. This is because the same physical motions could be described as intelligent or unintelligent in different contexts. And that is done by noting that an intelligent performance is an exercise of knowing how.
Intelligent capacities are not habits; the ability to calculate mathematical products is didifferent from the ability to recite times tables from memory. The latter type of action cannot be done intelligently (although it can be done badly) – it is automatic. Habits are learnt by mindless repetition; intelligent capacities are the result of training and can go beyond the instruction. Both will lead to different dispositional properties.
We must look beyond the physical details of a performance to establish whether it is intelligent or not; but we must not attempt to look into a hidden mental realm. The context and dispositional properties of the performer are more important. Single shot performances could be hard to judge in this context: we would need to see more operations in order to distinguish luck from skill. We would not allow that a drunk man playing chess who happens to play a brilliant move is a good player. We would instead consider his dispositional properties which would be such that in similar situations he would be unlikely to play a similarly good move. Crucially, the question as to whether his performance was intelligent or not would not hinge on whether ghostly processes accompanied the move. Know how is dispositional whereas habit is one-shot.
Similarly, someone producing an intelligent argument is not doing so by habit or consideration of rules. This is because the situation is too fluid and flexible to allow for a rules-based approach and also people had the ability to argue intelligently before Aristotle produced his rules of logic. This is true of other intelligent operations such as boxing or surgery. Description of mental processes does not relate to shadow operations but to parts of physical operations. The mind is not located.
Someone not knowing the rules of chess could watch a game in that he could observe the players making moves. This observer would not however have any ability to assess the quality of the players. That type of ability would not consist in inference to hidden mental operations of the players. This can be seen by noting that such inferences could never be confirmed except by other inferences of the same type.* It would also mean that we could never know whether people were idiots because their behaviour would have `come adrift’ from their putative occult causes.
People are not like signal boxes in two ways: we cannot see the levers and they are not standardised. Understanding whether someone is good at an activity is related to whether the observer knows that activity itself and not related to understanding other mental states. If I know how to play chess I also know how to say whether you are good at chess.
An ability to perform an operation well is not the same as the ability to describe how it should be done.** Otherwise, historians could not comment on the doings of Kings without being regal themselves. The solution proposed that I can understand Plato by mimicking his mental operations cannot hold because I am less intelligent than Plato and not alive in ancient Greece. A claimed form of `resonance’ or `harmony’ between minds falls to the previously argued point about signal boxes: we could never know whether the resonance holds between hidden states.
Learning how is again not like learning that because the former takes a long time in comparison with the latter.
Misunderstanding someone is not a case of imputing the wrong mental motivations to him; it is to mistake one exercise of a capacity for another.
The dogma produces solipsism, which alone should count against it. In fact, I can see that other minds exist merely from observing behaviour.*** Some mental processes may not be accompanied by any behavior: these can only be discovered by testimony. But these are a small fraction of the whole.
*This is an analogy to Wittgenstein’s Private Language Argument.
**This is a further example of the know how/know that distinction: the ability need not be capable of being broken down into a set of propositions.
***Ryle would be most desirous to distinguish this from behaviorism. That doctrine holds that the mental does not exist; Ryle claims alternatively that there is nothing to mental processes that is not exhibited in behavior: he does not deny the existence of mental processes. But the next point acknowledges a difficulty for him.