Wednesday, November 28, 2012

Project 42

"Why do you call it Project 42?"
"I see you haven't read The Hitchhiker's Guide to the Galaxy."
"The what?"
"The Hitchhiker's Guide to the Galaxy--Douglas Adams' five-book trilogy."
"Oh, that thing. I heard a couple of the BBC radio episodes back in med school when they first came out. I noticed it came out as a book, later, but I don't have time for fiction. I'm a doctor, not a comedian."
"Sometimes fiction gets ahead of science. Remember Jules Verne? H.G. Wells? Even Galileo turned to fiction when the Church wouldn't let him be a scientist."
"Fair enough. Permit me to revise my remarks--I don't have time for silly science fiction. The whole thing was a spoof, right? Like I'm supposed to take a  five-book trilogy seriously? Besides which, you still haven't told me why you call it Project 42."
"In the books, the Earth is destroyed because it is on the brink of discovering the ultimate question."
"See--that's silly, right there. You'd think they destroy the Earth because it was about to discover the ultimate answer."
"No--they'd already discovered that. The answer to life, the universe, and everything is 42. That's why they had to construct a planet-sized thinking machine to figure out what the question was."
"I assume that planet was Earth?"
"Exactly. Which brings us back to Project 42. What we have assembled here is enough hardware and software to explore the collective intelligence of the planet."
"What do you mean?"
"You've heard of data mining, right? We're mining everything. We've searching through encrypted messages, government reports, conspiracy theory blogs, Sunday School periodicals--everything. We're even using Google's street view software to read graffiti off the walls."
"That's a lot of information, but what do you do with it? Data isn't knowledge."
"We're treating it as a problem in mimetics."
"Mimetics--I can never decide whether that is science or science fiction, but at least it isn't silly."
"Not silly at all. Mimetics treats words like genes, reproducing across information-space the way genes reproduce in physical space."
"Yes, I know the theory.  If I recall correctly, Richard Dawkins suggested it in The Selfish Gene, way back in the 70s."
"You have an unusually good memory!"
"Yes--because I don't waste it on fluff. Dawkins said that genes use organisms to propagate themselves, and he speculated that memes might do the same thing. I remember being startled by the thought that ideas might actually use humans to spread themselves."
"Project 42 doesn't go that far. What we are doing is mapping the mimetic 'DNA' to see if we can find a mimetic 'missing link.'"
"What do you mean by that?"
"In evolutionary biology, we assume that every living species comes from a single common ancestor. Biologists are decoding the DNA of so many living species every day that it's just a matter of time before software can start to tell us how and when the different organisms arose. The information is in there--it's just a matter of teasing it out."
"Sure--I can see how you look for a missing evolutionary link. But how does this apply to memes? We know where memes come from. People just make them up."
"Do they? That's the question. There are lots of memes that might be just made up, or they might be clues to something big."
"Like what?"
"How much time do you have?"
"It's your nickel. I've got as much time as you want to pay for."
"At $1000 an hour, I'll keep it brief. Here are three of the memes we're investigating--ghosts, dragons, and worldwide floods. You and I would agree that there's no such thing as a ghost or a dragon, and there's never been a worldwide flood, right?"
"So what do we find when we explore the 'ghost' meme? Since there are no actual ghosts, we would expect to find a more-or-less random assortment of 'spooky' elements in ghost stories, with some kind of Darwinian selection that conserves the 'spookiest' elements and loses the bits that don't frighten people."
"What we would not expect is a recurring pattern of non-spooky elements in ghost stories across different cultures. That would suggest that there's something more going on than mere imagination."
"Are you saying you're finding such a thing?"
"No--at least not yet. Results are still preliminary. I'm just explaining how mimetics help us explore such mysteries."
"You had me worried."
"Don't stop worrying yet."
"You listed ghosts, dragons, and floods. Keep going."
"Right--dragons.  With dragons, we have enough texts over time to watch mimetic evolution happen. In the old stories, there are dragons that walk, dragons that talk, and dragons that fly--but they are all different creatures. Over time, the 'standard dragon' in the West has morphed into the winged, greedy, fire-breathing monster you see in the Saturday morning cartoons."
