Cerebrum Article

A Debate on “Multiple Intelligences”

The scientific study of intelligence began more than a century ago with Sir Francis Galton, an aristocratic polymath whose own I.Q. has been estimated posthumously at 200. A cousin of Charles Darwin, Galton became fascinated with inherited abilities. In 1869, with the publication of Hereditary Genius, Galton advanced the thesis that intelligence was an inherited human characteristic. He also brought into play the statistical approaches to measuring it that have dominated psychometrics—the science of measuring mental attributes—ever since.

Published: October 1, 1999

At the beginning of this century, a Frenchman, Albert Binet, director of the Sorbonne psychology lab, introduced widespread testing of school children in Paris to ascertain their intellectual abilities. Binet and his colleague, psychiatrist Theodore Simon, gave us the concept of a “mental age” and the nascent industry of intelligence testing. To round out the story, a German psychologist, William Stern, suggested that an “Intelligence Quotient” could be calculated by dividing mental age by chronological age (and multiplying the result by 100 to get round numbers). To emphasize the supposed unity of the trait being measured by intelligence tests, psychologists came to refer to it as “G” for general intelligence.

Summing up these early years of psychometrics, Jack Fincher, in Human Intelligence (G.P. Putnam, 1976), wrote: “The siren song of an intelligence that was stable, distinct, and independent caught the public ear as few seductive sounds from science have before or since.”

I.Q. testing has exerted its influence in America at least since World War I. But in recent decades, critics of I.Q. testing and of the concept of a “stable, distinct, and independent” trait called intelligence have proliferated in psychology and education. They charged that the field uses instruments that discriminate against (to name a few) minorities, the exceptionally bright, the creative, and the poor. None of those challenges has had the impact of Howard Gardner’s 1983 book Frames of Mind, which introduced the concept of “multiple intelligences,” relatively autonomous faculties including the linguistic, logical-mathematical, and musical. Although Gardner, a professor of developmental psychology at Harvard University, put forward his work as a psychological theory, not a platform for school reform, its impact on schools in America—indeed worldwide—has been great. The outpouring of books, articles, conferences, courses, and curricula invoking the concept of multiple intelligences might bring to mind Fincher’s melodious phrase, “caught the public ear as few seductive sounds from science have before or since.”

But Gardner, too, has his critics. James Traub, a writer on education, recently challenged Gardner’s ideas and their effect on education in an article entitled “Multiple Intelligences Disorder” in The New Republic (October 26, 1998). Traub is the author of City on A Hill: Testing the American Dream at City College (Addison-Wesley Publishing Co., 1994), named a New York Times Notable Book of the Year and winner of the 1994 Sidney Hillman Foundation Book Award. He has written on education topics for The New Yorker, Harper’s, and other magazines and is now a contributing writer for The New York Times Magazine.

Cerebrum invited Gardner and Traub to debate the concept of multiple intelligences in these pages. Here they debate three major questions: “What is the evidence for the theory of multiple intelligences?” “What has been the impact on education?” and “What is the potential of this idea for the future?”

Gardner and Traub addressed these questions in initial statements. They then exchanged statements, and each wrote a rejoinder. Finally, these rejoinders were exchanged, so each could write a short closing statement. We begin with Gardner.


What Is the Evidence for the Theory of Multiple Intelligences?

For nearly a century most psychologists have embraced one view of intelligence. Individuals are born with more or less intellectual potential (I.Q.); this potential is heavily influenced by heredity and difficult to alter; experts in measurement can determine your intelligence early in life, currently from paper-and-pencil measures, perhaps eventually from examining the brain in action or even scrutinizing your genome. Richard Herrnstein and Charles Murray defended this position in their controversial book, The Bell Curve (1994).

Recently, criticism of this conventional wisdom has mounted. Biologists ask if speaking of a single entity called “intelligence” is coherent and question the validity of measures used to estimate heritability of a trait in humans, who, unlike plants or animals, are not conceived and bred under controlled conditions.

A research psychologist by training, I began to doubt the standard view as a result of my work with both normal and talented children and with brain-damaged adults. I was struck that individuals can be strong (or weak) in certain skills, but that strength cannot predict skill in other areas. If a person is strong in telling stories, solving mathematical problems, navigating unfamiliar terrains, tracing the transformations of a fugal theme, or understanding the motivations of others, one does not know if comparable strengths (or weaknesses) will be found in other areas. That intuition lies behind the “theory of multiple intelligences” (MI theory).

While other scholars have also had that intuition, most have tried to support it by an examination of the correlations (or, rather, the lack thereof) of scores on paper-and-pencil-style tests of these skills. I approach the scientific challenge differently. I define intelligences as biological potentials to process information in certain kinds of ways to solve problems or fashion products valued in one or more cultures. To see if there is evidence for a possible intelligence, I examine eight separate criteria, drawn principally from biology, anthropology, and psychology; these criteria range from the existence of special populations that are strikingly strong or weak in an area (e.g., prodigies, savants) to the existence of symbol systems (e.g., maps, dance notation) that capture specific forms of information.

