2.7 — Consciousness, and the Sense of Being Someone

A patient is having surgery to control severe epilepsy, and the bundle of fibres connecting the two halves of her brain has been cut. She is awake, recovered, and appears entirely normal.

Now a word is flashed to her left visual field, which is processed by the right hemisphere only. Language sits mostly in the left hemisphere, so the speaking half of her brain does not receive the word. Asked what she saw, she says "nothing."

Then her left hand — controlled by the right hemisphere — reaches out and picks up the object the word named.

Asked why she did that, she does not say "I don't know." She produces a reason. A confident, plausible, entirely invented reason. In Michael Gazzaniga's studies of these patients, the left hemisphere reliably manufactured explanations for actions it had not initiated and could not have known about, and it showed no sign of noticing that it was making them up.

Gazzaniga called this the interpreter. It is the part of you that narrates, and this page is largely about the discovery that it is a commentator rather than a captain.

Two different questions

Keep these apart or the subject turns to fog.

The easy problems — easy only by comparison — are about mechanism. What distinguishes a signal you become aware of from one you process without awareness? What does anaesthesia switch off? Why does damage here abolish consciousness and damage there not? These are hard scientific problems and real progress is being made.

The hard problem, named by David Chalmers in 1995, is different. Why is any of it like something from the inside? A machine could in principle do all the processing — detect the red, avoid the obstacle, report on its own state — with nothing it is like to be that machine. Why is there an experience, rather than just information moving?

Nobody has an answer that satisfies people who disagree with them. Positions range from "it will dissolve when we understand the mechanisms" to "it is a genuinely different kind of question" to "consciousness is a basic feature of matter". The honest state of play is that this is unsolved, and any book that tells you otherwise is selling something. What follows is about the easy problems, where there is real knowledge.

What is known about the mechanism

Consciousness is not one thing and it has levels. Wakefulness and content are separable. In a vegetative state a person cycles through sleep and waking with no evident awareness. In locked-in syndrome the opposite holds — full awareness, almost no ability to move. Some patients diagnosed as unresponsive have been asked, inside a scanner, to imagine playing tennis, and have produced the correct brain activation on command, which is a way of answering yes when nothing else works.

Most processing is unconscious, and the conscious part is small and late. Vision, grammar, balance, face recognition, the retrieval of a word — all done for you. What arrives in awareness is a summary.

Awareness looks like broadcast. The most productive current framework holds that a signal becomes conscious when it is made available widely across the brain rather than staying in one specialised region — so that memory, language, planning and action can all use it. This is the global workspace idea, and it does predict things: contents you are conscious of are exactly the ones you can report, remember and act on flexibly, and the transition into awareness is associated with a sudden widespread wave of activity rather than a gradual increase.

And the timing is strange. Benjamin Libet's experiments in the 1980s found brain activity preceding a spontaneous movement by a few hundred milliseconds before people reported deciding to move. This was widely reported as disproving free will. It does not — the paradigm has serious methodological criticisms, the early signal may be ordinary background fluctuation rather than a decision, and a trivial finger flexion is not a model of choosing a career. What survives is the more modest and still unsettling point that your sense of having initiated an action is itself a constructed experience, generated after the fact and occasionally wrong.

The self is a construction, and here is the evidence

This is the part with practical consequences, and there are four independent lines of evidence.

The body boundary can be moved. In the rubber hand illusion, your real hand is hidden, a rubber hand is placed where a hand would plausibly be, and both are stroked with a brush in synchrony. Within a minute or two most people report the rubber hand feeling like theirs, and they flinch when it is threatened. Ownership of your own body is an inference from consistent evidence, and it can be fooled by supplying consistent evidence about something else.

The narrator confabulates. The split-brain result above, and many others. When people are induced to make a choice they did not make — through techniques where a card is swapped without their knowledge — they will explain their preference for the card they did not choose, in detail, with feeling. This is choice blindness. The explaining machinery does not have access to the deciding machinery; it takes the outcome and generates a story.

