Neither Larry Bird of the Boston Celtics nor Wayne Gretzky of the Edmonton Oilers has a reputation as a mental giant. Yet Bird, the National Basketball Association's Most Valuable Player for the '83-'84 and '84-'85 seasons, and Gretzky, the National Hockey League's leading scorer for the last five seasons, excel at their respective sports for reasons that go far beyond physical prowess.
It is doubtful that either Bird's or Gretzky's special abilities would show up on conventional intelligence scales of "smart" and "dumb." But they would unquestionably fare better if their intelligences were assessed according to a new theory proposed by Harvard University psychologist Howard Gardner.
Gardner argues that human beings possess not one but several different intelligences, one of which is the "bodily kinesthetic" ability of the athlete, the dancer and the mime. Judged by Gardner's standard, Bird and Gretzky would undoubtedly be classed as out-and-out geniuses.
Gardner's theory of multiple intelligences is the latest and most provocative challenge to accepted notions of intelligence and intelligence testing. Intelligence-quotient (IQ) tests and their progenitor, the Scholastic Aptitude Test (SAT), have been the center of controversy for years. In the heated atmosphere of the late '60s and early '70s, IQ tests were regularly condemned as culturally and racially biased. Those opposed to widespread use of the SAT to assess college applicants have argued that high scores on that test have more to do with skill at test-taking than anything else.
But if critics in the past questioned whether standard intelligence tests measured what they purported to measure, new intelligence theorists are taking the argument one step further. They are challenging accepted concepts of what constitutes intelligence itself.
Yale University psychologist Robert J. Sternberg, for example, has evolved a three-part theory of what makes up mental ability. Gardner and Tufts University psychologist David Feldman question to an even greater extent whether intelligence is the single measurable quantity it has generally been assumed to be.
Although these theorists often disagree with one another and therefore cannot be said to make up a particular school of thought, they all share a humanistic approach to what constitutes intelligence and how it might be tested. At a time when artificial intelligence researchers are teaching computers to mimic operations of the human brain and some proponents of the primacy of IQ as the true yardstick of one's capabilities are measuring intelligence by means of brain waves and reaction times, the new theories of intelligence are a refreshing change.
The "father" of intelligence testing was the French psychologist Alfred Binet. In 1905, at the behest of the French government, he designed the first such test -- to ensure compliance with a new law requiring special classes and programs for the retarded.
In 1916, the term "intelligent quotient" -- one's mental age divided by chronological age and then multiplied by 100 -- first came into existence with the revision of Binet's test by Stanford University psychologist Lewis Terman. The Stanford-Binet Test, as it came to be called, rapidly emerged as the prototype for future intelligence assessment. In the 1920s, Terman launched the most extensive intelligence research program of the time, quizzing 250,000 California schoolchildren in order to find a sample of 1,500 boys and girls with IQs above 140. Follow-up studies on the same subjects were conducted over the years, the last as recently as 1972, with another expected to take place soon. This kind of scientific study went far to establish IQ testing as a tool for identifying particularly bright children.
Such testing was grounded in a theory developed in 1904 by the British psychologist Charles Spearman, who argued that every intellectual act, from boiling an egg to memorizing Latin declensions, required general intelligence, or "g." He said that an individual has the same degree of "g" at his disposal for all intellectual acts: if a person was "smart," he was "smart" across the board. It is Spearman's "g," essentially an amalgam of language and logic skills, that for decades we have called intelligence or brain power and that IQ tests seek to measure.
In the '60s and '70s, controversy arose over whether "g" was totally inherited, partly inherited, or totally or partly a product of one's environment. This nature-versus-nurture debate took on great importance as IQ experts argued about whether "compensatory" education programs for poor, mostly black children could have any effect.
The debate turned ugly as proponents of nature and IQ testing in general, such as Arthur Jensen of the University of California at Berkeley, were shouted down as racists and picketed at public appearances. With IQ tests being denounced as racially and culturally biased, the state of California even banned IQ testing of black children. Although the controversy dissipated after a while, the hegemony of IQ testing had clearly suffered a setback.
Still, the concept of a general, single intelligence remained enshrined as a cultural and educational orthodoxy -- that is, until today's generation of theorists arrived to try and break down "g" into multiple parts and, in some cases, to dispense with it all together.
Although Yale's Sternberg believes that IQ tests are "reasonably good measures of what you know and of analytical-, critical-thinking ability," he contends they are "too narrow." "There are lots of people with high IQs who screw up all the time," Sternberg says. "Other people don't do so well on the tests, but they are extremely effective people. I don't think we really care what Lee Iacocca's IQ is."
