Edge 139—May 13, 2004
(5,100 words)

In the domain of bodies, most of us accept that common sense is wrong. We concede that apparently solid objects are actually mostly empty space, consisting of tiny particles and fields of energy. Perhaps the same sort of reconciliation will happen in the domain of souls, and it will come to be broadly recognized that our dualist belief system, though intuitively appealing, is factually mistaken. Perhaps we will all come to agree with Richard Dawkins and Daniel Dennett and join the side of the "brights" [*]: those who reject the supernatural and endorse the world-view established by science.

But I am skeptical. The notion that our souls are flesh is profoundly troubling to many, as it clashes with religion. Dualism and religion are not the same: You can be dualist without holding any other religious beliefs, and you can hold religious beliefs without being dualist. But they almost always go together. And some very popular religious views rest on a dualist foundation, such as the belief that people survive the destruction of their bodies. If you give up on dualism, this is what you lose.

This is not small potatoes.

A Talk with Paul Bloom


Pamela McCorduck on "Aristotle" by W. Daniel Hillis


The New Humanists:
Science at the Edge


Jared Diamond • Steven Pinker • Helena Cronin • Andy Clark • Marc D. Hauser • Richard Wrangham • Daniel C. Dennett • Stephen M. Kosslyn • Jordan B. Pollack • David Gelernter • Rodney Brooks • Hans Moravec • David Deutsch • Marvin Minsky • Ray Kurzweil • Jaron Lanier • Seth Lloyd • Alan Guth • Paul Steinhardt • Lisa Randall • Lee Smolin • Martin Rees • edited, with an introduction by John Brockman
Science at the Edge

A Talk with Paul Bloom


As a teenager, Paul Bloom worked extensively with autistic children, and when he majored in psychology at McGill University, he expected to end up as a clinical child psychologist. His interests shifted when he met John Macnamara, a professor who studied the interface between psychology and philosophy. Bloom worked with Macnamara as an undergraduate, and then did his graduate work at MIT with Susan Carey, on cognitive development and language acquisition.

As a professor—first at University of Arizona, and then at Yale—Bloom explores how children learn the meanings of words, and he developed a theory of word learning that has social cognition (also known as "theory of mind" or "mindreading") at its core. More recently, Bloom and his students have started to explore a set of related puzzles having to do with the nature and development of art, religion, humor, and morality.

PAUL BLOOM is a professor of psychology at Yale University who works on language and development, and with Steven Pinker coauthored one of the seminal papers in the field. He is co-editor of Behavioral and Brain Sciences and the author of several books, the most recent of which is Descartes' Baby: How the Science of Child Development Explains What Makes Us Human.


Paul Bloom's Edge Bio Page


Content on this page requires a newer version of Adobe Flash Player.

Get Adobe Flash player

(PAUL BLOOM:) For the last few years I have been interested in common sense dualism, which is the notion that people have two ways of looking at the world. We see the world in terms of material bodies, including our own bodies, and in terms of immaterial souls. And we are dualists; we see bodies and souls as distinct.

Our dualistic conception isn't an airy intellectual thing; it is common sense, and rooted in a phenomenological experience. We do not feel that we are material things, physical bodies. The notion that we are machines made of meat, as Marvin Minsky once put it, is unintuitive and unnatural. Instead, we feel as if we occupy our bodies. We possess them. We own them. Because of this, we talk about my brain, or my body, using the same language of possession that we use when we talk about my car, or my child. These are things that we possess, that we are intimately related to—but not what we are.

This dualist perspective explains certain intuitions that we have about personal identity. We readily accept and make sense of situations, real or fictional, where a person stays the same but their body undergoes radical changes. In 13 going on 30, a teenager wakes up as Jennifer Garner, just as a 12-year-old was once transformed into Tom Hanks in Big. Characters can trade bodies, as in Freaky Friday, or battle for control of a single body, as when Steve Martin and Lily Tomlin fight it out in All of Me. This body-swapping is not Hollywood invention. We make sense of Kafka's Metamorphosis where this guy goes to sleep one night and wakes up as a cockroach, or Homer’s Odyssey, where sailors are transformed into the bodies of swine. In such cases, the soul is unchanged, only the body is different.

In fact most people around the world believe that an even more radical transformation actually takes place. Most people believe that when the body is destroyed, the soul lives on. It might ascend to heaven, or descend to hell, go off into some sort of parallel world, or occupy some other body, human or animal. Even those of us who do not hold such views have no problems understanding them. But they are only coherent if we see people as separate from their bodies.

