Wednesday, May 30, 2007

The evolution of intelligence and the tribal mind

We humans are proud of our intelligence.

We hold competitions to find the smartest of our kind: spelling bees, chess tournaments, University entrance exams, quiz shows, trivia nights, poker, as well as informal tests like riddles, crossword puzzles.
We collect symbols of intelligence, such as degrees and qualifications.
We read literary books (or say we read them) that others think are a sign of intelligence. If we don't read them we put them on our bookshelves anyway, so that others might think that we did, or at least intend to one day.
We take great interest in animals that are said to be "as smart as humans" such as dolphins.

So, given all this self-love of the apparent intelligence of humanity, a puzzling question arises. If intelligence is so great, how come more animals don't have it? Or, on the other hand, if it isn't all that important, how did we come to get it in the first place?

Ants get by very nicely with a minimum of circuitry. And no-one ever accused a sheep of being too smart for its own good. In fact, I have it on good authority from a behavioral scientist who has done extensive research on sheep, that they are indeed as stupid as they seem. (I wonder how you would word such a finding in your abstract? "in keeping with our hypothesis, sheep are not very bright")

Human intelligence is an anomaly. It is a mystery of nature. Some people might say "That's just human arrogance. Animals are much more intelligent than we give them credit for." And it is true that we have found many surprising abilities in the animal kingdom that in previous times would have been unthinkable: Dogs can learn up to 50 human words; pigs can do three-dimensional spatial reasoning; pigs also have complex social interactions; primates build and use tools; monkeys hunt in coordinated groups and communicate with non-verbal gestures (including a clip on the back of the head to a monkey that steps on a twig when the prey is near); chickens have different signals for approaching aerial and ground predators (the prototypical chicken bock-bock-bock-bock sound is actually the signal for "ground predator approaching" - which is usually you!); not to mention the mysterys of dolphin and whale cognition.

But all of this is itself just another example of human hubris. We look for examples of smarts in other animals because we think it is so important. But it really isn't. Steven Pinker uses a hypothetical example. Imagine, he says, that elephants invent science. Elephant scientists write books about how wonderful the elephant's trunk is. Elephants, and elephants alone, have a trunk. No other animal is as lucky! None can claim ownership of this symbol of animal perfection.

It's the same way with intelligence. (actually, in The Language InstinctPinker uses his example in relation to language, not intelligence, but I think it extends). How intelligence in humans came about is a question that has recently been the subject of much debate.

According to one theory of the evolution of intelligence, the story goes like this. Modern humans evolved on the African Savannah. This took place over a period of about a hundred thousand years, during which the environment was stable and reasonably congenial- the twin requirements for us to firstly, survive and secondly, adapt.

Our ancestors wandered around the savannah, fed themselves, avoided predators, and had children. They did other things, too. For one thing, they talked. Language is an innate ability to humans, and arises spontaneously in any human society. But on the whole, there was very little progress: one year was much like the next. Seasons came and went. Children became adults, and have children themselves. Their children lived as they did.

This is the starting point for perhaps the best known account of the evolution of intelligence: the theory of Cosmides and Tooby(2002). They draw a distinction between fixed intelligence and improvisational intelligence. Fixed intelligence is intelligence that you were literally born with (or at least, you are born with the genetic code to develop it as you mature): the ability to throw a rock at a moving target; the ability to identify a suitable mate.
This is how most animals get by. If you take some calves from their mother cows and put them in a separate pasture, those calves will soon figure out without any help that all they have to do is wander around and eat the grass. Nobody has to tell those calves what to do.
Because this environment was so stable, fixed intelligence abilities could evolve and develop, and they would be useful for generation after generation. High-level problem-solving intelligence, was also needed from time to time, and it, too, evolved.

But this is the crux of the matter. Where did this “improvisational intelligence”, this mental imp that leaps from idea to idea, that creates laws and films and bridges? According to Tooby and Cosmides, it emerged from a conglomeration of mental skills, a new ability built out of existing brain functions.

An alternative theory has been proposed by Satoshi Kanazawa (2004), at the London School of Economics. Kanazawa claims that intelligence is not an emergent property of all the different parts of the brain, but a very specific skill: problem solving in novel situations. Kanazawa says that this skill was probably not used as much on the African Savannah as it is in modern society, because most of the time, humans could rely on their fixed intelligence.

As evidence, Kanazawa notes all the skills that are uncorrelated with general intelligence, or g, as measured by intelligence tests. Emotional intelligence, wayfinding, mate selection, parenting, and making friends are all abilities that seem to have no relation to intelligence. That is to say, empirical studies of the relationship between these things and general intelligence tend to produce a correlation of zero. The reason, Kanazawa says, is that these things are all “evolutionarily familiar”; they are problems that have been occurring since the dawn of time.

By the way, wayfinding here is defined as finding your way without any aids at all. Sure, smart people can use a compass and a map, and maybe even follow fences better. But if you drop a bunch of people in the woods with no artificial aids at all, the people who are best at finding their way out are not necessarily the most intelligent.

On the African savannah, people could survive quite nicely if they were proficient in a number of important skills: but intelligence – or rather, “problem solving in novel situations” – was not critical. Kanazawa draws the contrast with today’s society, where everything requires intelligence. All our environments are artificial, and all our skills and tasks are novel, so we need intelligence constantly. And this is why it has become such a prize ability. Kanazawa notes that there are only four things in your environment right now that were probably on the African savannah of long ago: men, women, boys and girls.

