Thoughts on “Consciousness Explained”, Part 1
February 23rd, 2010This post is part of a little series of thoughts on the book “Consciousness Explained” by Daniel Dennett. I was having a lot of problems the first time through and gave up in a rage, but enough people I respect recommend the book. So to find out if it’s just me and my personal bias, I started to read it again, giving Dennett more credit than before. I plan on commenting on the whole book, but might skip parts I simply agree with and have nothing to say about. If all else fails, I’ll have at least a detailed criticism this time.
Part 1 – Hallucinations
The Brain in a Vat
They say you only get to make first impressions once and oh boy did Dennett make some! The book starts off with a little introduction to the old “brain in the vat” thought experiment. Just 5 pages in and I’m already raging about Dennett’s sloppiness and faulty reasoning. Let’s take it one mistake at a time:
He begins by differentiating between “possible in principle” and “possible in fact”. (As a little side note, he did the same thing when arguing that “free will” still exists in a deterministic world, another topic I completely disagree with him. Our world is not deterministic (it is, at best, probabilistic) and his re-definition of a concept of free will that is useful in practice is very weak. That’s like arguing that, while impossible in principle, I can still measure the momentum of an atom with enough accuracy I would ever need in practice, therefore I can ignore all the implications of quantum physics. A weak excuse to save his own world view instead of facing the weirdness of reality. Also, Aaron Swartz pretty much says what I think already.)
Anyway, he goes on, saying that while an incredibly (or even infinitely) powerful entity could keep your brain in a vat and fool you into believing their illusion, any remotely plausible being couldn’t do so, therefore we can safely dismiss the argument. I’m going to address the plausibility next, but first something about the argument itself.
If you are the prisoner of a powerful trickster, then you can not tell what tools they have available. You don’t know anything about their universe. They main idea of running a convincing simulation is exactly that you do not give the victim any external reference! You do not get to assume that “yesterday was real”, but “today looks different, maybe I was kidnapped by mad neurologists?”. Any information you have ever been given can be part of the simulation; that is exactly the point of running one.
Maybe they have access to infinite energy? Their universe could very well be infinite. You have no way of knowing how many resources they have because, by definition, you can not see their universe. You can estimate a lower bound, but that’s about it. You can not even tell if any property of your simulation is like the world the trickster is in. They can impose any logic, any amount of resources (provided they have more) they want. Want to run the simulation as a finite world? No problem. Impose fake concreteness, enforcing quantization of any property? Makes the source code a whole lot easier! Let information travel only at a limited speed to simplify the calculations? Sure. Because you don’t even have to run it in real time, you can enforce any speed you want, even a faster one than you have in your world! The “real” world could look so utterly alien to us that we would have to call it supernatural. And then all bets are off. But Dennett doesn’t even pretend to address this. In fact, it looks like he isn’t even aware of the literature. This is a staple of gnostic teaching, at least 3000 years old, and he gets it fundamentally wrong. The book certainly doesn’t start on a good note.
But how hard is it really to lie to a human brain? Imagine some human scientists wanted to pull this off, could they do it? Well, sure. Maybe not today, but easily in 20 years. One great simplification they could employ, that Dennett never even mentions, is taking senses away. If you have never experienced something, then you won’t miss it! If I take a fresh brain without memories and never provide it with visual feedback, then it won’t develop vision and never miss it. The necessary complexity of the simulation has just gone down a lot. We know that blind people are just as consciousness as the rest of us and I don’t think Dennett would dare argue against it, so why doesn’t he address this? Nonetheless, there is a limit here, as demonstrated by Helen Keller. If you cut away too many senses, no consciousness will develop. But we don’t need movement, we don’t need vision and we don’t need pain. Sound and speech, plus a few easy parts like smell, should be enough. We could also add touch as long as we limit movement. The human brain is also quite flexible and will adapt to new senses, like magnetism, as long as we can input it. Some body hackers have achieved neat things in that regard.
Even better, you can do this even after the person has experienced a “real” world, as long as you modify their memories as well. There are plenty of documented cases of people losing parts of their brain and not noting it. Losing a whole direction, like “left”, is not that unusual for a stroke victim. They don’t notice at all that they don’t see anything to their left, the very concept is gone. Ask them to get dressed and they only put on one sock. So if vision is too complex for you, just cut it all out. Once technology has improved, you can add it back in again.