Our dualistic perspective also affects how we think about such moral and political issues as stem-cell research, abortion, animal rights, and cloning. These are complicated issues, but the way people tend to address them—often explicitly, but always implicitly—is in terms of the question: Does it have a soul? If so, then the being in question is worthy of protection, a precious individual. If not, it is a mere thing.

In the case of abortion, our common-sense dualism can support both sides of the issue. We use phrases like "my body" and "my brain", describing our bodies and body parts as if they were possessions, and some people insist that all of us—including pregnant women--own our bodies, and therefore can use them as they wish. On the other hand, the organism residing inside a pregnant body might well have a soul of its own, possibly from the moment of conception, and would thereby have its own rights. President Clinton, one of our most scientifically literate presidents, was at a town meeting ten years ago, and he discussed abortion. He described the controversy as a reasonable disagreement among moral people. Nobody doubts the preciousness of human life. What they disagree about is an empirical issue: Precisely when does the soul occupy the body?

Or take cloning. About three years ago the Vatican issued a statement against human cloning. There are all sorts of reasons to be against human cloning, but the Vatican raised an interesting point. They claimed that clones would not have identical souls. While doctors and scientists might be able to create new bodies, only God could create new souls. I think some people take the next step here, and the worry lurking in a lot of minds is that with human cloning, you might end up with soulless bodies. And if you believe that, which is kind of scary, you don't want human cloning.

Admittedly, some people wouldn’t be caught dead talking about souls or spirits. But even for those people who would explicitly reject the notion of a body-soul split, dualist assumptions still frame how these issues are thought about. You can see this when people appeal to science to answer the question "When does life begin?" as if this is an empirical question, and an objective answer would settle the moral debate once and for all. But the question is not really about life in any biological sense. It is instead asking about the magical moment at which a cluster of cells becomes more than a mere physical thing. It is a question about the soul.


I am a developmental psychologist by training and by inclination, and so the question that fascinates me is this: Where does common-sense dualism come from?

One reasonable answer is it is learned. Children are raised in environments where they hear dualistic stories, they see movies where souls are depicted as independent from bodies, and they usually get some sort of religious training. And this dualism is inherent in the language that they learn; when we talk about the relationship between a person and his or her brain, we use the language of possession, not of identity.

There are also certain universal experiences that support a dualist worldview, such as the sensation of leaving one’s body in a dream, or the experience of our bodies disobeying our will. Saint Augustine uses the examples of impotence and involuntary sexual arousal here, seeing them as divine punishment after the Fall. As Garry Wills states, "the chanciness of arousal shows the loss of the integrity, the unison, of body and soul". But the experience of the unfaithfulness of our bodies need not wait for sexual dysfunction. It is experienced by any child who howls in frustration at learning to crawl.

So it is perfectly plausible that children start off innocent of any body-soul separation, and come to be dualists through experience. But I want to defend a very different view. I think children are dualists from the start. Even babies start off with this sort of body-soul split. To put it somewhat differently, they start off with two distinct modes of construal, or systems of core-knowledge, one corresponding to bodies, the other to souls. Because these systems are distinct, common-sense dualism emerges as a natural by-product.

For much of our recent intellectual history, in philosophy and psychology, this claim about babies would be thought to be utterly ludicrous. Total madness. It was said that babies and young children know nothing about bodies, and know nothing about souls. They are blank slates. Rousseau called the baby a perfect idiot. And Jean Piaget, the Swiss psychologist who got developmental psychology going in the last century, was adamant that babies have no notion of what an object is, and no notion of what a person is.

But over the last two decades there is decisive evidence showing that this minimalist perspective is wrong. In fact, babies, before they hit their first birthday, have a rich and intricate understanding of bodies and of souls.

First, there is a lot of research from psychologists like Elizabeth Spelke, Renee Baillargeon and Karen Wynn showing that young babies have a powerful understanding of physical objects. They understand that physical objects obey gravity. If you put an object on a table and remove the table, and an object just stays there - because there's a hidden wire - babies are surprised, they expect the object to fall. They also expect objects to be solid, and that objects should move on continuous paths over space, And, contrary to Piaget, they don't think that once you look away objects go out of existence. They understand that objects persist over time even when they aren't being observed. Show a baby an object, and then put it behind a screen. Wait a little while and then remove the screen. If the object is gone, babies will be surprised.

