I can't remember exactly when I wrote this, but I recall sitting in a yakitori joint in Tokyo in August 1998 reading the book, so I must have posted it to the PandA mailing list shortly after.

"The Origin of Consciousness in the Breakdown of the Bicameral Mind"

by Julian Jaynes

A thick book (details) with a starkly monochrome cover, I'd rate this a "good read". It's quite heavy going, and to review it properly I think you'd need a degree in psychology, another in ancient history and languages, plus some. His basic thesis is that "consciousness" is a modern development, arising in the last 3000 or so years.

I can probably only give a gross caricature of his idea, but it suggests that the age preceding consciousness, the bicameral era, included the world portrayed in Homer's Iliad and the beginning of the Old Testament. In that age, one half of the brain (left) organised the ordinary business of doing things, eating, seeing, and ordinary sort of stuff, while the other (right) half did more, long-term sorting out of strategy. The brain's owner more or less trugged along on autopilot under the direct control of the left side until a totally new or unexpected situation, (in which we would use conscious thought) occurred. At this point, if the right half had any good ideas, it communicated them to the left, in a linguistic way. In other words, of course, the autopilot heard voices, giving instructions on what to do next.

The book is very historically based, and he draws on much indirect evidence to support the bicameral hypothesis, and the subsequent breakdown. (He does also, though, relate the idea to modern evidence of what happens in schizophrenia, which includes the same sort of hearing of voices.) His suggestion is that gods are basically projections of voices heard in this bicameral era; that much of the Old Testament is the story of coming to grips with the fading of the voices, and lamenting thereof.

So what sort of characterisation does this give to consciousness itself, which is after all what interests me most? It means for a start that consciousness is mediated by language. He starts with a list of things consciousness is not: it is not perception, but much more like the ability to narratise bits of perception. He gives an interesting example of how consciousness is not a simple "copy of experience". Remember the last time you went swimming. You probably remember a pool or the sea off a beach, and 'see' yourself swimming, as though you were someone else observing. Moreover, your conscious self does not have access to your complete set of memories, so it isn't necessary for memory. Just think about all the places you know you can drive to on autopilot, yet can't run through in your head. Even reasoning isn't necessarily conscious. Here's a sequence of numbers:

    11  7  11  7  11  7  ?

What is the next number in the series? How did you arrive at the answer? I find very convincing his claim that actually you don't have a clue how you found it. You were conscious of looking at the problem, and (presumably almost immediately) the answer popped into your head. The closest you get to a story about what happened in between is a story you made up after the event.

Now a few doubts. If you accept a number of currently widely touted hypotheses, you soon get stuck. If consciousness is the ability to narratise, to go through 'in your head'* the plan of how you're going to move the box underneath the too-high bananas, climb on it, and thus enjoy them, then this is just what chimpanzees do; yet (at least some of the time) we are supposed to agree that chimpanzees do not have language. Almost equivalently I might ask whether what the right half of a bicameral mind, generating more abstract plans, doesn't necessarily involve this kind of narratization, and thus be conscious. This would leave the bicameral hypothesis as an interesting complication, but unfortunately mean that nothing had been contributed to the understanding of consciousness itself.

* Of course, he points out that consciousness isn't really located anywhere; some ancient Greeks might say "in your cardiac organ."

I'm uneasy about the way he invokes evolution as a sort of helping hand to the change from bicameral operation. Apart from anything else, it would seem to require exactly parallel genetic changes happening all round the world, which seems implausible.* I think the historical reasons for the breakdown of bicameral society -- increasing community sizes, and a few historically documented disasters, causing huge numbers of people to be thrown suddenly into situations where the old, largely auto-pilot behaviour falls to bits -- are much more plausible, if the biological capability of unicameral, conscious thought is already present.

* By the way, I've never understood how the, what's it called, the opposite of the "out of Africa" hypothesis would work. It seems to require homo whatnot scattered round the globe to undergo neatly parallel chance mutations in the same direction, while remaining totally interbreedable. Mystery.

A final quibble. He says pretty much zilch about China. This strikes me as a big gap.