A comment on two biologically oriented criteria. One is “a distinct evolutionary history.” Existence of a specific intelligence becomes more plausible if one can locate its evolutionary antecedents, including capacities (like bird song or primate social organization) that humans share with other organisms. Specific capacities may operate in isolation in other species but have become yoked in human beings. For example, discrete aspects of musical intelligence may appear in several species but be joined in human beings. In The Prehistory of the Mind, Stephen Mithen describes how the several intelligences may have evolved one after the other.

Another critical criterion is “potential isolation by brain damage.” Scientists used to think that the brain had the same potential throughout, but now agree that different regions of the brain serve different functions. While avoiding a simple-minded phrenology, it is reasonable to speak of the middle frontal and temporal regions of the left hemisphere as language areas, and the posterior regions of the right hemisphere as spatial areas (in right-handed adults). When one can point to regions dedicated to certain processes, and when injury to these regions compromises those processes, one has persuasive evidence of a separate intelligence.

Based on these criteria, I originally identified seven intelligences. Each can be exemplified by a vocation that draws heavily on that intelligence. Thus the poet is strong in linguistic intelligence, the composer in musical intelligence, and the salesperson in interpersonal intelligence.

This way of conceptualizing intelligence has two corollaries. First, all human beings possess each of these intelligences; this, in fact, constitutes a definition of being human, cognitively speaking. (Rats may have more spatial intelligence, computers may have more logical-mathematical intelligence, but neither has intrapersonal intelligence). Second, no two individuals, not even identical twins, possess exactly the same mosaic of intelligences because we each have different experiences.

My reading of the scientific literature following the 1983 publication of Frames of Mind provides strong confirmation of the approach I took and the intelligences I identified. With this new evidence, I could identify an eighth (naturalist) intelligence, and a possible ninth (existential) intelligence. Suggestive findings about a possible relation between musical and spatial processing may require reformulation of those two intelligences. New in vivo methods of examining the human brain enable us to specify capacities more accurately. Finally, brain studies reveal two important correctives to my original approach: first, the greater idiosyncracies across individual persons; second, the involvement of many neural regions in nearly all non-elementary skills. Were I rewriting Frames of Mind, I would look for evidence that each intelligence involves specific (but not necessarily identical) regions of the brain in the majority of individuals examined.

What Has Been the Impact on Education?

I in no way anticipated the enormous interest in MI theory on the part of educators. The interest initially came from those involved with special populations (children with learning problems, “gifted” children) and those working with young children. More recently, interest has extended to middle schools, secondary schools, colleges, museums, and even the work place. This interest has been notably sustained. Whatever its scientific merits, clearly MI theory has struck a responsive chord in educators in the United States and abroad. It is important to indicate that this interest has developed with little push from me. I wrote initially as a psychologist, not an educator. I have neither endorsed any program nor had any commercial interest in any applications. The response to MI theory has arisen, without prodding, from the worlds of education and training.

Given this almost laissez-faire atmosphere, it is not surprising that many applications of the theory to education have been proposed. (In a new book, Intelligence Reframed, I list about 500 books and articles on MI.) Indeed, the theory has functioned as an ink-blot, with individuals reading into it their own hopes or anxieties. There is no way for me to track all of these applications. Recently, in her SUMIT study (Schools Using Multiple Intelligence Theory), my colleague Mindy Kornhaber has systematically studied 41 schools using multiple intelligences theory.1 Her survey reveals that a majority of schools indicate improvements in both scholastic (test-score) and other measures (parent involvement, student discipline). The theory has proved especially productive for students with learning difficulties.

Applications coming to my attention have been largely benign. MI theory has been used as a rationale for including a wider range of cognitive activities in the school day (and after-school); setting up “learning centers” or “flow rooms” where students pursue their own interests; encouraging students to carry out rich projects; and broadening the basis on which student understanding is assessed.

On occasion, however, the applications have been pernicious. Most upsetting was a curriculum in Australia that labeled children from different ethnic groups as having (or lacking) specific intelligences. Not only does this statement have no scientific basis; it offended my (and presumably most other people’s) sensibilities. I went on television to dissociate myself from the curriculum and was relieved when it was canceled.

Observing several dozen MI schools (mostly elementary schools in the United States) and conducting my own collaborative studies over 15 years, I have developed my own ideas about suitable MI applications. They are presented in the newly published Intelligence Reframed: Multiple Intelligences in the 21st Century and other publications, notably the three volume Project Zero Frameworks for Early Childhood Education (Teachers College Press, 1998).

I have concluded that MI theory is most useful under two conditions. First, one must define a specific adult end state that is desired. If one wants adults who are civil to one another, one must develop the personal intelligences. If one desires adults who are sensitive to the arts, then one must develop one or more of the intelligences that are crucial to artistic performance and perception.

Second, one must decide which kinds of understanding are most important. I personally seek students who can understand the world in terms of the major disciplines —science, mathematics, history, the arts. This understanding is most likely to emerge from deep immersion in specific topics and, if one is willing to spend time on those topics, an MI approach becomes possible. In The Disciplined Mind, I examined three rich and complex topics: the theory of evolution, the music of Mozart, and the Holocaust. I showed how understanding arises if one broaches these topics in a number of ways; draws on comparisons and analogies from complementary domains; and captures the key ideas in different symbol systems (e.g., linguistic, numerical, graphic, dynamic). Such applications show the power of an “MI approach” to serious scholarly work. Because students differ cognitively from one another, one reaches more students; and because disciplinary competence involves the capacity to represent the world in multiple ways, students gain a feeling for what it is like to be an expert.

