tHURSDAY, MARCH 6, 2003

Steven Pinker, PhD, Peter de Florez Professor,
Dept. of Brain and Cognitive Sciences,
Massachusetts Institute of Technology

CHAIRMAN KASS:  Well, I think we should get started.  We have a few stragglers who will, I'm sure, wander in promptly.

The topic this afternoon is for the first time in our meetings the subject of human nature and human nature in the age of biotechnology.  The subject crops up now and then in our conversations and is very often just below the surface as one talks about various kinds of technical innovations which at least some people claim might produce certain kinds of changes in human nature.

And in these kinds of conversations very quickly one gets into questions of whether there is such a thing as human nature, whether there is something fixed and hard wired, or whether it's primarily plastic, or whether, as is frequently said, it is the course of the essence of human nature to change human nature, and on and on and on.

There have also been some conversations here where human nature has functioned not simply descriptively, but also normatively as something which is either thought to be sacrosanct and offering some kind of guidance or, on the contrary, as something which is so filled with flaws that it needs, in fact, to be improved, our senescence in mortality being one amongst those flaws that we discussed improving in the session before lunch.

And it seems to me that we thought it was worthwhile to actually make this a subject of explicit conversation and spend some time on it.  This is not a public policy question.  There are not going to be recommendations.  There are not even going to be "thou shalts" and "thou shalt nots" coming out of this, but it does seem to me it's worth our while to pay some attention to this larger theme in an explicit way, and especially to think about how to think about human nature in an age of genomics, in an age of neuroscience, both how we should understand it, to understand what might be possible in the way of altering it and ultimately what those alterations might mean and whether they would be a good thing, large questions all, and we are very lucky to have as our special guest someone for whom these large questions are, to say the least, not daunting because he's willing to step forward and speak about them, and that's Professor Steven Pinker, a neuroscientist and evolutionary psychologist and a very gifted and prolific, popular author about these matters.

He's the Peter D. Florez Professor in the Department of Brain and Cognitive Science at MIT, the author of recent books, How the Mind Works, and more recently The Blank Slate, and he has very kindly agreed to come and introduce us to this topic with a formal presentation after which all of us look forward to having conversation with you.

Thank you very much and welcome,

DR. PINKER: Thank you very much.  I'd like to thank Dr. Kass for the opportunity to speak to this group.  It is really an honor and a privilege to share these ideas with you.  Thank you.

I'm going to talk about the modest topic of the past, present, and future of human nature with an emphasis on the future.

What about the past?  In much of the 20th Century, there was a widespread denial of the existence of human nature in Western intellectual life, and I will just present three representative quotations.  "Man has no nature," from the philosopher Jose Ortega y Gassett.  "Man has no instincts," from the anthropologist and public intellectual Ashley Montagu.  "The human brain is capable of a full range of behaviors and predisposed to none," from the evolutionary biologist Stephen Jay Gould.

I think, however, that in recent times there has been a rediscovery and a reacknowledgement of the idea that humans have a nature as well as a history.  Partly it's an acknowledgement of common sense.  Anyone who has had more than one child knows that children are not indistinguishable lumps of putty waiting to be changed, but come into the world with certain talents and temperaments.

Anyone who has both children and house pets has surely noticed that the children exposed to language will develop language, in turn, whereas the house pets will not.

There has also been a reacknowledgement of universals across human societies, although it's undeniable that human societies and cultures differ from one another in countless ways.  There is also a large stock of universal behaviors and emotions that can be found in all of the world's 6,000 cultures.

Here is a list recently compiled by the anthropologist Donald Brown that goes from aesthetics, affection, and ambivalence all the way down to fallow contrasts, weaning, weapons, and attempts to control the weather.

