Print
Discuss
Send to friend
RSS feeds
The brain is a kind of computer, although the more we learn about it, the less it appears to work like the machines on our desks. Paradoxically, this comparison also seems not to go far enough. The brain is just one part of a chain of information processing stretching back hundreds of millions of years. Life is a processor of information. Life is a computer. Evolutionary biologist Richard Dawkins was among the first to use a computer to illuminate its awesome history. His simulations of natural selection demonstrated how a succession of simple mutations can lead to a complex and functional beauty. He put Darwinian evolution into a 20th-century framework with a series of best-selling books.
These attacked the common misconception that evolution is random, that it works like a monkey stabbing blindly at a typewriter and somehow coming up with Shakespeare. Mutation is random but selection has a cruel algorithm: only the fittest survive. We are accustomed, by the enormity of evolutionary time, to see this process as slow but it can produce drastic change in the space of a single generation, as is amply demonstrated on a new CD, the Evolution of Life, based on Dawkins’ ideas (see the review in our CD-ROMs Reviews section).
Genetic algorithms using the same principle are routinely used by programmers, which is to say that “life” and the computer are doing precisely the same thing. The parallels go further: Dawkins has likened the gene to a program and the body to a computer. I asked him to expand on the idea.
He began by restating in computer terms the ideas form what is probably his most influential book, The Selfish Gene (1976), in which he refocused Darwinism on the gene rather than the organism. “The genes are very like computer code,” he told me. “They use a digital code and [they program] the process of embryonic development. The genes are the only things which pass from generation to generation, and that is why, in The Selfish Gene, the genes are selfish and at centre stage… If natural selection is the selection of anything, which it is, it is the selection of the code, the selection of alternative versions of the code.”
The versions are chosen on the basis of the body’s ability to survive and reproduce. “A wing is a device for propagating genes. It does so by the indirect means of keeping birds aloft in the air, which increases their effectiveness in surviving and therefore reproducing. If you ask why the wing is the shape it is: the proximal reason is that it is aerodynamically efficient and the ultimate reason is that it is the most effective shape for propagating the genes…and the genes did the programming of the development of the wing in the first place.”
So effective is this process that you can easily gain the impression that the wing has been designed by the Great Engineer in The Sky, says Dawkins. Even in the late twentieth century this is dynamite, cutting to the heart of conventional ideas of God the Creator. Dawkins, a militant atheist and Oxford’s first professor of Public Understanding of Science, has been accused of making a religion of science. (The result, ironically, is that he is best known as a figure on TV arguing with vicars). I asked him to what extent computer simulations of these processes go beyond being analogues to being the same phenomena. He said: “It is incidental that in our information technology we use a binary code using voltages that are either high or low. The other aspects of information technology are important and they are the same in genetic code as in a computer program, or approximately the same.”
I said I was interested in the nature of the insights we gained from computers. That the graphical projections of mathematical models, which have enabled us to see what we could not otherwise see are as real, in a sense, as our image of the world, an image which is itself only a construct of our biocomputer.
“When you see a solid object like a cup” said Dawkins, “you construct that solid object as a model inside your head, and occasionally you can fool it. That’s what visual illusions are. You can draw a Necker cube, a cube that flips. That’s because the software in the brain takes that two-dimensional retinal image and constructs a three-dimensional model inside the head with it. And there are two equally compatible interpretations of the image and it flips between the two. So…you could say, in a way, that the digital computer is a sort of extension of your brain, doing the same kind of things your brain has been doing all along.”
Many of these computer images, such as fractals and cellular automata, are beginning to help us understand living structures and how they develop. Did Dawkins feel that a general theory of living systems would emerge based on information processing? “I’m a little bit worried about getting too enthusiastic about unifying theories because sometimes there really isn’t that much unifying them,” said Dawkins.
Newton appeared, for more than 200 years, to have come up with an all-embracing theory until Einstein came along with a compatible but far more comprehensive view. Is it possible that Darwinism missed something out, that it would one day have its Einstein?
Dawkins commented: “This graduation from Newton to Einstein is an absolutely favourite [question and] leads us to a general suspicion that present-day science is only a special case of future science. I just wonder a little bit whether that is always true.
“Now, Darwin has a theory…that all living creatures alive today are descended from ancestors who are very different from them. As a matter of fact, we think all living creatures are descended from a single common ancestor which lived some thousands of millions of years ago. That’s not an approximation to some future truth. That’s never going to be any more or less true than it is now.”
I observed that Darwinism is a purely mechanistic explanation. Dawkins replied: “Natural selection is a mechanism and it is certainly likely that, in the future, our detailed view of natural selection and evolution will change. I don’t think it’s quite right that it’s going to turn out to be a special case of something more general.” But did he not consider there might be some kind of meta-reality that would set Darwinism in context? For instance, is current Darwinism not geocentric, rooted on our planet? We surely have to allow for the possibility that it is going to happen on other planets.