What Is the Potential for the Future?

The idea of multiple intelligences is deceptively simple: “Human beings have eight or nine separate intelligences, not just the one or two that are highlighted in schools.” Yet, even 20 years after proposing the theory, I discover new implications.

Part of my own interest centers on the theory itself. In recent years, I have pursued theoretical distinctions; for example, the differences among an intelligence (biological construct), a domain or discipline (an epistemological construct), and a field (a sociological construct). I have considered additional intelligences (naturalist, spiritual, existential). I keep up with data from biological, cognitive, and anthropological sciences, monitoring with special interest studies using PET scans, functional MRIs, and other ways of examining the human brain in vivo. With the theory as background, I have begun to investigate extraordinary humans (see Creating Minds, Leading Minds, Extraordinary Minds, and most recently, my ongoing collaborative study of persons who carry out “good work”). I have also been stimulated—if not always delighted—by critiques of the theory.

The world of educational practice has a life of its own. Practitioners often do not follow the theoretical work carefully; many individuals involved in MI have read neither my work nor that of other reliable writers. This state of affairs seems inevitable.

Clearly, ideas of multiple intelligences will continue to be put to use in a range of spheres (schools, other educational institutions, businesses) in many regions and countries. Nearly all applications will initially be superficial. I am interested in whether or not the experiments last; whether or not insight deepens over time; and, most of all, whether or not the MI ideas help to bring about greater success in the educational missions of the institutions that embrace them. Thus, I welcome institutions reflective about their practice and actually employing measures (“hard” or “soft”) indicating whether MI practices have been productive.

It is difficult to prove that improvements (or declines) in educational achievement are due to MI practices. In the real world of education, one cannot carry out controlled experiments where everything save MI practices is held constant. I have been criticized for not making more decisive statements about the efficacy of MI approaches. I have not done so not because I doubt their efficacy, but because I know how difficult it is to prove that successes are due to MI. Surveys like SUMIT are pivotal.

I suspect that, as time passes, ideas about multiple intelligences will gain more acceptance among psychologists, though perhaps not my particular formulation. Psychologists favor research that relies on tests, while I question if this is the way to study the several intelligences.

Within formal education, I expect that a small number of schools will be developed based explicitly on MI beliefs and practices. More broadly, the work of my colleagues will encourage practitioners to enhance their teaching approaches, curricula, and modes of assessment. Whether or not MI theory is explicitly acknowledged here is not important.

For me, the scientific and practical importance of MI theory centers on the idea that individuals differ cognitively from one another on a range of relatively autonomous dimensions. This variety is important in terms of evolution; it also spices up life for all of us on this planet. For most of recorded history, educators have ignored this variety in favor of uniform schools, where all students have been taught and assessed in the same way.

I may have helped to undermine this institutional reflex, but computers will do far more. Already it is possible to individualize education extensively by using software that engages a range of intelligences. In the future, that range will surely be extended; moreover, “intelligent systems” will adapt teaching to the strengths and interests of each student. In 50 years, our successors will laugh at the notion that there is but a single way to teach and assess. Instead, they will seek the best way to teach this concept or subject to this student and the best way for this student to demonstrate understanding. If MI theory turns out to have kept apace scientifically, it will be able to explain in biological, psychological, and cultural terms why this educational approach has worked.


What Is the Evidence for the Theory of Multiple Intelligences?

“Intelligence” is one of the most tendentious words in the English language. Human beings possess innumerable gifts and attributes, and yet we apply the word “intelligence” only to one tiny cluster of them. Howard Gardner has pointed out that the definition of this core of intellectual attributes changes from one era to the next, and is different in traditional cultures from what it is in technocratic ones. It seems to me that the simplest way to dispose of the unending debate over the validity of I.Q. tests and the like would be to concede that “intelligence” is not an entity but a construct, and then to proceed directly to an open discussion about which intellectual attributes we consider essential at this particular moment in history.

That, however, is not Gardner’s position. Instead, he has argued, since the publication of Frames of Mind in 1983, that the pantheon of intelligence has been far too restrictive. To the traits measured by I.Q. tests, which he describes as “linguistic” and “logical-mathematical” intelligence, Gardner would add musical, spatial, interpersonal, intrapersonal “bodily-kinesthetic,” and what he calls “the naturalists’ intelligence.” Gardner says that he has considered, but not yet granted admission to, the “existential intelligence.”

Unlike the proponents of “moral intelligence” or “emotional intelligence,” Gardner has not used this terribly loaded term, intelligence, simply for effect; he has furnished a set of objective criteria that may be used to distinguish it from a mere aptitude. In Frames of Mind, Gardner writes that he calls a trait “intelligence” only if “it can be found in relative isolation in special populations,” “may become highly developed in specific individuals or in specific cultures,” and if “experts in particular disciplines can posit core abilities that, in effect, define the intelligence.” Elsewhere, he distinguishes between information-processing skills and values, or attributes of character.

