BEING SMART AND FEELING GOOD
Published in slightly different form in Tomorrowsf (http://www.tomorrowsf.com), No. 10 (September-October 1998).
Thomas A. Easton
You are visitor 6224
What is intelligence?
To the CIA, it means data and inference, knowing what is going on in sufficient detail to forestall political disaster.
To a biologist, it means something similar. To a biologist, every feature of every living thing is a product of natural selection. That, is, those past members of a species who had any particular genetically specified feature--or who had better versions of it--reproduced more successfully and passed their genes on to future generations. Those who did not, did not. Therefore, present generations have the feature, and generally a version of the feature much improved over what anyone had many generations ago. (Note that "improvement" is relative; natural selection adapts creatures to their environment; when the environment changes, so does the nature of the needed adaptation; thus--for instance--animals may gain and lose thick fur as ice ages come and go.)
What is intelligence? Many people seem to think of it as how much one knows, and many intelligence tests actually test one's knowledge base, at least in part. But it isn't really knowledge alone, for it is easy to find people with little schooling--both children and adults--who are clearly very smart. They learn quickly, integrate what they learn, and apply it promptly to the solution of whatever problems face them.
Is problem-solving the key to a definition of intelligence? It necessarily includes the ability to catch on or learn quickly. Over time, it therefore leads to an enlarged knowledge base. Thus someone who is problem-solving smart must also be fast-learning smart and be or become knows-a-lot smart.
Many people seem to resist this idea, but consider that for a great many years psychology textbooks have been telling students of Wolfgang Kohler's 1913-1917 studies of chimpanzees (see also this). Among the chimps Kohler studied at his North African field station was a male named Sultan, who one day was led into a high-ceilinged room where a banana had been suspended from the ceiling. On the floor lay wooden boxes and a stick. Sultan tried to grab the bananas by jumping. Then he tried to knock them down with the stick. "After failing to get the banana with the stick alone, Sultan sat down 'with an air of fatigue... gazed about him, and scratched his head.' He then stared at the boxes, suddenly leaped up, seized a box and a stick, pushed the box underneath the banana, [climbed upon the box,] reached up with the stick and knocked the fruit down. Kohler was struck with the apparently thoughtful period that preceded Sultan's solution, as well as with his sudden and directed performance."
It is very hard to say that Sultan was not displaying intelligence, and indeed a common test of an animal's intelligence is to pose problems for it to solve. For animals with hands (primates, raccoons, etc.), these problems often take the form of boxes containing food and closed with complicated, puzzle-like latches. The more complicated the latch an animal can open, the smarter it is said to be.
We gauge the intelligence of pets in much the same way. A dog or cat can't open the fridge, but a smart one knows that you can and may come to you, grab your hand with its teeth, and tug you toward the kitchen. Raccoons are even smarter; they're good at opening latches, but their true genius emerges in a story my father used to tell about his boyhood in Bridgton, Maine. A neighbor had a pet raccoon tethered by a chain to the top of a flagpole. The coon could climb up and down the pole and range in a circle around it. And the neighborhood dogs would come to bark at it, positioning themselves in an arc just beyond the coon's snarling reach when the chain was taut. The poor coon just couldn't quite reach those dogs to exact revenge for being tormented.
One day, Dad and his grandfather were walking past when they noted the coon burying some of his chain at the base of the pole. They stopped to watch. When the dogs showed up, they took their usual positions and began to bark. The coon lunged toward them, but only as far as the unburied portion of the chain would let it. It would stop its lunge before it tugged any of the buried chain out of concealment. The dogs soon realized that the coon was on a short tether that day and moved closer. When they were close enough, the coon stopped pretending, the chain came out of the dirt, and the coon got his revenge. It also proved that dogs aren't too bright.
If we accept that intelligence means the ability to solve problems, we must next ask whether this means all kinds of problems, or only certain kinds of problems. Here the biologist must say that an ability to solve only some problems may be fine if those are the only problems around, but life tends to present problems of more than one kind. Those whose problem-solving abilities are limited in degree or kind are necessarily weeded out as the generations pass.
That is, intelligence makes evolutionary sense only if it means generalized problem-solving ability. It is a characteristic of creatures whose ancestors solved all the problems that came their way; those predecessors who missed any one of the problems they faced did not become ancestors.
