News You Can Bruise for 2009 January 5

Permalink [Comments] (16) Mon Jan 05 2009 23:05 The Origin Of Consciousness In The Breakdown Of This Really Long Title:

I'm reading clusters of related books: Consciousness Explained, I Am A Strange Loop, now Origin of Consciousness Yada Yada. I knew about this book since it was name-checked in The Big U, and afterwards I think I gathered that the same concepts had been deployed in Snow Crash, both of which books I want to re-read now.

But my interest was seriously piqued when I read Richard Dawkins referring to Origin as (non-exact quote) "one of those books that is either complete brilliance or utter rubbish". There are many such books, but most of them have been classified by now, and usually as "utter rubbish." I was intrigued by this thirty-year-old book whose ideas never became mainstream but which could still stand up to someone like Dawkins. If there are any other such books lying around I would like to hear about them.

It's a fascinating book because even if it's totally wrong, it's an excellent work of science fiction. And reading these books in rapid succession it feels to me like the bicameral mind theory is compatible with, or even a special case of, Dennett's Multiple Drafts theory of consciousness. And what do you know, Dennett has written an essay about bicameral mind theory which ties it together with his own writing in about the way I expected.

Anyway, the cluster-reading continues, as I now need to read a bunch of books about game design for the upcoming big project. I've experimented in the past with choosing a next book that has some relationship to the book I just read, but it didn't work because I was trying to make a chain chain chain, chain of books. Clusters make more sense. For instance I was just going to read Rules of Play next, but also on my bookshelf are Dungeons & Desktops and Magister Ludi, both of which I think will be important to this project, so I've put the three books in a stack. A STACK, I tell you!

literature philosophy

Comments:

Posted by Evan at Mon Jan 05 2009 23:55

I keep my books in named pipes. Thats how I roll.

Posted by Zack at Tue Jan 06 2009 02:10

The problem I always had with Jaynes (and keep in mind that this is not having read the book, only criticisms and summaries of it) was that it posited a revolutionary change (from "bicameralism" to "consciousness") at a totally implausible point in history. He has to do that in order to adduce evidence for bicameralism from things that the Romans said and stuff, but then you can come back at him with: how is it that when the Europeans went out and conquered the world, in the sixteenth and seventeenth centuries, they found nothing but conscious humans?

Posted by Leonard at Tue Jan 06 2009 08:25

Actually, he claims (something like this) that the big Mesoamerican civilizations went through boom-bust cycles of bicameralism and something like our consciousness, and that the Incas were bicameral when the Spanish encountered them. If you accept the basic argument, then some of the pre-contact North American societies (as seen in "1941") also look bicameral.

He claims the breakdown is triggered by stresses like invasion and increasing societal complexity. I don't think all the pieces fit together terribly well, but the point he chose for the breakdown in Eurasia (1000 BC) doesn't seem too implausible. If anything I was thinking it was a little too convenient that it happened right before there would have been a lot of written evidence for it, and that anything that would lead to really good evidence (such as widespread literacy or contact between bicamerals and consciouses) also triggers the breakdown. There's at least one convenient destruction of physical evidence during the transitional period, too.

Like I said, the pieces don't fit together terribly well, but I don't need much convincing to believe there were transitional states as consciousness was being built as a virtual machine atop our brains, and Jaynes makes a decent case that there could have been a transitional state within historical times where the two hemispheres of the brain had separate abstraction layers.

Also, and this is important to me in a meta sense, he's pretty humble about what he considers to be his discovery. He doesn't say that bicameralism is an inevitable stepping-stone on the way to "consciousness" (except at the end, in self-deprecation, where he's making fun of theories that make that kind of claim), or that "consciousness" is the last word in abstractions of brain function. He posits many variations of bicameralism over time and in different places, all of which eventually got wiped out or suppressed by the invasive species of today's consciousness, which might itself get wiped out in the future by something better suited to the times. And I don't think he'd object to the idea that some old societies might have run on some third kind of abstraction; he'd just say that bicameralism was the most common.