Gardner’s signal achievement is to bring the findings of brain research into a world hitherto defined by test outcomes. He premises his “multiple intelligence” theory on the “modular” or “vertical” picture of the brain that is now widely accepted in neuroscience. We now know that mental activities are parceled out into separate regions of the brain. This, Gardner says, argues for a view of intelligence as consisting of relatively autonomous faculties, rather than for the traditional view of a bundle of aptitudes highly correlated with one another and together constituting a larger whole called “general intelligence,” or G.

Most psychologists who study intelligence scoff at Gardner’s work, pointing out that he has merely posited his various intelligences; neither he nor anyone else has yet done the psychometric work that would be needed to prove that they are distinct, rather than aspects of one another or of something else. Gardner’s argument for the relative autonomy of intellectual gifts is broadly accepted, up to a point. Brain researchers recognize, for example, that there are many types of memory and even learning capacity that are scattered throughout the brain. Research in brain-damaged patients has shown that it is possible to lose one core intellectual ability and leave others unaffected. But to abandon G is to accept a view of the brain as having little or no executive capacity to direct and integrate the mind’s activity. Gardner has suggested, alternatively, that each intelligence might, in effect, have its own G; but at this many cognitive psychologists balk. Can we really say that “musical intelligence” is so autonomous from, and uncorrelated with, the skills that constitute G?

Gardner’s “multiple intelligence” theory, perhaps like all theories about intelligence, begins in science and ends in cultural politics. G is a construct used to explain why the various aptitudes measured by I.Q. tests correlate with one another so highly. It would hardly be a surprise if aptitudes not measured by I.Q. tests—say, the bodily-kinesthetic ones—are completely independent of I.Q. scores. We measure some qualities, and not others—and call the ones we measure “intelligence”—because we think they matter more in determining success in a technocratic culture, as they have since Binet’s time. They do not, of course, determine human value, though we, and the psychometricians, often talk as if they do.

Gardner, who once trained to be a concert pianist, recoils at this narrow conception of man-the-logician. In one of his first books, The Arts and Human Development, he rebuked the developmental psychologist Jean Piaget for studying only “those mental processes that culminate in scientific thought.” Creative thought, he pointed out, is just as fundamental a constituent of human behavior as logic. Gardner is, at bottom, a moral philosopher; he wants to change the way we measure human worth. That is a profound project, if perhaps only incidentally a scientific one.

What Has Been the Impact on Education?

Frames of Mind landed in the school world with the force of revelation. Though Gardner had written scarcely a word about pedagogy, “MI schools” began to spring up within a few years of the book’s publication, and were soon followed by MI teaching guides, consultants, and publishers. Gardner had offered a theoretical framework, and a scientific explanation, for an intuition shared by many teachers: that children have different ways of learning, and must be approached with different kinds of teaching. Gardner had also validated progressive education’s focus on the learner, rather than on the knowledge being imparted, a principle which had already attained the status of orthodoxy in education schools and among many teachers.

The common thread of MI schools is the use of what Gardner calls “multiple entryways” to curricular knowledge. Thus, at one elementary school described in an educational journal, students learning about photosynthesis “might act out the process at one [learning] station, read about it at another station, and, at others, sing about photosynthesis, chart its processes and, finally, reflect on events that have transformed their lives, just as chloroplasts transform the life cycle of plants.” Two things are at least allegedly happening here: The school is stimulating all of the child’s intelligences, and the child is mobilizing those various faculties to gain a deeper understanding of subject matter.

Gardner has made a point of writing that everything ought not be taught seven or eight different ways; and many schools find ways of integrating the MI philosophy without being so literal-minded about it. Nevertheless, the question this and similar descriptions raises is: Have many students been failing to learn about photosynthesis, or fractions, or American history, because they happen not to have the mix of intelligences suited to the traditional text-oriented curriculum? If this were true, one would have to imagine that the Asian and European students who regularly out-perform Americans were benefiting from a more personalized form of pedagogy—which, of course, they are not. The educational system both in Taiwan and France, for example, is determinedly knowledge-centered, and is unambiguous about the primacy of logic and language.

But perhaps this is the wrong question. Progressive reformers argue that our schools aim too low, that the knowledge imparted even by apparently successful schools is superficial; they draw a sharp distinction between “knowledge” and “understanding.” In The Unschooled Mind, Gardner observes that even students at the highly selective Massachusetts Institute of Technology provide childishly naive explanations of ordinary real-world phenomena. What they have learned with paper and pencil— and their logical intelligence—has left them with a merely formulaic grasp of the subject matter. The mark of true understanding, Gardner writes, is the ability to make “multiple representations” of a given subject. He would have us teach subjects in their own medium as well as abstractly.

This provocative thought about true understanding echoes the work of educational philosophers like Seymour Papert. Nevertheless, American schools are in crisis not because even the most expert students suffer from a flimsy grasp of subject matter, but because so many students have almost no grasp of subject matter, nor the skills they need to remedy the situation. Indeed, the progressive distinction between knowledge and understanding seems peculiar, and almost perverse. A persuasive case can be made that American students are doing poorly as a result of the progressive aversion to a specified curriculum, not because of a Gradgrindish preoccupation with “drill and practice.” Gardner’s harshest critics, in fact, are not the psychometricians, but educational traditionalists like E. D. Hirsch, and educational psychologists like Harold Stevenson, who believe that he has things backwards.