“Good. That’s interesting and I do think it’s important,” said Dawkins. “We can do it now: to think about which properties of Darwinism as we know it on this planet are…somehow necessary features of any kind of evolution, anywhere in the universe… For example, will all evolution be Darwinian? Interesting question. I think the answer’s yes, but one could argue about it.”
Alternative forms might be Lamarkian, with parents passing to their offspring the characteristics acquired during their lifetime. “Another question is: genetics as we know it is digital…does that just happen to be true, or is that another universal? Could you devise an artificial or imagined form of life whose genetics was analogue?
“Or, if it’s digital, does it have to be a linear, one-dimensional, array of digital code elements or could it be a two-dimensional matrix?…It could be analogue rather than digital. And does there have to be sex? Well, we know there doesn’t have to be sex, but I’m just adding a list of questions we could ask.”
So how did consciousness fit into all this? Can it be explained in terms of some kind of self-referential programming? “I think that’s one of the most difficult questions we are facing. And not the least of the difficulties is even deciding what it is…
“My bias is towards thinking that it is a product of brains. Brains that have evolved. It is something which seems to emerge under the influence of natural selection, when brains become large and complicated. But I find it very hard to come up with a good theory of the biological role of consciousness. It is easy to see what is the biological role of complicated behaviour. But it is difficult to see actually why it has to be self-reflective, why it has to be conscious.”
But surely all characteristics don’t necessarily have to have a biological role — consciousness might be an accident, something that arises when the brain reaches a certain level of complexity.
“The more complicated [a characteristic] is, the less easy that is to sustain,” Dawkins said. “Some people have tried to argue that…language is somehow an accidental and useless by-product of large brains. I find that hard to swallow. It has to me all the hallmarks of evolved adaptation and I suspect that natural selection has favoured language.”
It seemed to me that this was the first time in our conversation that Dawkins had faltered in his certainties. That “bias” towards seeing consciousness as a product of the brain recognised a possibility that it might come from outside. Was he therefore not accepting at least the possibility of the transcendental?
“I didn’t say it was outside the brain,” he replied quickly. But did he accept the possibility? “I think it is a product of brains,” he repeated. I said: “That is an opinion. That is not something you can prove.”
“It is not something that I have proved. I am not saying it could never be demonstrated,” replied Dawkins.
So if you get consciousness with a certain level of computational complexity, will machines become conscious? “Yes. I think I would have to be committed to that view…I am not saying it is likely to happen in the next 200 years but I don’t think there is anything, in principle, to rule it out.”
Did he subscribe to the view that consciousness is incremental? That there are degrees of consciousness? “I suspect so. But again I don’t know. All our experience of evolution is that everything is incremental and I would think that consciousness is no exception.”
I began to comment: “But say there’s a level at which one bit of life isn’t conscious and there’s another bit that is…” but Dawkins stopped me there.
“The word ‘incremental’ implies that it is a gradual thing and that some forms of life are, say, a quarter conscious and that there are glimmerings of consciousness in other forms of life.”
So at what point might a computer get a glimmering? And how did Dawkins view the debate about whether computer networks can be a habitat for artificial life, seen now in the primitive form of viruses?
“Life doesn’t have to have anything to do with consciousness,” said Dawkins, surprisingly. “But I think that at the very least there’s a very interesting analogy…that there really can be something analogous to viruses on the internet, and that they could evolve. As to whether they are alive, that’s not a question that troubles me because I don’t care to let words be my master…it’s like arguing about whether so-and- so is a tall man. Some people think that you have to be over six feet to be tall.”
I suggested that here was an example of evolution that might not necessarily use the same rules as the ones that produced us. “I think that artificial life on computers is a good opportunity to look at that very question…because we’re unlikely to have the privilege of visiting other planets where there is life. Artificial life on a computer is the next best thing.”
Almost everything Dawkins says makes perfect sense to me, yet I suspect I am far from alone in wanting to prick his scientific balloon. At one point in our conversation I cited the Hindu/Buddhist concept of maya as a religious image of truth. Dawkins was dismissive but on replaying my tape of the interview I realised that he had taken this to be a reference to the much romanticised Mayan indians of South America. This is a pity because “maya”, often translated as “illusion”, more deeply means the world as perceived via our senses — a centu ries-old image of what we had been talking of as a biocomputer model. Trivial, perhaps, but a by no means unique example of the fact that science is not the sole repository of knowledge.
At times, Dawkins teeters on hubris. He says at the beginning of the new Evolution of Life CD (mentioned above): “Our existence was once the greatest mystery. It is no longer. It has now been solved.” Well, evolution does answer many questions, but it is a long way from solving the mystery of life.
All Online Tags: Evolution, Creationism, Philosophy
del.icio.us
Digg this
reddit!