Karen Wynn showed that babies can even do addition and subtraction, of a rudimentary sort. You put an object down, a screen rises to hide it, and then you put another object behind the screen. Then the screen drops, and there is one object, two objects, or three objects. If there's one object there, babies are surprised. If there are three objects, babies are surprised. They know one plus one equals two.

Second, even young babies are social creatures. They prefer to look at faces over just about anything else. They quickly come to recognize different emotions—anger, fear, happiness. They imitate people. As soon as they begin to move their bodies in different ways, they can do clever things to manipulate emotions and behaviors of other people.

Then there's a lot of recent work from people like David Premack and Gyorgy Gergely and also from my own lab, that before children learn how to talk, they can make sense of social interaction. A typical experiment involves showing them movies where a character moves in a way that makes sense from an adult perspective—pursuing a goal, moving away from something—or in a way that doesn't make sense from an adult perspective. What we find is that even children before their first birthday get it. They expect people to act in certain ways. They expect people to pursue goals.

In some work that I've done with Valerie Kuhlmeier and Karen Wynn, for instance, we found that when babies see one character help somebody, and another character hurt that person, they later expect the person to approach the one that helped it, and to avoid the one that hurt it. This is sophisticated social knowledge. (The article is available on my lab web site, along with much of the other work I am talking about here.)

That the foundations for a body-soul understanding are present early on does not make children dualists. It's possible that they have some understanding of bodies and souls, but they don’t view them as separate. This issue is far from settled. But I want to give two examples that suggest to me that young children do have a dualistic understanding of the world.

The first involves their understanding of the brain. Young children start off now knowing what the brain does. That makes sense; it was a scientific discovery that the brain was involved in thinking, But once they do learn about the brain they develop an odd and interesting misunderstanding. I can illustrate this with a story about my son, Max, when he was six years old.

We were having an argument because he had to go to bed. I told him, You have to go to bed, it's very late, and he said, You can make me go to bed, but you can't make me go to sleep. It's my brain! This got me interested. I said, Okay, fine, stay up, let's talk. I got out a piece of paper, and started asking him questions about what he thinks the brain does. I gave him lists of things - does the brain do this, does the brain do that?

His answers showed an interesting split. He agreed the brain is involved in perception - in seeing and hearing, in tasting and smelling. He'd been taught that. And he agreed that the brain is particularly important with regard to conscious problem-solving: Solving a math problem, making sense of a story, planning what to do. But he said the brain didn't do certain things: it didn't do dreaming, it didn't do loving his brother, it didn't do pretending to be a kangaroo. Max said, that's what I do, though my brain might help me out.

I later found out when I looked at this literature that there are several experiments supporting this conception as typical for a child his age. Once children learn that the brain is involved in thinking, they don't take it as showing that the brain as the source of mental life; they don't become materialists. Rather they interpret "thinking" in a narrow sense, and conclude that the brain is a cognitive prosthesis, something added to the soul to enhance its computing power. In other words, there's Max, the person, and then there's his brain, which he uses to solve problems just as he might use a calculator or a computer.

I don't think that this is an entirely immature way of thinking. For instance when adults read that a certain part of the brain is involved when you think about sex, or race, or politics, they are often surprised. And I think this surprise is revealing. After all, the details of the studies may be of scientific interest, but the mere fact that the brain is involved should be boring. In fact, it would be the discovery of the century if some sort of thinking went on that did notinvolve the brain. And when they describe such results, even experts slip into dualistic language: "I think about sex and this activates such-and-so part of my brain"—as if there are two separate things going on, first the thought and then the brain activity.

A second sign of early dualism concerns when children think about the afterlife. The best study on this was done by Jesse Bering and David Bjorklund. They told young children a story about a mouse and an alligator, where the mouse is running around and the alligator comes up to him and chomps down, and the mouse is no more. The mouse is dead. Then they asked the children questions. Some of these are questions about the biology of the mouse. Now that the mouse is not alive, does its brain still work? Now that the mouse is not alive, will it grow up to be a big mouse? For these the children tended to say No. They understand that once the mouse is dead, his brain is not going to work; he isn’t going to need any food.