The proliferation of the MI model may well have the positive effect of giving music and art—and even physical education—a more central place in the curriculum. It may, on the other hand, have the harmful effect of reducing the sense of urgency needed to ensure that all children master basic skills. MI can offer a pretext to put self-esteem ahead of the hard work of learning: Rather than harp on what a child doesn’t do well—i.e., reading or math— why not focus on his or her own special gifts? Isn’t singing about photosynthesis, or putting on a play about it, a valid form of learning? Many schools will not need much encouragement in this regard. The guiding principle of one well-known MI school is, “Who you are is more important than what you know.” That seems like an excellent formula to ensure that you do not know much.

What Is the Potential for the Future?

Gardner believes strongly that we are living at the tail end of an exhausted paradigm of human nature. Invoking the work of Binet, he asks why we should continue to live with a definition of intelligence based on a “scholastic skill—what it meant to be a good bureaucrat a hundred years ago.” Why do we continue to distribute so precious a good as a seat in our greatest universities on the basis of minute distinctions in performance on the SAT exams, our ubiquitous surrogate for the I.Q. test? Gardner believes that the walls of the I.Q. meritocracy are crumbling (though the psychometric establishment has not yet noticed). “There is a new definition of human beings, cognitively speaking,” he has grandly said. “Socrates defined man as a rational animal; Freud defined him as an irrational animal. What MI theory says is that we are the animal that exhibits the eight and a half intelligences.” (“Existential” is the half.)

This raises two basic questions: Is the old paradigm really on its last legs, and what would it mean to have a new one based on Gardner’s conception?

Of course, the reason why we still live by Binet’s definition of intelligence is that yesterday’s good bureaucrat is today’s good mid-level executive, or corporate lawyer. This is what former Secretary of Labor Robert Reich means when he says that the skill increasingly in demand in the world is that of “symbolic analyst.” An increasing portion of the population works with abstractions for a living; that is why the minimal definition of school success has advanced so far beyond the achievement of literacy. It is quite true that the universalization of the computer will automate much of the brute work of logic; but the proliferation of spreadsheets will probably have the effect of making spreadsheet-type thinking more conventional, not less so.

At the same time, we no longer view rationality as the sine qua non of modernity. That era seems to have lasted about from the time of Francis Galton (1822-1911) to the time of Vietnam-era Secretary of Defense Robert McNamara. We have been forcibly disabused of the wisdom of such wise men, and of their machines. So Gardner is quite right to say that the sense of human personality has changed. The immense popularity of Daniel Goleman’s theory of “emotional intelligence” shows how widespread is the wish for a less hyperrational view of character. The same may be said for the incredible reception which Gardner’s work has enjoyed far from the worlds of science. The restless and often rootless search for meaningful forms of faith reflects the inadequacy of the old secular gods. We grow more estranged from rationality as ever more rationality is demanded of us.

Will we turn instead to a more diverse, less hierarchical view of human gifts? The answer, to a certain extent, must be “no.” So long as the commanding heights of the culture require gifted symbolic analysts, the institutions which serve as sorting devices for the elite—say, Harvard University—will keep selecting candidates based mostly on Binet’s hundred-year-old criteria. In certain quarters, Gardner’s paradigm will continue to be seen as a pretext for refusing to accept distinctions of merit. And yet it is not implausible that our sense of merit will change—that the grade-point average of that girl applying to Harvard will be measuring her reflectiveness, her social gifts, her classificatory skills, her aesthetic sense, as much as by her linguistic and mathematical talents. We may, in effect, come to think of well-roundedness as itself the supreme expression of merit.

One is tempted to growl, as Robert Frost once allegedly did, “Why d’ya wanna be well-rounded—you wanna roll downhill?” But perhaps one ought not be so curmudgeonly in defense of the hegemony of logic and language. We are, after all, a Philistine culture; we would be less so in Gardner’s ideal world. We would also, presumably, be more self-knowing, more socially adroit, more aware of our bodies. But we would also, necessarily, be less something—a little less intellectually nimble, perhaps. Gardner likes to poke fun at what he calls the Alan Dershowitz model of intelligence. In Gardner’s future, we will have more well-rounded folks, and fewer like Harvard law professor Alan Dershowitz. One need not excessively respect Claus von Bulow’s lawyer to say that this does not sound like the answer to America’s prayers.


In his provocative statement, James Traub wears a number of hats (cultural critic, science reporter, capsule biographer) and assumes a number of tones (respectful, neutral, ironic). Rather than responding to each, I have chosen to clarify some points for both Traub and the readers.

Science and Politics

Traub suggests that all theories of intelligence begin in science and end in cultural politics. It may be true that, in our culture, theories of intelligence are drawn upon to undergird social recommendations (e.g., The Bell Curve). But one must draw a sharp distinction between those who encourage blurring of boundaries and those who strive to keep the realms discrete. For example, Arthur Jensen and Hans Eysenck are both psychologists who believe in the explanatory power of G. Throughout his career, Eysenck moved almost too effortlessly between psychometrics and policy; but after one disastrous foray, Jensen has focused on G, avoiding policy debates altogether.