To lie convincingly, we really only need to be consistent. If movement and touch is only binary (I touch you or not; you push or not), then the brain will think of it as normal.
Furthermore, we already have brains in vats! There are already complete simulations of neurons. Some primitive animal brains (worms, mostly) have already been simulated! As of 2010, the best we can do are small parts of a rat’s brain, but in less than 30 years, we will be able to do human brain’s as well. So his claim of this being “beyond human technology now and probably forever” is utterly ridiculous.
Strong Hallucinations
Because brains in a vat are impossible in fact, we have a problem with strong hallucinations, he continues. He defines a strong hallucination as “a hallucination of an apparently concrete and persisting three-dimensional object in the real world – as contrasted by flashes, geometric distortions, auras, afterimages, fleeting phantom-limb experiences, and other anomalous sensations. A strong hallucination would be, say, a ghost that talked back, that permitted you to touch it, that resisted with a sense of solidity, that cast a shadow, that was visible from any angle so that you might walk around it and see what its back looked like”. My first reactions to this was: “I had such hallucinations! Multiple times!” But he concludes that they must be impossible, as the brain is clearly not powerful enough to create them. This puzzled me, to say the least. I can understand him here, but my own experience seems to contradict this. In fact, because my hallucinations were so convincing, I was often reluctant to call them hallucinations at all. They were the primary reason why I was a gnostic theist. If I talked to a god, saw it, touched it, had it transform the whole world and so on, how could I possibly have hallucinated that?
Before I address this, a little side note. I didn’t notice it at first, especially when reading Breaking the Spell (a more sensible, but too careful book), but Dennett mentions Carlos Castaneda as an example of someone describing such strong hallucinations and how that fact “suggested to scientists that the book, in spite of having been a successful Ph.D. thesis in anthropology at UCLA, was fiction, not fact.”. And then it dawned on me: Dennett is an exoteric thinker. Let me explain what I mean by this. The terms esoteric and exoteric, in this context, refer to where knowledge comes from: esoteric knowledge is derived from within oneself, while exoteric knowledge is drawn from the outside world. The perceived duality is false, but this is irrelevant. What I mean when I say that Dennett is exoteric is that he looks at consciousness as an outside phenomenon, something you approach like an anthropologist, taking notes of other people’s behaviour and so on. This approach is utterly alien to me. I have always favored the esoteric approach, in which you think of consciousness (and related phenomena) as something that can only ever be addressed in your own mind. The insights of any other person are, ultimately, useless to you. This is similar to the difference between orthodox religions, that value history, authority and literalism (You can only learn about God from his Chosen.), and gnostic religions, that value personal revelations and experiences (You can only learn about God yourself.).
The consequence of this difference is that Dennett seems to me so completely inexperienced about the topic of consciousness. As far as I can tell, he never took any drugs, never meditated, never learned any spiritual teaching or anything like this. How could anyone not do this? I would never trust a chemist that never tried to build a bomb, nor would I ever trust an engineer that didn’t took apart a complex machine (like their microwave or car engine) for fun (and to see if they could put it back together again). Those would be the most natural first impulses for anyone remotely interested in the fields (and not just doing it for the profit), and they would be valuable first insights and opportunities to learn essential skills (like, “don’t get burned” for all three fields I mentioned). For example, Susan Blackmore has extensive drug and meditation experiences, as has Sam Harris and almost everyone else I know that is interested in some aspect of their own mind. I find it really hard to imagine the mindset of a person that wants to understand minds, yet doesn’t start hacking their own one right away. The term “ivory tower academic” never seemed more appropriate.
But back to the book itself. As I mentioned, I was still, at least partially, convinced I had experienced strong hallucinations before. So is Dennett’s conclusion just bullshit? Well, no. He goes on to explain how they actually might come about, and provides a great analogy in the form of a party game called “Psychoanalysis”:
In this game one person, the dupe, is told that while he is out of the room, one member of the assembled party will be called upon to relate a recent dream. This will give everybody else in the room the story line of that dream so that when the dupe returns to the room and begins questioning the assembled party, the dreamer’s identity will be hidden in the crowd of responders. The dupe’s job is to ask yes/no questions of the assembled group until he has figured out the dream narrative to a suitable degree of detail, at which point the dupe is to psychoanalyze the dreamer, and use the analysis to identify him or her.