What is more interesting is that they also asked children about the mental life of the mouse. Now that the mouse is not alive, does he still love his mother? Does he still like cheese? Does he know he's dead? For these questions the children tended to answer Yes. So they tacitly believe that even though the mouse is not alive any more, its mental life persists. This is the foundation for the more articulated view of the afterlife you usually find in older children and adults.

I take this evidence as supporting the claim that children are natural-born dualists. I should add that I have done some other research—in collaboration with Valerie Kuhlmeier and Karen Wynn—that suggests that 5-months-olds see bodies and souls as entirely distinct. They are super-dualists. Normal dualists appreciate that people are both body and soul (or more precisely, a soul inhabiting a body), while babies see people as entirely immaterial beings, and hence not subject to physical constraints such as solidity and gravity and continuity. As you might imagine, this is a somewhat controversial claim, and there is a bit of a back-and-forth about our research that will appear in the journal Cognition (see again my web site if you want to look at the papers).


A different aspect of my work looks at the implications of common-sense dualism. Some of this research concerns how children and adults think about artwork, and I’m also exploring humor (particularly slapstick humor, which is rooted in a sense of body-soul duality). But the work that I am most excited by right now concerns its implications for moral reasoning.

The framework that I adopt is the philosopher Peter Singer's notion of a moral circle—the circle of things that matter to us. This circle can be very small, including just your kin and those you interact with on a daily basis. Friends and family. Or it can be extremely broad, including all humans but also fetuses, and animals, and even complicated computers. For most of us, it falls somewhere in the middle.

What causes the circle to expand and to shrink? I think you can make some progress towards an answer—both historically and developmentally—by recognizing that we have two ways of seeing a person, as body and as soul. Normally when we interact with people we see them as both. We appreciate that they have beliefs, desires, and consciousness, and so on, and we appreciate that they're physical things that take up space, that are subject to gravity, and can be moved around. Both stances coexist pleasantly enough in the normal course of things. But when you emphasize one perspective over another, you get moral consequences.

Social psychologists, for instance, have shown that simply getting an experimental subject to take another person’s perspective will make the subject care more about the person and be more likely to help them. Focusing on the soul, then, leads to moral concern. This can expand the circle.

Then there is the more sinister side, the shrinking of the circle. One route to this is what happens when you see someone solely as a body, and one emotion that supports this construal is disgust. Paul Rozin has done a lot of work showing how, as Charles Darwin first said, disgust is an adaptation towards veering us away from bad meat, and so it is naturally triggered by animals and animal waste products. And so some things are universally disgusting: Rotten meat, feces, urine, blood, vomit. But disgust can readily extend to people. People, after all, are made of meat.

It has been long observed that every movement designed to stigmatize or malign some group - Jews, black, gays, the poor, women - has used disgust. Once you get somebody to view a group of people as disgusting, the attention shifts away from them as people, as moral individuals. They become soulless bodies, and the circle closes in to exclude them.

This response to soulless bodies is one reason why I call my recent book, "Descartes’ Baby". The more obvious reason for the title is that I claim that people are mind-body dualists, and so they naturally hold a philosophy that's most famously associated with René Descartes. In this sense, we are all Descartes’ babies. But there is another reason for the title that relates to disgust and similar emotions.

It is based on a story that was told about Descartes after he died. Descartes was never popular; he had a lot of enemies when he was alive, and even more after he died, and the story about him is weird and faintly nasty. It was known that Descartes had an illegitimate daughter, Francine, who died when she was five years old; this was said to be the greatest tragedy of his life. The story went that Descartes was so struck with grief that he created an automaton, a mechanical doll, built exactly identical to his dead daughter. The two were inseparable. He took this mechanical doll with him wherever he went. It was kept in a small trunk, and wherever he slept she was by his side. The story is that Descartes was crossing the Holland Sea, and the captain of the ship became very curious about the contents of this trunk that Descartes always had by his side. One night the captain crept down to his cabin while he was sleeping and opened up the trunk. To his horror the robot Francine arose. The captain, struck with revulsion, grabbed her, dragged her up to the deck of the ship and threw her overboard.

I like this story because it captures how disturbing—in some cases, revolting—we find a body without a soul. It is a nice illustration of the emotional pull our common sense dualism could have. But it also raises a serious problem. Common-sense dualism is wrong. There is no consensus as to precisely how mental life emerges from a physical brain, but there is no doubt that this is its source. If by ‘soul’, then, you mean something immaterial and immortal, then souls do not exist. All of us are soulless bodies, no less than the robot Francine. In this sense too, we are Descartes’ babies.