"I'm not surprised."
"The thing is, the older texts are the ones most like creatures that have actually existed."
"Not during human history, though."
"How do we know that? Sure, we don't have any fossils that overlap--but we still don't really know what lurks in the deepest parts of the Congo. Who knows when the last of the dinosaurs died? If the first human story-teller saw the last living dinosaur, we might still be hearing about it."
"That's absurd! The gap is too great. There are too many millions of years between them."
"Maybe. Or maybe not. Mimetics enables us to dig where the fossils aren't."
"So much for ghosts and dragons. What was your third example, again?"
"Worldwide floods. There's a meme with legs, for you."
"Right--I've heard about that. Don't most cultures have a flood story?"
"Yes--one reason why these fundamentalists insist that Noah's Flood really happened."
"I thought Noah's Flood had been proved to be a local thing--I thought a couple of geologists demonstrated that the Black Sea was suddenly submerged about ten thousand years ago."
"Funny you should mention that--I never heard about that until the Project turned it up. William Ryan and Walter Pitman discovered that the Black Sea was a freshwater lake at the end of the last Ice Age, with a water level several hundred feet below sea level.  They proved it suddenly filled with salt water around 6000 BC."
"Right--and they said that every farming community was descended from the refugees, didn't they?"
"Yes. Their book, Noah's Flood, was mostly about the geological evidence, but it included a few mimetic arguments for their claim."
"What ever happened to them? I expected that book would make more waves, if you'll pardon the pun."
"I would call them victims of mimetics. Nobody had much of an argument with their scientific claims, but nobody picked them up and ran with them either. The scientific community wasn't interested in repeating the claim that a big chunk of Genesis might actually be a historical event. The fundamentalist community wasn't happy with a local flood, even if the Black Sea flood could be a literal eyewitness account of exactly what happened."
"So mimetics explains why stories don't spread as well as why they do?"
"Exactly--memes don't necessarily reproduce because they're true. They reproduce because they're popular. The more popular the idea, the faster it spreads."
"Do you think they were right about the flood?"
"I don't know--yet."
"But you think Project 42 can help us find out?"
"I do--if we can get the kinks out of it. Which is why you're here."
"Which is why I'm surprised. This sounds like a computer project. Why hire a shrink?"
"Because it's not the computers we're having trouble with. It's the operators."
"What kind of trouble?"
"Three suicides, several psychotic breakdowns, complete catatonic paralysis, and one operator who can't stop shouting, 'I see Jesus! I see Jesus!'"
"That's some kind of trouble! How long has this been going on?"
"It started about two weeks ago and stopped last Wednesday."
"It stopped on Wednesday? What changed?"
"It didn't stop. We stopped letting anybody near the interface."
"So as far as you know, it's just as deadly as ever?"
"As far as we know."
"You didn't have any troubles before two weeks ago?"
"No, quite the opposite. We were getting excited about the results. We had just completed a huge data add of religious materials--everything from the Egyptian Book of the Dead to the Church of the Three-Eyed Toad--and there were all kinds of convergences popping up."
"That's what we call it when apparently unrelated memes appear in different contexts. You see something similar with genes--octopuses have the same kind of eyes that humans have. That either means the camera eye evolved twice (which is hard to believe) or that squids and humans share an ancestor with eyes."
"You were seeing convergences between religions?"
"Yes--all over the place."
"I'm not surprised. I've always suspected all religions basically say the same things--different brand names on the same basic product."
"That wasn't quite what we found. We mapped the memes and came up with something that looked just like the Mandelbrot set."
"The what?"
"The Mandelbrot set--one of the most familiar fractal functions."
"I thought that's what you said. It's that math thing that has all the pretty patterns, right? Seahorses and curlicues that keep reappearing as you zoom in at ever higher magnitudes?"
"That's the one. Although our map didn't have any seahorses--what we saw looked like the basic Mandelbrot plane--a squiggly circular sort of black space surrounded by infinitely intricate patterns."