There has also been an increasing body of data from behavioral genetics and cognitive neuroscience, suggesting that the human brain has a complex inherent structure.  This is a recent study from Paul Thompson and colleagues based on earlier work from your own Mike Gazzaniga, which used magnetic resonance imaging to measure the distribution of gray matter in different parts of the cerebral cortex and correlated it across a large sample of pairs of individuals.  They coded the correlation in false color so that zero correlation was represented in shades of blue and purple, and statistically significant correlation in shades of green, red, and pink.

Now, by definition if you pick people at random and correlate the gray matter in different parts of the brain, the correlation will be zero, and so in unrelated subjects you have view of the left hemisphere, the right hemisphere, and a top view that is uniformly zero.  This is what happens in pairs of people who share half their genes, namely fraternal twins.  As you can see, most of the cortex shows statistically significant correlations in how much gray matter is found in different areas.

This is what happens in people who share all of their genes, namely, identical or monozygotic twins, and as you can see, they're even greater extents of cortex that are highly correlated across pairs of individuals.

Now, these correlations are not just meaningless anatomical shapes like the shape of your earlobes, but have behavioral consequences, and studies of twins and adoptees have shown substantial genetic influences on personality and intellect.

My favorite summary is from the Charles Addams cartoon in The New Yorker whose caption is "separated at birth, the Malliefert twins meet accidentally," showing a pair of inventors with identical contraptions in their laps in the waiting room of a patent attorney.

The cartoon is not such an exaggeration on the data.  Studies of identical twins who are separated at birth and reunited in adulthood show that they share astonishing similarities in their personalities, in their intellects, and even in individual quirks, like dipping buttered toast in coffee and wearing rubber bands around their wrists.

Well, that's the past and the present.  Given that human nature exists as common sense and the empirical data tell us, does that mean that we can change it?

Now, there have been some notorious attempts to change human nature that we've seen in the 20th Century.  There has been the attempt to socially engineer a new man, in particular, a new socialist man leading to the totalitarian regimes in the Marxist dictatorships in Russia, China, and Cambodia.  I think it's fair to say that this is no longer a topic of debate among decent individuals.

Equally horrific has been the attempt to change human nature through eugenics both in the case of mandatory of sterilization that was widespread in many Western countries, including the United States, until the 1930s, and even more horrifically, the Nazi genocide, which was predicated on the desirability of changing human nature through sterilization and mass murder.

I'm going to talk about the ability to change human nature that's of more direct interest to the members of this committee, namely, voluntary genetic engineering, popularly known as designer babies, and that will be the topic of the rest of my presentation.

I don't have to remind you that this is ethically fraught, and there are vociferous voices arguing that this would be a bad thing or that it would be a good thing.  I'm going to address a common assumption both of people who are alarmed and people who welcome genetic enhancement.

The assumption that this is inevitable, that science has reached the point where it's only a matter of time before genetic enhancement is routine and possibly the human species will change unless we intervene and regulate the science and practice now.

I'm going to present a skeptical argument about designer babies to give you an overview.  I'm going to suggest that genetic enhancement of human nature is not inevitable.  Indeed, I would be willing to venture that it's highly unlikely in our lifetimes.

Why?  First of all, because of the fallibility of predictions about complex technology in general.

Secondly, impediments to genetic enhancement from what we know about the human behavioral genetics.

And, third, impediments from human nature itself.

Well, let me begin with the frailty of technological predictions in general.  There's a wonderful book called The Experts Speak by Victor Navasky and Christopher Cerf which has some delicious quotations about what is inevitable in our future, such as the following one.  "Fifty years hence we shall escape the absurdity of growing a whole chicken in order to eat the breast or wing by growing these parts separately under a suitable medium," Winston Churchill in 1932.  That should have happened by 1982, and we're still waiting.

Nuclear powered vacuum cleaners will probably be a reality within ten years, a prediction made in 1955 by a manufacturer of vacuum cleaners.  A few other predictions that I remember from my childhood, and in fact, from newspapers of just a few years ago.  Dome cities, jet pack commuting, mile high buildings, routine artificial organs, routine consumer space flights, such as the Pan Am shuttle to the moon featured in 2001, interactive television, the paperless office, and the dot-com revolution and the end of bricks-and-mortar retail.  All of these predictions we know to be false, and a number of them are not even developments that have not happened yet, but things that we can say with a fair amount of confidence never will happen.