This applies to people, too, for people are animals as much as are chimpanzees and raccoons. We too are the products of an evolutionary process that defined intelligence broadly. To a biologist, those whose intelligence is fragmented, nongeneral, good for solving only some types of problems, thus do not deserve the label of intelligent.
I put it to you, those we recognize as the most intelligent are indeed very sharp across the board. Those who are intelligent only in certain areas get a "Yes, but..." appraisal.
And now that I've insulted most of my readers, let me admit that the above definition of intelligence may make evolutionary sense, but it's really pretty useless in the modern age. For one thing, humans have an advantage in that we live in social groups whose members may specialize in solving different sorts of problems. For another, this definition doesn't lead to any very good way of testing intelligence, and our society loves to test, measure, and rank to the point where psychologists are being only slightly facetious when they define intelligence as the score on an intelligence test. Intelligence (IQ) tests have their roots in military aptitude tests and are very good at generating scores, but they have severe problems, ranging from dependence on the culture of the test creator to poor prediction of academic, career, and life achievement, partly because they don't recognize "partial" intelligence (excellence in one area but not in others) very well.
Efforts to defeat these drawbacks have included attempts to make IQ tests "culture-free," meaning free of dependence on any particular social context. This issue first arose around the question of racial bias, and a famous illustration of the problem involved giving a standard IQ test to Koko, the first gorilla to learn American Sign Language. As I recall, one of the questions asked the testee to identify shelter from a list of choices including a house and a tree; humans choose the house, and that is indeed the answer rated correct; Koko chose the tree. Another question asked the testee to identify the edible item among apple, brick, flower, etc. The right answer is apple; Koko chose flower, as might any human being aware that nasturtiums, day lilies, and squash blossoms are tasty, or that broccoli and cauliflower are flowers. As for the apple, a botanically knowledgeable person might decide the picture looks more like the fruit of the deadly nightshade and thus reject it. That is, with these tests, knowing too much or having a different knowledge base could lower your score.
Defeating the "culture bias" problem has been difficult, for it is hard to come up with a written test that does not depend to some extent on acquired knowledge, which in turn depends on experience or culture. One interesting non-written approach involved asking testees to press buttons to match a pattern of flashing lights; the more complex and faster the pattern they could match, the higher their score. This approach did indeed seem to test something akin to pure information processing capability, it correlated well with results from other kinds of tests, and it matched up well with our everyday sense that smart people are "quick" and dumb ones are "slow," but it never caught on.
Another approach has been to subdivide intelligence. Howard Gardner's original formulation of multiple intelligence theory listed seven different "intelligences"--linguistic, spatial, musical, logical-mathmatical, bodily-kinesthetic, interpersonal, and intrapersonal. More recent lists add "naturalist" or scientific intelligence. For a slide show review of the theory, click here. For a page of links to numerous related sites, click here.
This approach has had considerable influence in education, for it seems to say that if people can be intelligent in different ways, then they may also learn best in different ways. The result has been much effort to individualize learning plans so that each student is offered material in the way best suited to his or her type of intelligence. No one is downrated for inability to learn in other ways.
But are there really different kinds of intelligence? In "The Genetics of Cognitive Abilities and Disabilities," Scientific American, May 1998, Robert Plomin (of the Institute of Psychiatry in London) and John C. DeFries speak of cognitive skills instead, and stress the major role of genetic inheritance in defining what skills one has. A sidebar by Karen Wright points out that most psychologists now accept "a global conceptualization of intelligence. Termed general cognitive ability or 'g,' this global quality" appears as the correlation of various cognitive skills (that is, people who are smart in one way are generally smart in others too). About the same time, Plomin and colleagues were reporting the discovery of the first genetic marker--not necessarily a gene, but a bit of DNA that is at least close to the gene--for that "g." The marker is found in many high-IQ children, and the associated gene may by itself account for as much as four IQ points. See the May 9, 1998, Science News, p. 292; for critical reaction, click here. The most recent reports were in the papers in June 1998.
Still, IQ remains a bad word among many educators. It is just the score on a test of questionable validity, they say, environment--encouragement, nutrition, etc.--is much more important than any possible genetic component, and a low score can stigmatize a child by labeling that child "dummy." Such labels can act as self-fulfilling prophecies; research has shown that if teachers are told that certain kids have high IQs, they treat them accordingly and the kids do well in school; conversely, if teachers are told that certain kids have low IQs, they treat them accordingly and they do badly.
