In the first two chapters, Paley summarises a huge pile of studies in experimental psychology that go to prove how "the unconscious does what the conscious thinks the conscious does". He gives potted accounts of many studies of circumstances which were found to affect subjects' memories and decisions without them consciously knowing it.
Witnesses to a video car-crash more often misreported seeing broken glass when they were asked about the cars that smashed into each other than when asked about cars that bumped. Subjects did not realise they more often chose Dasani branded water based on product-placing they could not consciously recall noticing. Academic psychologists learned to do better at a seemingly randomised computer game without knowing they had learned anything at all. What you consciously hear can be changed by what your eyes see. Subjects holding pens between their lips so they can't smile think a cartoon is less amusing than subjects holding pens between their teeth. Shoppers have a bias towards goods on the right-hand side of a stall, yet they explain their preference in terms of the quality of the goods, even when all the goods are identical. And so on and on.
The most interesting experimental finding which was new to me, was that autistic children who fail to appreciate that other people can have false beliefs, also fail to appreciate that they themselves previously held a false belief. In other words, their theory-of-mind or mind-reading faculty that malfunctioned when turned outward on other people also malfunctioned when turned in on themselves, as if the same faculty that interprets the minds of others also interprets our own. That is huge. (Reference - pdf.)
There is a brain-zapping experiment with a Transcranial Magnetic Stimulation wand that I won't get into, and a very persuasive experiment on somebody with their brain hemispheres disconnected. (I should probably check out some of these studies for myself, because the relentless flood of findings is highly persuasive.) Paley's evidence bears out his statement:
If consciousness determined our behaviour we'd already know that was how we worked; but as it doesn't, we don't.The next question is, "Why do we have the conscious impression that we know why and what we are doing?" Why do we "infer a decision to act from the fact that we have acted"? Why do conscious minds bother to do this?
Paley's theory is interesting (although I can't tell you how original it is). To summarise it in my own words, he proposes that, just as we have a theory-of-mind program running in our minds that models the minds of other people, so we model our own minds in the same way. This lets us know how other people might be modelling our minds. Which let's us predict how they will respond to our actions, since their reactions depend on their assessment of our motives. And if we can predict how they will respond to us, then we can better manipulate them and thus acquire a selective advantage that will promote the development and spread of consciousness as an adaptation.
As Paley puts it:
The role of consciousness is to explain plausibly why we made the decision we did in the way that someone else (with an equal lack of knowledge about how decisions are really made) might also explain our choices.And:
Our conscious model of our self is primarily for adjusting how we are perceived.The evolutionary logic is intriguing. As a higher animal evolved a theory-of-mind faculty for modelling in its own mind the beliefs and intentions of its conspecifics, there would be a corresponding selective pressure for any given individual to model those conspecifics' models of its own mind. How better to do this than by reflexively applying the theory-of-mind faculty to the behaviour of the individual itself—even though, nay because, this would imply turning the faulty, misperceiving interpreting faculty back on oneself?
My model of your model of my mind is consciousness.This would neatly explain why a conscious human mind does not accurately model the mind's own knowledge and motivations, but rather interprets and models itself as if from the outside. That is, as if it were an outsider. Consciousness should be as bad a mind-reader as the externally directed theory-of-mind faculty. (Why the theory-of-mind faculty evolved to be inefficient I will leave for another day, but it's another vital question.)
The theory makes sense on the face of it. It is almost shockingly, simply, right as an explanation of the evidence. It raises questions, but I think Paley's answers are cogent.
If the function of consciousness is to model one's mind like an outsider does, then why does the conscious mind also receive internal experiences of vision, hearing, taste, warmth, pain, and so on? Consciousness of sensation would seem to be a whole other area of conscious life that Paley's theory-of-mind theory fails to cater for.
Now, people with brain-damage that has caused them to have "blindsight" can perform visual tasks, as Paley says, better than chance without consciously sensing anything visual, and without being able to consciously explain how they are getting tasks right. So consciousness is not necessary to vision, just as one would expect it is not (cf. a robot with a camera). Then why present a visual experience to the conscious mind at all?
Paley's answer is that sensory data needs to be reflexively mind-read because it is often sensorily accessible to other minds too—in the sense that an outsider models whether you are seeing what they can see—and so has to be incorporated into one's model of their model of your mind. We have consciousness of the aspects of our minds that are important to other people. This is a clever extension of the mind-reading theory to accommodate what seemed at first glance to be a separate domain of conscious experience. Paley supports it with evidence from various experiments showing, for example, that subjects can be fooled into thinking they are simply drawing a straight line on a computer screen, when in fact they are adjusting to keep the line straight as the computer subtly rotates its output compared with their movement. The subjects accept what they can see happening over the signals coming from their own body.
It is not a great step to see that a strategic theory-of-mind ought to know what information its own mind has that other minds lack. This could explain why consciousness includes access to our long-held memories as well as contemporary sensations. Not just because they are helpful for developing a more exact theory-of-mind, but in order to make use of the fact that you know things that other minds do not. Paley's theory seems to work.
However, one thing Paley's theory does not really do is to get us any further on the Hard Problem of consciousness, i.e. why we have internal, subjective experiences of anything at all. Could not a faculty as unconscious as that which makes our decisions for us also handle representation of what other people construe our minds as thinking? Maybe I am asking too much, and Paley is for the vast majority of the book not trying to answer the Hard Problem. He has a theory of what consciousness if for, not of what it is, or of how it arises from dumb matter. In not attacking the Hard Problem, he fails to solve what I think most people would regard as the greater mystery of the thing. But I can hardly criticise him for not answering a different question from the one he seemingly set himself.
He does make a brief sally against the Hard Problem, though:
The abstract states of emotions, desires, field of vision, etc., which perhaps only exist in our self-model, are naturally experienced. We do not think that there is a need to presume a gap between having these states and being conscious of them. They are our experience.I think that is equivalent to saying: conscious sensations just are subjectively experienced. Which is not really a theory at all as to how they arise. Not that anybody else has ever done any better!
I don't want to finish on a note of disappointment, though. It seems to me that Paley has done a great service in publishing a theory of what consciousness is for that is derived so deeply from experimental findings rather than armchair philosophising. Read it.