We're not going to have domed cities, at least not in the future that's worth worrying about.

Now, why are technological predictions so often wrong?  First, there's a habit of assuming that technological progress can be linearly extrapolated.  If there's a little bit of progress now, there will be proportional progress as we multiply the number of years out.

Engineers sometimes refer to this as the fallacy of thinking that we can get to the moon by climbing trees.  A little bit of progress now can be extended indefinitely.

Secondly, there's a tendency to underestimate the number of things that have to go exactly right for a given scenario to take place.  Most technological changes don't depend on a single discovery, but rather on an enormous number of factors, scores or even hundreds, all of which have to fall into place exactly right. 

Both technological developments, psychological developments, namely, whether individual humans will opt for the technology both in developing it and in adopting it, and sociological factors, namely, whether there will be a multiplication of those choices society-wide that will lead to the economies of scale and the social pressures that would lead to some technological development becoming ubiquitous.

Third, there's a widespread failure of futurologists to consider the costs of new technologies, as well as the benefits, whereas in reality the actual users faced with a particular technology consider both the benefits and the costs.

Finally, there is an incentive structure to futurology.  Someone who predicts a future that's radically different from our own, either to hype it or to raise an alarm against it will get the attention of the press and the public.  The chances are The New York Times won't call you up if you say either that the future is going to be pretty similar to the present or we haven't a clue as to what the future will be.

The second part of my talk, reasons for skepticism about designer babies is that there's a considerably bracing splash of cold water on the possibility of designer babies from what we know about behavioral genetics and neural development today.  There's a widespread assumption that we have discovered or soon will discover individual genes for talents such as mathematical giftedness, musical talent, athletic prowess, and so on.

But the reality is considerably different, and I think an Achilles heel of genetic enhancement will be the rarity of single genes with consistent beneficial psychological effects.  I think there's a myth that such genes have been discovered or inevitably will be discovered, but it isn't necessarily so.

Indeed, I would say that the science of behavioral genetics at present faces something of a paradox.  We know that tens of thousands of genes working together have a large effect on the mind.  We know that from twin studies that show that identical twins are far more similar than fraternal twins who, in turn, are more similar than unrelated individuals, and from adoption studies that show that children resemble their biological parents more than their adopted parents.

But these are effects of sharing an entire genome or half of a genome or a quarter of a genome.  It's very different from the existence of single genes that have a consistent effect on the mind, which have been few and far between.

Anyone who has kept up with the literature on behavioral genetics has noticed that there's been a widespread failure to find single genes for schizophrenia, autism, obsessive-compulsive disorder, and so on.  And those, by the way, are the areas where we're most likely to find a single gene simply because it's easier to disrupt a complex system with a single defective part than it is to install an entire complex ability with a single gene.  The failure to find a gene with consistent effect on, say, schizophrenia means that it's even less likely that we will find a gene for something as complex as musical talent or likability.

And though there have been highly publicized discoveries of single genes for syndromes such as bipolar illness, sexual orientation, or in perhaps the most promising case, a gene that appeared to correlate with four IQ points in gifted individuals; all of those discoveries have been withdrawn in recent years, including the four point IQ gene withdrawn just last month.

Now, it's really not such a paradox when you think about what we know about biological development in general.  The human brain is not a bag of traits with one gene for each trait.  That's just not the way genetics works.

Neural development is a staggering complex process which we are only beginning to get the first clues about.  It involves many genes interacting in complex feedback loops.

The effects of genes are often non-additive.  The effect of one gene and the effect of a second gene don't produce the sum of their effects when they're simultaneously present necessarily.