The theory of multiple intelligences has given educators an escape from this IQ trap by letting them say to a child, in effect, "Your bridge fell down, but that's okay--you're a whiz at tying your shoes." Everybody's smart at something, nobody has to be told they're a loser, and everybody feels good about themselves.
Is feeling good about oneself--self-esteem--important? Research has shown that people who have high self-esteem learn better and accomplish more in school and out. As a result, an entire industry has grown up to help people strengthen their self-esteem -- just try a 'Net search on the term "self-esteem" -- and educators have pushed the idea that we must do nothing to damage children's self-esteem. We shouldn't use corporal punishment or other harsh discipline. We shouldn't give them tests they might score low on. And we shouldn't encourage the idea that heredity has anything to do with intelligence. We should do everything we can to build up their self-esteem, because once the little hellions feel good about themselves, they will buckle down and learn.
You think so? My own history suggests that corporal punishment can have a salutary effect. As a child, I spent two years in France, where one teacher used a ruler repeatedly on a student who kept making errors. That student was sobbing piteously, terribly traumatized by the experience, but I and presumably others in the classroom vowed to bust our little tails to avoid that fate. As the French have said before, you hammer one pour encourager les autres. Alas, that's easy to overdo.
It's also worth pointing out this comment: "The most creative and gifted people are not models of high self-esteem. Their inner lives are often plagued by self-doubt, worries and fears, feelings of inferiority." Perhaps that should be glossed to say that self-esteem is for the mediocre.
And here's a report from the University of Guelph in Ontario, Canada, whose Professor Bruce Ryan, Family Studies, "examined the relationship between self-esteem, achievement, student characteristics and family relationships, [and] found that self-esteem does not independently affect children's achievement when the full context of the child's development is considered. In fact, in study after study that looked at the effects on behaviors such as acting out, internalizing problems, peer sociability and rule compliance, self-esteem was found to be irrelevant."
In May 1998, the New York Times reported that the self-esteem building movement in education is dying, largely because "Research indicates that self-esteem is not in and of itself a strong predictor of success [, and] criminals and juvenile delinquents, it turns out, often have high self-esteem." It is being displaced by "the movement in education toward higher standards [which] has shifted the emphasis toward achievement." That is, there seems to be a shift toward viewing self-esteem not as a prerequisite for learning but as a consequence. First you do something to feel good about. Then you can feel good about yourself.
That makes sense. But it does seem wise to bear in mind that, especially with children, self-esteem can be a fragile thing and it can be damaged by cruel or even clumsy adults. Thus we don't really want to label kids as dummies or pound them with rulers just because they don't satisfy adult expectations.
Nor should we try to define those expectations out of court, as with the theory of multiple intelligences. Call those "intelligences" "cognitive skills" instead, as do Plomin and his colleagues, and consider that a serious problem with efforts to tailor teaching to a student's particular "intelligence" or preferred learning mode may be that such efforts fail to exercise the student's weaker modes and strengthen the ability to learn in other ways. Too, I find that teachers generally agree that the very brightest kids learn well in more than one way--indeed, for such kids the most effective way to teach may be simply to point them at something new.
As for the lesser lights in the classroom, perhaps we should stop pretending that everyone is equal, or that everyone could be equal if only they were taught or coddled or assisted in just the right way.
Some folks are smart, and some folks aren't.
Dr. Thomas A. Easton is Professor of Life Sciences at Thomas College in Waterville, Maine. He has been the Analog book columnist for almost 20 years. His latest novel, Unto the Last Generation, is available only on-line, from Mind's Eye Fiction. Last year's Silicon Karma (White Wolf, 1997) was well received. His latest nonfiction books are Taking Sides: Clashing Views on Controversial Issues in Science, Technology, and Society (Dushkin Publishing Group, 1995, 2nd ed., 1997, 3rd ed., 1998) and Periodic Stars: An Overview of Recent Science Fiction (Borgo Books, 1997).
Previous columns are available at http://www.sff.net/people/teaston/tomorrow.htm .
If you enjoy these columns, you should also enjoy the Science à Gogo site.