Once the dupe is out of the room, the host explains to the rest of the party that no one is to relate a dream, that the party is to answer the dupe’s questions according to the following simple rule: if the last letter of the last word of the question is in the first half of the alphabet, the questions is to be answered in the affirmative, and all other questions are to be answered in the negative, with one proviso: a non-contradiction override rule to the effect that later questions are not to be given answers that contradict earlier answers. For example:Q: Is the dream about a girl?
A: Yes.
but if later our forgetful dupe asks
Q: Are there any female characters in it?
A: Yes [in spite of the final t, applying the noncontradiction override]When the dupe returns to the room and begins questioning, he gets a more or less random, or at any rate arbitrary, series of yeses and noes in response. The results are often entertaining. Sometimes theprocess terminates swiftly in absurdity, as one can see at a glance by supposing the initial question asked were “Is the story line of the dream word-for-word identical to the story line of War and Peace?” or, alternatively, “Are there any animate beings in it?” A more usual outcome is for a bizarre and often obscene story of ludicrous misadventure to unfold, to the amusement of all. When the dupe eventually decides that the dreamer — whoever he or she is — must be a very sick and troubled individual, the assembled party gleefully retorts that the dupe himself is the author of the “dream.”
This is, in a way, very close to how some parts of the human brain actually work. Most processing doesn’t start with the facts and derives a hypothesis that it then tests (as science should work), but rather is overeager to find patterns. Instead, you get a face recognition system that is totally convinced that this is a face, no doubt about that! Oh, it was just some toast, oh well. But it totally look like a face! Like the Virgin Mary, even! You just need to slightly disorient this part, or feed it random noise, and it will see faces everywhere, in the walls, the trees, your hand, everything. Or nowhere, of course, depending on the exact disturbance.
And I began to think, if you just disturb a few crucial areas involved in parsing important objects (like faces, intentions, geometric patterns and so on), and this isn’t particularly hard, you really only need to cut off the regular input (as when sleeping), then the narrative parts of the brain are in quite a tricky situation. Their job is to make sense of all that, rationalizing both the outside world and your own behaviour. This is crucial in social situations; you really wanna figure out fast who is plotting against you and whom you can trust. In fact, it is so useful, that even quite a bit of false positives isn’t so bad. Some paranoia or thinking your PC hates you isn’t so bad and can even help you analyze situations (like thinking that “the fire wants to eat up all the oxygen”). Dennett calls this particular analysis the “intentional stance”.
Now, if the narrator is only given (pseudo-)random noise, it will impose any story it thinks is most natural, i.e. most of the time other human(oid)s, recent emotions and so on. This is exactly how dreams work and, in fact, most drug-induced hallucinations as well. The exact distortion and resulting flexibility in making up a good story depends on the drug, of course, and is quite interesting in itself.
But does this really explain my own strong hallucinations? I was reluctant to accept this at first, but now have to agree with Dennett here. Thinking back, and based on the most recent experiments, I am forced to concede this point. I never met an agent, or phenomenon at all, that was able to act against my own will. James Kent, of tripzine.com, describes this well:
However, the more I experimented with DMT the more I found that the “elves” were merely machinations of my own mind. While under the influence I found I could think them into existence, and then think them right out of existence simply by willing it so. Sometimes I could not produce elves, and my mind would wander through all sorts of magnificent and amazing creations, but the times that I did see elves I tried very hard to press them into giving up some non-transient feature that would confirm at least a rudimentary “autonomous existence” beyond my own imagination. Of course, I could not. Whenever I tried to pull any information out of the entities regarding themselves, the data that was given up was always relevant only to me. The elves could not give me any piece of data I did not already know, nor could their existence be sustained under any kind of prolonged scrutiny. Like a dream, once you realize you are dreaming you are actually slipping into wakefulness and the dream fades. So it is with the elves as well. When you try to shine a light of reason on them they dissolve like shadows.
And so I gave up on believing in them, as reality, as Philip K. Dick said, “is that which, when you stop believing in it, doesn’t go away”.
One last thought one the topic, though: Dennett contradicts himself here. If it is so relatively easy to lie to the brain, to convince it to see patterns that aren’t there – and he even provides a mechanism: don’t lie to the senses, lie to the interpreting part – how can he still dismiss the brain in the vat so easily? He has just described, in detail, how you would go about setting up a relatively easy simulation! He really hasn’t thought about the simulation argument nor does he have much experience with hallucinations, and it shows. But the mechanism is clever and I hope he will elaborate more on it.
This covers the Prelude. Chapter 1 will follow soon.