The pattern of expression of genes is often as important as which genes are present, and therefore, it's a good idea not to hold your breath for the discovery of the musical talent gene or any other single gene or small number of genes with a large, consistent effect on cognitive functioning or personality.

As an analogy, we know that the code that comes with a software package, that is a software package obviously determines the operation of a computer, and we know that properties of a computer package, such as how easy it is to use depend intimately, completely on the sequence of instructions in the software.

That doesn't mean that there is a single instruction that you can insert into a computer program that will make it easy to use, nor a single instruction that you can remove that will automatically make it hard to use.

I think there are other genetic impediments to the possibility of genetic enhancement.  One is that the genes, even acting across an entire genome, have effects that are, at best, probabilistic.  A sobering discovery is that monozygotic twins reared together who share all of their genes and most of their environment are imperfectly correlated.  When it comes to personality measures, such as extroversion or neuroticism, correlations are in the range of .5.

Now, that's much, much bigger than correlations among non-identical twins or, let alone, unrelated individuals, but it's much less than one, and what that tells us is that there is an enormous and generally unacknowledged role for chance in the development of a human being.

Secondly, there's a phenomenon of pleiotropy that most genes have multiple effects, and in general, evolution selects for the best compromise among the positive and negative effects that come from an individual gene.

A vivid example of this is aside from the four point IQ gene, probably the best candidate for a gene with the potential for enhancement is the knock-in mice reported two years ago that were given extra MNDA receptors, receptors that are critical to learning and memory.  These were artificially engineered mice that had an enhanced ability to learn mazes.

On the other hand, it was later discovered that these mice were hypersensitive to inflammatory pain.  So a genetic change had both a positive and negative effects.

Because of this, it means that there are ethical impediments to research on human enhancement, namely, how can you get there from here.  Are there experiments that a typical human subjects committee would approve of, given the likelihood that any given gene will have negative effects on a child, in addition to the positive ones.

Finally, most human traits are desirable at intermediate values.  Wallace Simpson famously said that you can't be too rich or too thin, and it may be true that you can't be too smart, but for most other traits, you really can have too much of a good thing.

Most parents don't want their child to be not assertive enough, to be a punching bag or a door mat.  On the other hand, most parents would also not want their child to be Jack the Ripper.

You want your child to have some degree of risk taking, not to sit at home cowering out of fear of negative consequences.  On the other hand, you don't want a self-destructive maniac either.

So if a given gene, even if it did have as its effect an enhancement, say, of risk taking, put it in a child and you'll have ten extra points on the risk taking scale; the crucial question is:  what are the other 29,999 genes doing?  Would they be placing your child on the left-hand side of the Bell curve, in which case an extra dose of assertiveness would be a good thing, or have they already put your child on the right-hand side of the Bell curve so that an extra dose of assertiveness is the last thing that you would want?

The third part of the argument is I think there are impediments in human nature to enhancing human nature.  Now, one feature of parental psychology that is often invoked in these discussions is the desire of parents to give their children whatever boost is possible, and lurking in all of these discussions is the stereotype of the Yuppie parent who plays Mozart to the mother's belly while the mother is pregnant, bombards the baby with flash cards, has them taking violin lessons at the age of three, and so on.  And the assumption is that parents would stop at nothing to enhance their children's ability, including genetic engineering.

Well, that obviously is a feature of parental psychology, but there's a second feature of parental psychology that also has to be factored in, namely, the aversion to harm your children.  Most parents know that even if they are not sure whether playing Mozart to a pregnant woman's belly will help their child, they have reasonable belief that it couldn't harm the child.  Likewise the flash cards, the violin lessons, and so on.

If it came to genetic enhancement where this was unknown, it's not so clear that parents would opt for the risk of doing their children genuine harm for the promise of a possibility of doing them good.

Also, one ubiquitous feature of human nature is intuitions about naturalness and contamination, sometimes referred to by cognitive psychologists as psychological essentialism, the folk belief that living things have an essence which can be contaminated by pollutants from without. 