Posted by Holly at Tue Jan 06 2009 09:22

Psch, you're just being cruel now, all this "I have a REALLY INTERESTING PROJECT coming up" and "it's a bit like The Future: A Retrospective but EXCITINGLY BETTER" and "I'm reading many books on game design for this REALLY EXCITING SECRET PROJECT THAT YOU DON'T KNOW WHAT IT IS".

Posted by Leonard at Tue Jan 06 2009 09:26

Holly, ask Kevan. I told him about it.

Posted by Zack at Tue Jan 06 2009 13:35

If that's what he says then the summaries and criticism that I read are seriously misrepresenting his argument. I'd come away with the impression that the transition was a one-time, irreversible thing and at least partially laid down at the genetic level -- to be true under that constraint, it would have to have happened no later than the Neolithic cultural explosion (at least 50,000ybp and constantly getting shoved further into the past)!

If "bicameral" and "unicameral" (let's not use the contested word "conscious") are both possible operating modes for a genetically modern human brain, with the pattern presumably set by infant experience, and perhaps even possible for a single individual to flip between the modes, it becomes a much more interesting hypothesis, one you can connect all sorts of anthropological and psychological data to. Eliade's studies of Siberian shamanism seem directly relevant, as does schizophrenia.

Posted by Leonard at Tue Jan 06 2009 14:38

Huge chunks of the book, the most enjoyable parts IMO, deal with anthropological and psychological data. Schizophrenia, religious ecstacy, hypnosis, etc. are fit into the model of people whose cultures went unicameral some time ago but whose brains still have exposed RS-232 ports left over from the bicameral days. Anthropological data and literary analysis (a red flag of crackpotism) are used to narrate the transitions from bi- to unicamerality in Greece and the Middle East, and to a lesser extent in Mesoamerica and Egypt.

There was a little genetic stuff that I didn't think was believable because of the timescales involved, as you said. There's a _lot_ of stuff in this book, some of the factual statements failed my web-searching spot checks (look at the Wikipedia entry for the Draw-A-Person Test), and in general it's a mess. But unlike just about every other crackpot-sounding book I've ever read, it feels like there's something there.

Posted by Anonymous at Wed Jan 07 2009 04:29

homo ludens

Posted by Leonard at Wed Jan 07 2009 08:10

I read Homo Ludens a couple years ago, before I started using LibraryThing.

Posted by Holly at Wed Jan 07 2009 13:20

Aha, let me now add to everyone else's frustration by saying that yes, that sounds like a very lovely project.

Posted by Jim Beauregard at Sun Jan 11 2009 07:38

I've always felt the big unsolved problem in consciousness, the one which has practical consequences, is questions like "is it possible that we are [a digital simulation, characters in a book, a dream, etc.]? If not, why not?"

Does any of the stuff you've read help to answer this? I really find it confounding, but I get the impression dennett and jaynes are not even aiming to solve it, and so I've not read much of their stuff.

Posted by Leonard at Mon Jan 12 2009 09:45

At the beginning of "Consciousness Explained", Dennett talks about hallucinations, about the difference between hallucinated and normal perception, and makes an argument against the proposition that our entire life is a hallucination or a product of brain-in-vat trickery.

The argument is fairly convincing but not airtight, and it also includes the assumption that the real universe is something like the simulated universe--that maybe you are a brain in a vat but the vat inhabits a world of human beings with brains who build things like vats. What if the world is very different?

This is the radical skepticism that begins Descartes's Meditations and there is no defense against it, because the only evidence the radical skeptic will accept is the stuff in meditations 1 and 2. I've talked about radical skepticism before.