All of this makes for an interesting and compelling case. Indeed, it even has the appeal of ‘popping the balloon’, as it were, and reducing the importance of intelligence in overall mental life. However, from this point, Kanazawa has indulged in speculation that undoes him. He observes the data from the 2002 report IQ and the Wealth of Nations, by Lynn and Vanhanan that reported that IQ in Sub-saharan Africa was significantly lower than in the rest of the world. Kanazawa says (2004, p521):

“This, of course, makes perfect sense from my perspective of general intelligence as a domain-specific adaptation for evolutionary novelty. Since our ancestors spent most of their evolutionary history in sub-Saharan Africa, it is evolutionarily more familiar to the human brain than the rest of the world, which is more evolutionarily novel. If general intelligence evolved as a means to deal with evolutionarily novel situations, then it follows that it should evolve more rapidly in the rest of the world than in the ancestral environment of the sub-Saharan Africa."

And with that one paragraph of idle speculation, Kanazawa does himself and his theory a great disservice. His observation is foolish on so many levels. For one thing – it is unnecessary. His theory of “intelligence as a domain specific adaptation” stands well enough without pondering the meaning of the Lynn and Vanhanan report.

But more importantly, it demonstrates an ignorance of the complex environmental and social issues surrounding intelligence. It is no surprise that IQs are low in sub-saharan Africa. IQ is adversely affected by poor nutrition, poor health care and lack of access to formal education: three factors that are present in sub-Saharan Africa. Indeed, the causal relationship between poverty and intelligence is so obvious as to be banal.

Yet here, and in papers since, Kanazawa has argued that the causal relationship is the other way around: intelligence creates wealth, so in his view, poor people are poor – if I may paraphrase crudely – because they are stupid. And with this move, Kanazawa has made himself the living embodiment of the tradition of Eugenics first proposed by Francis Galton. Kanazawa is not a fascist, nor is he advocating any of the abhorrent social policies that stemmed from Eugenics in the mid-twentieth century. But his treatment of the relationship between wealth, poverty and intelligence is, to put it bluntly, naïve.

This is a pity, because his domain-specific theory of intelligence does not require any such position. He has undermined a not-so-controversial idea with highly controversial musings on poverty and race.

And so, (as always in this topic) we come back to the issue of race. It seems that scholarly investigations of intelligence always get pulled back into the whirlpool. Stephen Jay Gould, in his classic book, The Mismeasure of Man, documented the long and ugly history of intelligence testing, and how it was used in attempts to prove one race was superior to another. This was done on Ellis Island, for example, when immigrants landed in America, to separate those who were desirable from those who were not. It was done in the studies of Samuel Morton, who threw out all the large black skulls from his collection – and the small white ones - on the grounds that they were abberations, then used the remaining skulls to “prove” that whites have bigger skulls than blacks. It was done with the children of American Indians, who were given “intelligence tests” loaded questions about pianos and violins and other aspects of western life about which they had no knowledge.

The evidence does seem to be that there is such a thing as g, or general intelligence. The evidence seems to be that some people are simply smarter than others. And the evidence even suggests that there is some genetic aspect to intelligence. But from that point on, our tribal instincts overwhelm us.

It is simply too great a temptation, faced with such data, to draw conclusions that depict people from other races, other societies, in a poor light. This perhaps is due to our deeply ingrained tribal tendencies. Every society is a tribe. Every tribe has an us and a them. Humans were forming tribes on the African savannah, so we can be confident that the tribal tendency is an instinct.
This instinct distorts the thinking of otherwise enlightened individuals.
Deep down, an ancient part of the brain latches on to the idea: Ha! This will show them! This shows that my tribe is the best tribe!

Tribalism is fixed intelligence. It drives friendships and wars.
It seems that even in our modern age, with our learning and our scholarly investigations, we cannot rise above the most primitive parts of our brain.

References:
Gould, S.J. (1981), The Mismeasure of Man.
Kanazawa, S. (2004), General Intelligence as a Domain Specific Adaptation, Psychological Review, 111, 512-523
Pinker, S.,(1994) The Language Instinct: How the Mind Creates Language (Perennial Classics)
Cosmides, L., & Tooby, J. (2002). Unraveling the enigma of human intelligence: Evolutionary psychology and the multimodular mind. In R. J. Sternberg & J. C. Kaufman (Eds.), The Evolution of Intelligence(pp. 145–198).

Wednesday, May 09, 2007

magnet milestone for black hole machine

CERN has reached a milestone. They installed the last magnet in the construction of the Large Hadron Collider (LHC), an underground machine that will be the largest particle accelerator ever built.

A large dipole magnet was symbolically lowered into the tunnel, completing the basic installation of more than 1,700 magnets that make up the collider, which measures 27km in circumference and is scheduled to be commissioned at the end of 2007.

There was no mention of previous setbacks in the press release.
One of the experiments planned for CERN is the creation of tiny black holes. While some people have expressed concern that this may destroy the Earth, experts such as Stephen Hawking have stated that if any black holes are created by the LHC, they will not be dangerous.
There is a question here of risk and the reliability of scientific theory.
One reason to believe that there is no risk is that the same theory that says that black holes could be created also says that tiny black holes are bombarding the atmosphere with high frequency. This could be called the "They're everywhere and they're safe" theory.
On the other hand, if the theory is wrong, then perhaps black holes are dangerous but rare. What's the chance of a hybrid theory turning out to be correct? One where, oh, the LHC machine creates a black hole that destroys the Earth?
"Nil", say the makers of a report on the LHC.
The consensus among the panel members is that there simply is no risk of that happening.