The self has parts that dissociate. Damage produces oddly specific losses: the sense that a limb is yours, the sense of being the author of your movements, the sense of being located inside your body. Each can go separately, which means each was being assembled separately.

And it changes with the state of the brain. Meditation practice, certain drugs, extreme exhaustion and some seizures can all temporarily reduce or dissolve the sense of a separate self, with experience continuing perfectly well without it. Whatever the self is, it is something the brain does — and it can stop doing it for a while without the lights going out.

What the self actually seems to be

Not a thing. A process with several components, running continuously, and normally so consistent that it feels like an object.

The minimal self — being the one having this experience, located roughly behind the eyes, owning this body. Built from the constant agreement of vision, touch, balance and the internal sense of the body's state.

The narrative self — the character in your autobiography. Made of episodic memory (2.4), and therefore rebuilt and revised every time you tell it. The story you have about who you are is not a record. It is the current edition.

The social self — the model of how you appear to others, which for most people is doing considerably more work than they realise.

Here is why this matters rather than being philosophy. If the self is a process, it can be worked on. A story that is rewritten every time it is retold can be retold differently — that is not positive thinking, it is the mechanism 15.6 and 16.9 rely on. A sense of self that can be quietened is one you can get some distance from, which is the entire skill of Part 7 and the precise thing that makes a trigger stop owning you (5.4).

And it explains the strangest and most useful thing about the mind: you are not identical to your thoughts. A thought arrives, produced by machinery you did not consult. You can watch it. The watching is not the thinking. Every contemplative tradition noticed this, and it is not mystical — it follows directly from the fact that the narrating part is a component rather than the whole.

Free will, in one honest paragraph

The question people mean is: given the exact state of your brain and the world, could you have done otherwise?

Probably not, in that strict sense, if the brain is physical and physics is what it appears to be. But this matters far less than it seems, because that is not the distinction anyone actually uses. In practice we distinguish an action that came from your values, deliberation and character from one that came from a gun to your head, a seizure or a delusion. That distinction is real, it is what responsibility is built on, and it survives intact whether or not the universe is deterministic.

The practical implication runs in a useful direction. If your choices come from your character, and your character is built by what you repeatedly do, then the leverage is in the building rather than in the moment of choice. You will not out-decide a bad system at 11 pm. You can arrange, over months, to be someone for whom the 11 pm decision is easier. That is Part 8, and this is its philosophical licence.

What to do with this

Distrust your explanations of your own behaviour — gently, and specifically. When you ask yourself why you snapped, the interpreter will supply a fluent reason within a second, and it may be an invention. Treat it as a hypothesis. Then look for the physical facts: sleep, hunger, how many demands there had been, what happened an hour earlier. The boring explanation is right far more often than the psychological one, and finding this out about yourself takes a log rather than an insight.

Practise the separation between the thought and the noticing. Sit for two minutes and let thoughts arrive without following them. What you are training is the distinction this page describes — that there is a difference between having a thought and being it. This is not preparation for anything mystical. It is the specific ability that makes "I am furious" become "I notice anger", and the second one is manageable in a way the first is not. Part 7 teaches it properly.

Notice the difference between the story and the events. Your narrative self is an edited account with a theme. Occasionally check it against something contemporaneous — old messages, an old photograph, someone else's memory of the same period. People routinely discover that a period they remember as uniformly bad contained a good deal that was fine, and that a person they remember as always difficult was not.

And take some pressure off. A great deal of self-criticism assumes a unified chief executive who should have known better and chosen differently. There is no such official. There is a large collection of processes, mostly automatic, some in conflict, with a narrator who explains the outcome afterwards. That is not an excuse — you are still responsible for what you do — but it is a far more accurate and far less punishing picture, and it points at the right intervention, which is changing the conditions rather than scolding the manager.

Next: 2.8 covers the third of your life when all of this shuts down and reorganises itself, and what happens to your mind when it does not get to.