As a generalization, there's also no way to disprove the idea that the whole universe is a simulation within some meta-universe. Unlike the brain-in-a-vat idea, this idea even sounds plausible. Our universe might appear in a meta-universe as a strange cosmological object like a supermassive black hole, or it might just be a simulation inside whatever passes for a regular computer.

The problem is there's no way to know one way or another. Even if you found evidence that you were a brain in a vat, the radical skeptic would disregard that evidence too! Because it was acquired through unreliable sense impressions, the same as the previous evidence to the contrary.

So no, Dennet and Jaynes aren't aiming to solve this problem, because 1) it's not solvable, and 2) neither of them would consider it a problem of consciousness. It's a problem of the nature of reality. If the universe is a simulation inside a meta-universe that doesn't mean we're not conscious. Though Searle might disagree.

It's pretty safe to say we're not characters in a book, because the information in a typical book is too low-resolution to support consciousness--see here for my thoughts on that.

Posted by Jim Beauregard at Mon Jan 12 2009 11:46

It doesn't intuitively SEEM like we could be characters in a book, but I have a hard time proving it. For instance, the conversation we're having now could be had by two characters in a book, and for them it would obviously be invalid, so how do we know it's valid for us?

Posted by Leonard at Mon Jan 12 2009 14:47

Well, I know that I'm conscious--that's one of the incorrigible facts from the early Meditations. So believing that I might be a character in a book would mean believing that there's some book that describes my mental state in enough detail that reading the book embodies my consciousness and experiences.

This would be a book even bigger than the book in the Chinese Room, but it's theoretically possible. You could imagine some superhuman capable of imagining, say, Huckleberry Finn in such detail that you would say the fictional character had consciousness inside this person's mind.

But this is just another way of saying that I might be a computer simulation. Ordinary books don't have this much information in them. A book that could actually embody the consciousness of a fictional character wouldn't qualify as a "book" by any normal standard.

I'm going to hit "Post comment" and my thoughts will become words on a page, words that a fictional character could have written, but words are only the residue of consciousness. Hundreds of years from now, once I'm dead and these words are all that's left, it could conceivably be difficult to determine whether I was a real person or a fictional character. But that's because my consciousness will be gone. This is why I say that dead people and fictional characters are similar.

Note that if I'm a brain in a vat, everyone else I think I know and talk to is actually a fictional character. This means that the most I can hope for is proving to myself that I'm not a fictional character. And the fact of my own consciousness is almost the only evidence I can take out of the first two Meditations.

Posted by Jim Beauregard at Tue Jan 13 2009 12:07

Is consciousness an objectively judgeable property, like critical mass, or a semantic distinction, like the difference between a stream and a creek?

If the former, what is the external change that we witness when a mind becomes conscious?

Posted by Leonard at Tue Jan 13 2009 16:22

My current opinion is that so far as "consciousness" means anything, it has to do with the amount of introspection possible within a feedback loop. So a rock would not be conscious at all and a thermostat would have the lowest possible level of consciousness. This would make consciousness an objectively judgeable property and also a matter of degree.

In the absence of magic MRI machines, we judge consciousness through the Turing Test technique. I hold the operational belief that there's some conscious being on the other side of these "Jim Beauregard" comments, because we're successfully having a very abstract and introspective conversation. I wouldn't be able to have such a sophisticated conversation with an eight-year-old, and I'd need to dial it down even more for a three-year-old, because their consciousness is still being formed.

Now, maybe you're a fictional character; maybe "Jim Beauregard" is a pseudonym or I'm hallucinating or a brain in a vat. But even so it looks like there's something conscious asking these questions, even if it's a Cartesian demon or my own hallucinating mind.

Incidentally, this is how Jaynes can hypothesize that people once built buildings, spoke to each other, and made up epic poetry without being "conscious". Such people are intelligent but lack introspection. They wouldn't understand questions about their mental states and wouldn't be able to justify their actions in ways we'd understand. I haven't read it, but I believe Peter Watts in Blindsight has some aliens that are like this.

Leave a comment