As an educational reformer and a citizen, I have described the schools I favor and the society I cherish. But I have sought to keep my scholarly analysis of intelligence separate from my policy recommendations. Indeed, I have insisted that policy never follows directly from scientific discourse. And I have always emphasized that the various intelligences are amoral in themselves: one can use interpersonal intelligence to resolve a dispute or to manipulate groups.

Intelligences and Domains

Many writers, including me, have sometimes confused “intelligences” with “domains” or “disciplines.” An intelligence is a biopsychological potential that can be drawn on for a variety of skills or roles. A domain or discipline is a culturally recognized area of performance, whose practitioners can be arrayed in terms of expertise. Thus spatial intelligence can be exploited in domains like chess playing, sailing, surgery; and the domain of law draws, variously, on linguistic, logical-mathematical, and interpersonal intelligences. Even “musician” and “musical intelligence” are not interchangeable; musicians exploit several intelligences in composing or performing.

This conceptual distinction is important. Ultimately, as educators or citizens, we should not care which intelligences individuals are using; the important goal is to achieve reasonable performances in domains that matter. An individual with modest logical-mathematical intelligence can still learn to carry out mathematical operations; he will have to rely more than others do on linguistic, spatial, and perhaps bodily intelligences. The challenge for educators, and for the individual, is to marshal and choreograph intelligences to obtain proficiency. Contrary to Traub’s implications, there is no reason whatever to avoid teaching literacy or the disciplines to youngsters deficient in a given intelligence. MI theory simply provides a convenient way of analyzing how best to approach a domain when customary teaching practices fail.

Indeed, unless we literally peer inside the mind/brain, we have no way of knowing which intelligences an individual is actually using. You are reading these words (linguistic) but you may be representing them mentally using various intelligences (ranging from an interpersonal debate between the journalist and the academic, to a numerical scorecard, to a spatial layout of arguments). An MI classroom in the United States may publicly evoke different intelligences in teaching about photosynthesis; but for all we know, students in France or Taiwan may also be encoding such biological systems in a spatial or naturalist way. Nor should we assume that other societies ignore individual differences. Early education in Japan focuses on social and personal development, and many Japanese families use individual or group tutoring to supplement schooling.

Symbol Analysts and Anti-Rationality

In various writings, Traub has suggested that I am opposed to logic and rationality. This is simply untrue. I try to operate according to those canons and hope others will as well. While I poke gentle fun at my colleague Alan Dershowitz, I have never denigrated his powerful mind; I have only pointed out that many schools and colleges are currently set up to select one “frame of mind” above others. Fortunately, recognizing that standardized tests predict but a modest proportion of future success, some selective institutions welcome other samples of student work.

Throughout much of this century, it is true, individuals who are expert in “symbol analysis” have been at a premium in our society. It is not unreasonable to assume that they are using linguistic and/or logical intelligence, but, again, we do not know this for sure. Charles Schwab, for example, takes a back seat to few in his financial acuity; yet he speaks freely about his great difficulties in learning to read. There are many ways to conceive of, and analyze, a financial market.

Two other points. First of all, it takes various gifts and mixes of talents to have a productive society. As Traub points out, the canonical “best and the brightest” have sometimes inspired disastrous policies. Second, the kinds of abilities at a premium in a society can change, sometimes quite swiftly. Inventions like the printing press, the computer, or the cinema bring certain intelligences to the fore, while (at least temporarily) de-emphasizing others. The smarter that machines become, in the Binet sense, the more emphasis society is likely to place on those intelligences that transcend standard computation. And so, just as evolution is friendly to diversity, we are well advised not to put too much stock in the nurturance of one or two intelligences.

Knowledge (Facts), and Understanding

Traub imposes on me a dichotomy that I do not make and will not accept. Of course, it is not possible to understand a topic unless you have considerable knowledge, including factual. Who could maintain otherwise? The distinction that I make, most recently in The Disciplined Mind, is twofold. First and foremost, I believe that facts ought to be picked up through intensive, deep study of consequential issues. Not only will one have the motivation to learn the facts; one will also be in a position to put them together in meaningful ways, to recall them, and to master vital disciplinary ways of thinking. Absent such organization and mission, the facts remain as “inert knowledge”—available, perhaps, for the next test but likely to be forgotten and, in any case, not leading to disciplined thought. Second, I have little sympathy for the delineation of a specific canon—specific facts, theories, or books that all must master. The most important disciplinary habits of mind can be obtained from a variety of topics; and once they have been obtained, one is free to master any canon or create one’s own.

False Dichotomies

While I have broad sympathy with the progressive tradition in education (of philosopher John Dewey), I would be happy to chuck the distinction between Progressive and Traditionalist. It’s too often used to vilify, rather than to illuminate. As for the contrast between “who you are” and “what you know,” after the Littleton, Colorado, school massacre and similar tragedies, I hope that we do not consider this an either/or choice.

Creativity should not be contrasted to science or logic. There are creative scientists and logicians, and (all too many) noncreative artists. –H.G.

And Finally…

  1. Discussion of an “executive function” is a separate issue from the existence or provenance of G.
  2. Creativity should not be contrasted to science or logic. There are creative scientists and logicians, and (all too many) noncreative artists.
  3. Never have I ever stated or implied that basic skills (the three Rs) should not come first. I continuously insist that they must.
  4. As a youngster, I was a serious pianist but never trained for the concert stage. I still, however, have dreams.