"I think of the black bit as more heart-shaped."
"Heart, circle, whatever--it was surprisingly similar to our map of religious memes. And if you know how the Mandelbrot set works, you'll understand why that fascinated us."
"I'm a forensic psychiatrist, not a professional mathematician, so you'd better jog my memory on the Mandelbrot thing."
 "The Mandelbrot set is generated by a relatively simple function in the complex plane--you do know what the complex plane is, right?"
"I'm a shrink. Of course I know about complexes."
"I mean the plane formed by the real numbers along one axis and the imaginary numbers along another."
"Oh--those complexes."
"Right. Tell you what, let's skip the math and get to the point. The black circle, or heart, or whatever in the middle of the Mandelbrot set is the area where the function can't function, in a sense. The squiggly boundary around that empty black center is where all the interesting patterns appear."
"And this reminds you of religious mimetics, for some reason?"
"Yes--our map showed religious convergences all around an essentially empty center. Most of the major religious memes were equally far from this black point in the middle."
"I'm not saying they were identically distant. Some Zen Buddhist memes were pretty close to the center. Surprisingly, some Christian memes got even closer--but they were on the opposite side of the map."
"Interesting. But you say that happened two weeks ago."
"Yes, that was when we mapped the existing memes. The trouble started when we tried to find the 'missing link.'"
"Let me guess--you tried to generate a meme to fill the hole in the middle of the map."
"Yes. How did you know?"
"And once you did, your operators started going mad?"
"Yes, they did. But how did you know?"
"Because I'm a psychiatrist, not a psychologist."
"What do you mean?"
"You're a scientist, so you're dedicated to discovering the truth, right? Well, I'm not. I'm a pyschiatrist--in Greek, a healer of souls--not a psychologist, who studies them. When a patient comes to me, I'm not looking for the 'truth' about his life. I help people find a lie that they can live with."
"You don't mean that!"
"I don't usually say that, but I mean every word of it. And that's why I'm not surprised at what is happening. Your machine may reveal the secret of life, the universe, and everything--but who says humans can survive it?"
"I thought the truth will set you free!"
"That's what Jesus said. But in my experience, the truth will drive you mad."

Monday, November 19, 2012

Defining Education

My son is trying to define "education" over at his blog. I agree with Mike that "education" is a "process," but a practicing metaphysician can't resist this opportunity to dig into the epistemological substructure of "knowledge." Let's start with the physical basis of thought.

The human brain is a "neural network" which can be thought of as a hyperdimensional phase space. Since each brain has about one hundred billion neurons, there is a one-hundred-billion-dimensional "space" for each. I visualize that as a very folded, twisty space--a hundred billion dimensional labyrinth. In a purely metaphorical sense, one can visualize the "intelligence" of this neural network as the volume of the space. One can increase the volume by making new connections, converting a twisty tunnel in the labyrinth into a chamber in a cavern.

Another way to increase the volume of the labyrinth is to connect it up to other labyrinths. That's the reason I start my study of knowledge with physical brains and neural networks--the network isn't trapped within one human skull. Two human beings form a single neural network with two hundred billion neurons. Ten humans have a trillion nodes. Our planet currently has half a sextillion living human neurons that are increasingly connected to each other.

And we're not just limited to living human neurons. The "synapse" from one human brain to another is language. That means that a library of books captures trillions more of the best and brightest neurons humanity has to offer. Opening these books and connecting them up to living minds turns the  twisty little labyrinths into vaulted chambers where many tunnels intersect.

There's much more to education than books, of course--but not enough time to get into it now.