According to Sternberg, the tests don't measure a number of crucial areas, including identifying problems, coping with novelty, thinking about old problems in new ways and separating out mental processes. Moreover, he says, most IQ tests focus on what you already know rather than how well you can learn. Sternberg suggests that a good measure of one's intelligence might be to go on a trip "to a completely different culture," for the experience would test both the practical and novelty aspects of intelligence.
Although he essentially accepts the traditional notion of "g," Sternberg modifies it in order to accommodate some of the often-ignored aspects of mental ability. To those ends, he has evolved a "triarchic theory" that postulates three components of intelligence.
The first consists of purely internal mental mechanisms, in particular, how one goes about planning and evaluating situations in order to solve problems. The second covers how one functions in the world, a faculty close to what most people would call day-to-day common sense. The third concerns the relationship of intelligence to experience, especially regarding how one deals with novelty -- making a decision about a new job or getting married, for example.
Sternberg believes all three aspects of intelligence can be evaluated, and he is currently developing a test for the Psychological Corp., one of the nation's leading educational testing companies.
The trouble with existing IQ tests is that they fail to measure "good," or rational, thinking, says University of Pennsylvania psychology professor Jonathan Baron. Rational thinking -- thorough and critical examination of issues as well as self-examination -- is the key component of what he terms a "new theory of a part of intelligence," and Baron insists it can be taught and broadly applied.
Such thinking, he says, would be easy to assess by means of an individually administered test. "You just give the student a problem and ask him to talk out loud. Is the kid open to alternatives, to new ideas? What happens if you make a suggestion?"
Sternberg disagrees: "Insight is part of my theory of intelligence, but I don't think it is a rational process." Baron counters that thinking in general almost always follows the same steps: formulating possibilities; weighing evidence; and considering goals. What changes is emphasis; for example, in artistic endeavors, considering intent rather than weighing evidence predominates.
Although Sternberg and Baron try to break down mental ability into its component parts, implicit in the conception of each is the traditional notion of "g," of general intelligence -- the "it."
Harvard's Gardner and Tufts' Feldman take a very different tack. Co-directors of Project Spectrum, a joint research study aimed at developing new ways of assessing mental capability, the two psychologists argue that there is not one but a number of intelligences. In short, they are looking not for "it" but "them."
In his book, Frames of Mind (Basic Books, 1983), Gardner sets forth seven human intelligences or competences. Among them are linguistic intelligence and "logical-mathematical" intelligence, areas IQ tests measure. Beyond these, he lists faculties that traditionalists would never call intelligences in their own right -- musical ability, spatial intelligence and bodily kinesthetic intelligence.
Outraging the testing establishment even more, Gardner adds intrapersonal and interpersonal forms of intelligence, the former roughly equivalent to sense of self and the latter similar to social grace or ability to deal with others. And one of Gardner's major contentions is that you can be "smart" in one or more of these areas and "dumb" in others.
Gardner's ideas emerged from working with brain-damaged individuals, who he found were capable of performing some mental functions but not others, and with child prodigies, who often excelled in one area but showed only average intelligence in others. Feldman also evolved his ideas of multiple intelligences from studying child prodigies. (His book, Nature's Gambit: Child Prodigies and Human Development, is scheduled for publication by Basic Books this fall.)
Given these human variations, how does one determine what is an intelligence and what is, say, a skill or a talent? Gardner says an intelligence should be able to be "triggered" by some internally or externally presented information. For example, a person might use his bodily kinesthetic intelligence to imitate another's movements. Each intelligence also possesses an evolutionary history that might be shared with other species (such as bird song, in the case of musical ability).
Feldman, on the other hand, has only one major criterion: The faculty in question must correspond to a role or profession or task in the adult world. "That limitation is how we keep it the number of intelligences from being a thousand or 10,000 or a million," he says. "You could imagine 100 maybe. But when you're dealing with human affairs, that doesn't seem an unreasonable number."
How to test for each of these intelligences requires as much imagination as devising the theory did, but unfortunately, the notions about testing are not as developed. Gardner is, however, currently working with the Educational Testing Service in an effort to design a method for evaluating students for arts and humanities instruction at the secondary and middle-school levels. He is toying with adapting and broadening the concept of portfolios, used in art schools, as a way of assessing students in different areas.
Overall, the new ideas are making some headway, although Gardner admits he is still something of a maverick as far as the testing establishment is concerned. But, at the grassroots level, he says educators are increasingly interested in his ideas. He recently met, he says, with a group of women in Indianapolis who are starting a school based on the principles of his theory, though he adds, "I don't want my book to be a prescription for what to do. I want to get people to think for themselves."
bybio Neil Miller is a staff writer for The Boston Phoenix, from which this article is excerpted.