Is it really true that “If a person is strong in telling stories [or] solving mathematical proofs…one simply does not know whether comparable strengths (or weaknesses) will be found in other areas”? Does this assertion’s contradiction of our intuitive sense of the world only show how captive we are to the psychometric tradition? In Frames of Mind, Gardner pulls together evidence from an extraordinarily wide variety of fields to show that these faculties have distinctive evolutionary histories, patterns of pathology, and brain localization. But one is left wondering exactly how autonomous is “relatively autonomous?”

We know, of course, of people who are especially gifted in “understanding the motivations of others” or “tracing the transformations of a fugal theme.” But do we know of people who have such talents, and yet lack the “general intelligence” measured by I.Q. tests? I do not. Relative autonomy does not, after all, preclude correlation. As Steven Ceci, a developmental psychologist at Cornell University says, “If you tested people in track and field and you found someone who was really outstanding in one particular event, like the hurdle or the high jump, you’d also find that they were above average in all the others.”

Learning for Living

Let us, however, stipulate for argument’s sake that Gardner is right at the level of neuroscience and psychometrics. What happens when we apply his thinking to the world around us? What does it mean to think of logic and language as merely two of eight distinct and epistemologically equal gifts? Gardner has observed that in some traditional cultures—in western Africa, for example—musical intelligence is counted the summum bonum of human achievement. In the future, he observes, with computers doing the work of logic for us, we ourselves may put more store by social graces—that is, by “interpersonal” intelligence. Gardner has demonstrated that cultures adapt the concept of intelligence to their own purposes. But what about our present purposes? Since we do not get around in dugout canoes, learning to navigate by the stars is not terribly relevant to most of us. It may be a real gift, and it may even rest on a distinct intelligence, but, so far as our acculturating institutions go, it is pretty marginal.

Americans have a terrible problem with their principal acculturating institution—the public schools. According to last year’s National Assessment of Educational Progress, 38 percent of fourth graders read at a “below basic” level—the same number as score “at or above proficient.” We perform middlingly in the lower grades, and at or near the bottom in the upper grades, on all of the indices of the Third International Mathematics and Science Study.

Harold Stevenson, a psychologist at the University of Michigan, compared the performance of students at demographically similar schools in the United States, Japan, and Taiwan, and found that, among fifth-graders, only 4 percent of Chinese students and 10 percent of Japanese did as poorly on math as the average American; the figures were almost as bad in eleventh grade. Stevenson explained the huge discrepancy by noting that the American students spent far less time on academic work than did the others; that American teachers, lacking a national curriculum or standards, rarely worked together; that American parents were far more easily satisfied by their children’s work than were Asian parents; and that while Asians stress hard work, Americans believe that success is due to innate ability. Gardner’s multiple intelligence theory, of course, offers a new and progressive-sounding version of the American preoccupation with innate ability.

A Stand on Standards

No one can question Gardner’s own intellectual seriousness, but his pedagogy seems designed to address some problem other than the one which our schools suffer from. He has criticized standardized tests generally, and recently wrote skeptically of New York State’s intellectually ambitious fourth-grade reading and writing test, asserting that we must not lose sight of “the most crucial skills: love of learning, respect for peers, and good citizenship.” Like many progressives, he opposes a national curriculum or national standards. His criticism of the hegemony of logic and language seems oddly malapropos in today’s schools. Stevenson observes:

If you’re giving this child excessive feelings of accomplishment because the child is good at music, and not giving the child the sense of need to become accomplished in abstract intelligence, then you’re depriving that intelligence for important parts of his or her future development.

In short, we cannot afford to be agnostic about the “specific adult end state” that we want—at least, not until we have achieved the adult end state of writing a coherent paragraph.

Stevenson and other traditionally-minded reformers, such as Diane Ravitch, argue that Gardner has given aid and comfort to educators who want to excuse mediocre performance and justify low standards. But Gardner’s actual influence on the schools is certainly more complicated than that. Many ambitious and thoughtful teachers and administrators have been inspired by his books, and have seen them as a justification for holding children to the highest, not the lowest, standards.

The two “MI” schools I have visited had their eccentricities, but they expected a high level of academic performance from children. At the Governor Bent elementary school, in Albuquerque, fifth-graders who had been defined as gifted, and yet were deficient in reading and writing—itself a kind of multiple intelligence diagnosis—had been assigned sophisticated architecture projects, including the construction of an ant colony that could withstand flood, hurricane, and enemy attack, and of a robot one of whose parts could rotate 360 degrees. “If I was offering the kids another kind of curriculum,” said their teacher, Marleyne Chula, “you would not see them engaged like this.” But children at Governor Bent are also expected to bark out the names of painters, composers, and Presidents on command—a most un-Gardnerian faith in “decontextualized facts.” Perhaps the school succeeds out of sheer inconsistency.

In The Disciplined Mind, Gardner made the startling admission that he could imagine approving of a national curriculum—save for his fear that a “Jesse Helms” would take control of it. It might be a salutary exercise for Gardner himself to write such a curriculum—not an airy “pathway of understanding,” as he limns in The Disciplined Mind, but a year-by-year, subject-by-subject syllabus. Some unexpected people might be happy to embrace it.