Some scholars are confident that people will come to accept the scientific world-view, and reject the notion of an immaterial soul. I am much less optimistic.

People do believe all sorts of things that violate common sense. Some philosophers have argued that everything in the world is made of water, others that there is no such thing as pain. It has been claimed that there is no objective morality, and that the external world does not exist. Some people think that thermostats have beliefs; others argue that rocks have a form of consciousness. Some suggest that each brain contains two conscious entities (one for each hemisphere), and some doubt that consciousness even exists. We can add to this list of crazy views what Francis Crick called ‘the astonishing hypothesis"—the view that dualism is wrong, that mental life is the product of a purely physical brain.

People might sincerely believe these things. (I certainly believe the last one.) But such beliefs exist at a different level than gut feelings. They are more fragile, and less embedded in our everyday lives. The most severe moral relativist, if he were to see someone murder a child, would feel that it is very wrong indeed. A radical behaviorist can’t help but wonder what other people think of her; and there really are no atheists in foxholes. People can reject dualism at a conscious level, but the intuitive sense that body and soul exist is here to stay.

What about the more modest proposal that people will come to reject dualism at an explicit conscious level? In the domain of bodies, after all, most of us accept that common sense is wrong. We concede that apparently solid objects are actually mostly empty space, consisting of tiny particles and fields of energy. Perhaps the same sort of reconciliation will happen in the domain of souls, and it will come to be broadly recognized that our dualist belief system, though intuitively appealing, is factually mistaken. Perhaps we will all come to agree with Richard Dawkins and Daniel Dennett and join the side of the "brights" [*]: those who reject the supernatural and endorse the world-view established by science.

But I am skeptical here too. The notion that our souls are flesh is profoundly troubling to many, as it clashes with religion. Dualism and religion are not the same: You can be dualist without holding any other religious beliefs, and you can hold religious beliefs without being dualist. But they almost always go together. And some very popular religious views rest on a dualist foundation, such as the belief that people survive the destruction of their bodies. If you give up on dualism, this is what you lose.

This is not small potatoes. The insights of neuroscience are a much harder pill to swallow than, say, evolutionary biology. A religion such as Judaism or Catholicism might survive even if it comes to reject a literal account of God creating man and animals. But it cannot survive the rejection of the immaterial soul. Pope John Paul II was clear about this. A few years ago, he famously conceded that our bodies may have evolved, and that the Darwinian theory of evolution might well be true. But he drew the line at souls, stating that theories "which consider the mind as emerging from the forces of living matter, or as a mere epiphenomenon of this matter, are incompatible with the truth about man." Over 300 years ago, a philosopher named Henry Moore expressed this view in even sharper terms, writing: "No spirit, no God."

When people hear about research into the neural basis of thought, they learn about specific findings: this part of the brain is involved in risk taking, that part is active when someone thinks about music, and so on. But the general assumption underlying this research, that of the physical origin of mental life, is not generally appreciated, and it is interesting to consider how people will react when it is. The clash between dualism and science will not easily be resolved, and the stakes are high. The same sorts of heated controversies that raged over the study and teaching of evolution over the last hundred years are likely to erupt over psychology and neuroscience in the years to come.

[* Ed. Note: See Bright!!]

Pamela McCorduck on "Aristotle" by W. Daniel Hillis

Pamela McCorduck

Danny's idea has indeed appeared before, and the reason why is because it's what we yearn for, what we desperately need. AI may not quite be ready to take on everything Danny suggests, so why not a set of gradually improving interim systems? By now, most adults know how we learn best and thus we can supply some of the missing intelligence. We can say show me a picture, or provide me with some analogies; we can ask about the authority of the source. Though I use it far more than I ever dreamed, the current web does not let me interact this way. So why not have adults who want and need to learn new things beta-test an incomplete, imperfect automated tutor? Sign me up!

We may not know everything about how a given mind learns, but after 30 years of serious research in cognitive psychology, we know far more than we once did. Tragically, precious little of this has migrated into mainstream teaching, but that's another story.

Procrastination has allowed me to see other people's responses to Danny's proposal. I'm surprised at how much skepticism it has evoked, even claims that it could probably never be done. My goodness, I think Edge needs to poll a younger group.

PAMELA McCORDUCK's classic Machines Who Think has just been reissued in a 25th Anniversary edition.