American Scientist

Descartes' Baby: How the Science of Child Development Explains What Makes Us Human. Paul Bloom. xvi + 271 pp. Basic Books, 2004. $26.

When I teach about the mind/body issue, I am often struck by how many of my students are dualists. I'm not talking about modern hedged positions such as property dualism or explanatory dualism; I'm talking about good old–fashioned Cartesian substance dualism, which maintains that our physical bodies/brains are inhabited by immaterial souls/minds and that body and soul are intimately linked, yet distinct and dissociable (at death, for example, when the soul may depart the body). And these students are not wild–eyed religious evangelists; they are sober–minded science majors. I pose what seem to me to be serious problems with this position: For example, how could material and immaterial substances interact? But many of these students seem unable even to see the problem. I end up perplexed by their lack of perplexity.

Paul Bloom has an explanation. In his new book, Descartes' Baby, he maintains that dualism is innate—that is, not learned. We naturally see the world as containing both material objects, which are governed by physical laws, and mental entities, whose behavior is intentional and goal–directed. Some things in the world, such as people, can be seen either way, as physical bodies or as intentional agents. However, as Bloom describes, we tend toward the latter interpretation whenever possible, even attributing intentions to animated shapes on a computer screen if they move in certain ways. According to Bloom, dualism is the product not of nurture but of nature—specifically, evolution by natural selection. It was adaptive for our ancestors to be able to predict the behavior of physical objects and social creatures (especially conspecifics).

This much of Bloom's argument is not new. What is new—and makes for fascinating reading—is his argument that many phenomena that we think of as uniquely and essentially human, such as art, morality, humor and religion, can be explained as by–products of our innate dualism.

To say that dualism is innate is not to say that it is correct. In fact, Bloom makes it clear that he is a materialist, not a dualist. However, he simply states his position without attempting to prove that materialism is correct—which is probably a good thing, considering that philosophers have gone around in circles for centuries on the mind/body issue. Bloom escapes this trap by changing the question slightly, from whether belief in dualism is justified to why belief in dualism is so prevalent (the question I have about my students).

Bloom marshals evidence from cognitive developmental psychology to show that infants have expectations about the behavior of inanimate objects that differ from their expectations about the behavior of animate beings. Furthermore, these expectations appear too early in life, he argues, to have been learned. The claim that some knowledge is "built in," however, does not preclude its revision with age and experience. We are born dualists, but we can learn to be materialists. Just as training in modern physics may radically alter our understanding of the physical world, training in modern cognitive science may lead us to believe that the mind is simply an emergent property or functional state of the brain. Such beliefs may be correct, Bloom argues, but they are fundamentally unnatural. Apparently I'm the one who has the weird ideas, not my dualistic students.

Paul Bloom is an excellent ambassador for cognitive developmental psychology. He shows how seemingly simple, even trivial, studies (such as ones that measure how long infants look at one display versus another) can actually reveal something important about human nature. He is certainly not the first to claim that children's cognition is domain–specific. Developmental psychologists such as Henry Wellman, Susan Gelman, Susan Carey and Alison Gopnik have described how children use different naive theories to reason about different types of phenomena (physical vs. psychological vs. biological). But Bloom is the first to explore the implications of this insight for a broad range of human characteristics and activities in a way that is accessible to readers outside the field.

The bulk of Descartes' Baby is an exploration of those implications for areas as diverse as autism, racism, modern art, psychopathy, altruism, genocide, disgust, slapstick humor, cloning and creationism. Some of these sections feel like detours from the main road of his argument, but they are detours through interesting terrain with an entertaining guide. Bloom is one of those frighteningly erudite writers, with an encyclopedic grasp of the research literature and the ability to throw in the perfect Jorge Luis Borges quote to illustrate his point. In fact, in his chapter on art, some of the references may go over the head of the less well–versed, especially because there are no accompanying illustrations.

Among current popularizers of cognitive psychology, Bloom has no peer except Steven Pinker. Pinker calls Bloom "the wunderkind of cognitive science," and their biographies are eerily similar. Both were born in Montreal (Bloom nine years after Pinker), both attended McGill University, both did time at MIT (Bloom as a doctoral student, Pinker as a professor), and both have settled at the Iviest of the Ivy League schools (Bloom at Yale, Pinker at Harvard). Both write well, but I found Descartes' Baby to be an easier read than most of Pinker's books, and it's certainly shorter (under 300 pages vs. more than 600 for Pinker's How the Mind Works). But Pinker has much cooler hair.

Descartes' Baby covers a number of different points relatively quickly—a welcome difference from many academic books, which tend to dwell on the same point ad infinitum. Inevitably, in a book of such scope there are a few false notes. For example, Bloom sets up a dichotomy between naive physics and naive psychology but is unclear as to whether children's naive biological concepts fall on one side or the other. He contrasts biological identity with physical appearance in his discussion of children's essentialism, but then he contrasts biological functioning with mental functioning in his discussion of children's beliefs about death. Many authors consider naive biology to be a separate core cognitive domain, but Bloom does not mention this possibility.

A few examples fall flat. To illustrate the point that genetic variability is greater within racial groups than between them, Bloom makes an analogy between races and families. So far, so good. But then he asks, rhetorically, "Is my child more genetically similar to every other Bloom than to every other non–Bloom, including his mother?" He seems to be saying that the genetic similarity between child and mother is evidence of similarity between different groups. In other words, his child's mother is not a member of the same family because she has a different last name (Bloom's wife is the developmental psychologist Karen Wynn). I'm not sure he wants to say this, especially when he goes home.

Nevertheless, a few false notes do not spoil the symphony. Descartes' Baby incorporates the most recent and provocative theoretical ideas in cognitive science, such as Simon Baron–Cohen's hypothesis that people with autism suffer from "extreme maleness." And Bloom's theoretical points are leavened with amusing and illuminating anecdotes, such as the story of the autistic boy who tried to climb him like a ladder to reach a toy on a high shelf.

Bloom, demonstrating the typical human tendency to interpret the behavior of others psychologically, initially mistook the contact as an expression of affection. The boy, however, was demonstrating the tendency of autistic individuals to view other people as physical objects rather than as intentional agents. As Bloom notes, "It would have been simpler if he had just asked me to get him the toy."—Ethan Remmel, Developmental Psychology, Western Washington University, Bellingham