Like the imminent hanging described by Dr. Johnson, a 500-word limit concentrates the mind. I am now convinced that Traub is not principally concerned with multiple intelligences, nor with an ideal education. Unrelentingly pragmatic, he believes American schools are not good and need to improve. MI is seen as at best a benign grace note, at worst destructive.

I am no apologist for American education. I share Traub’s impatience with its often mediocre accomplishments (though I may have more empathy for the difficulties faced by many teachers). But I am convinced there is no one right way to achieve better schools. Widely divergent schools succeed in different countries and in ours. Our much-admired colleges are distinguished by their diversity. In The Disciplined Mind, I concluded that so varied a country as ours cannot agree on a single gritty and challenging curriculum. We will either fight endlessly (is creationism a theory?) or settle for pabulum. Accordingly, I sketched six alternative K-12 pathways. I even conceded that I would rather send my children along a pathway whose philosophy is not appealing, but which pursues that philosophy consistently than a pathway that applies “my” philosophy poorly.

A Personal Vision

The Disciplined Mind outlines my personal vision: a pathway that begins by teaching the three Rs; project work that addresses essential questions and engenders motivation to pursue those questions; work that begins to introduce the major disciplinary divisions; and then a deep exploration of several major disciplines: science, history, mathematics, the arts, and considerations of ethics. Students will acquire facts, but not acontextually; rather, they will learn facts naturally, as they investigate consequential matters, and are stimulated to read, write, and explore widely. If they lack a fact, they can look it up or use tomorrow’s Palm Pilot that issues facts on oral request.

For me, the irreplaceable essence of an education is a mastery of the major ways of thinking that have developed over the centuries in several disciplines. These provide the “mental furniture” which alone can make sense of past materials and allow one to understand new materials, phenomena, events. Absent the disciplinary capacity to apply or assimilate information, one has only “inert knowledge”—Christmas tree ornaments without the fundamental tree and branches.

In The Disciplined Mind, I detailed an educational investigation of three major topics—evolution, the music of Mozart, and the Holocaust. In the process, one would learn what it is like to understand consequential materials and would also be introduced, respectively, to the disciplines of biological science, music, and history. I elaborated how the fact of our multiple intelligences can be used to make this material accessible for students and ultimately yield a rich set of representations of a topic. My discussion provides a curricular model for these topics. I hope that it will inspire others to create and “own” their own curriculum. But I am not a curriculum writer, any more than James Traub is a research social scientist.

Notes as the Noose Tightens:

Traub seems fixated on how to improve our standing in international comparisons. I am concerned with the formation of disciplined minds, which, I believe, are the best preparation for the future. Current tests (and ways of testing) are inadequate.
Brain study will help us understand the mind, but you can never go directly from brain to education. Education involves considerations of epistemology (factual accretion vs. disciplinary analysis) and values (wide cultural literacy vs. in-depth understanding of key topics). These decisions must be made by communities, not by scientists or pundits.


I have to wonder why I would bother to pick a fight with someone as thoughtful, and as genuinely creative, as Howard Gardner. (Here I am in my “respectful” mode.) I recognize that this probably has less to do with his own work than with the effect I fear it will have on the world. It is absolutely true that Gardner has resisted the idea that intelligence can be something other than a form of cognitive processing—that it can be inherently moral, for example. It is also true that he has stressed his own faith in intellectual rigor. But many of his readers have no such faith, and are not so inclined to distinguish between the cognitive and the non-cognitive. And so I cannot feel much confidence about how they will use the concept of “bodily-kinesthetic intelligence” or “the personal intelligences.” I fear they will be used at least as excuses, and not just as opportunities. And though I do not wish to sound hard-hearted, I do not agree the lesson of Littleton was that schools are failing to teach that who you are is more important than, or as important as, what you know. I do, however, believe one very important lesson of Littleton was that if you want to have a culture of respect and mutual understanding, you should not have 1,800 kids in a high school.

What Should We Measure?

I can not help bridling at the phrase, “An individual with modest logical- mathematical intelligence…” Of course, I do not dispute the idea that intelligence is significantly heritable. But is there not something retrograde about going from the idea of I.Q. to the idea of seven or eight I.Q.s? Would we not be better off focusing on effort, rather than innate ability? I agree with Gardner that the SATs have vastly too much power in our society, but that is because they measure, or pretend to measure, underlying capacities rather than disciplinary knowledge. We should encourage and reward effort by measuring what students have actually learned. This, by the way, is another argument for a national curriculum, or at least for national “standards.”

I honestly do not see why the particulars of a specified curriculum would inevitably be reduced to “inert knowledge.” That has to do with how one teaches, not what one teaches. At a school I visited in Cambridge that follows E. D. Hirsch’s highly detailed Core Knowledge curriculum, kindergartners were learning about Moussorgsky while dancing around to “Pictures At An Exhibition” (very MI), third-graders were reading books to one another, and sixth-graders were attending a local play in order to bring the practice of commedia dell’arte to life. Is it really unreasonable to think that the lion of factualness can lie down with the lamb of understanding?


  1. For details, see http://www.pz.harvard.edu/projects/project-sumit