This has been an impediment to the acceptance of other technologies.  Famous examples are nuclear power, which is notoriously aversive to large segments of the population.  As you all know, there hasn't been a new nuclear power plant built in this country for several decades, despite the possibility that it could be an effective solution to global warming.

In Europe and in large segments of this country, there is a widespread repugnance to genetically modified foods for reasons that are probably more irrational than rational, but nonetheless cannot be gainsaid.  If people have a horror about genetically modified soybeans, it's not so clear that they would rush to welcome genetically modified children.

Finally, anyone who knows someone who has undergone IVF knows that this is a traumatic, painful, and rather unpleasant procedure, especially in comparison to sex.  While there are undoubtedly extremists who would use IVF, we know that they would use IVF for things as trivial as having their child born under a certain astrological sign; it's certainly not true that everyone would shun IVF for trivial reasons.  There is reason to believe that this would not necessarily catch on in the population as a whole.

So the choice that parents would face in a hypothetical future in which even if genetic enhancement were possible would not be the one that's popularly portrayed, namely, would you opt for a procedure that would give you a happier and more talented child.

When you put it like that, well, who would say no to that question?

More realistically, the question that parents would face would be something like this.  Would you opt for a traumatic and expensive procedure that might give you a very slightly happier and more talented child, might give you a less happy, less talented child, might give you a deformed child, and probably would do nothing.

We don't know the probabilities of those four outcomes.  I think this is a more realistic way of thinking about the choices that parents might face.

For genetic enhancement to change human nature or to lead to a post human future, not a few, but billions of people would have to answer yes to this question.

So to sum up, changing human nature by a voluntary genetic enhancement I would say is not inevitable because the complexity of neural development and the rarity or absence of single genes with large, consistent, beneficial effects, and because of the tradeoff of risks and benefits enhancement that will inevitably be faced by researchers and by parents.

The conclusions that I would draw are the following.  I am not arguing that genetic enhancement will never happen.  If there's anything more foolish than saying that some technological development is inevitable, it's saying that some technological development is impossible.

And corresponding to the silly predictions about the inevitable future of domed cities and jet packed commuters, one can find equally silly quotes from people who said things like we will never reach the moon.

So it's not that I am arguing that genetic enhancement is impossible.  Rather, it's an argument that bioethics policy should acknowledge the frailty of long-term technological predictions which have a very spotty track record at best.  The bioethics policy should be based on fact, not fantasy.  Both our positive and our negative fantasies are unlikely to come true, and that policies predicated on the inevitability of genetic enhancement should be rethought.

I thank you for the opportunity to present these views to the council.

CHAIRMAN KASS:  Thank you very much.  A very crisp, clear and interesting presentation.

The floor is open for discussion.  Robby George.

PROF. GEORGE:  Yes.  Thank you, Dr. Pinker, for that wonderful presentation.

You mentioned at one point the publicity surrounding claims that genes had been identified which operating just on their own have certain determinable effects, like the four IQ points.  You yourself are a person who's very much in the public media.

One thing that I notice about these sorts of claims is that when they're made, they get an enormous amount of publicity, and when they're withdrawn, you hear about it later, if at all.

I didn't know about the four IQ points had been withdrawn.  It seems to me that that's an enormous problem on the public education side, and it's not one that we don't face in the bioethics area as well because so much of what needs to be done really does require the public to have a realistic picture of what's going on in the sciences.

Do you have any reflections about that?  I mean particularly about the question of communicating scientific information that's relevant to bioethical decisions in the public media?

DR. PINKER: Yes.  It's something I have thought about a great deal.  There is an inherent, I think, problem in science journalism, which is that it is journalism, and science doesn't work on the same timetable.  It doesn't work on the same kind of database.

Editors, not surprisingly, want news.  They want to hear about things that have happened yesterday or this morning, and many of the scientific journals go along with this mentality by having embargoes and building suspense on the development, releasing it at a particular time, knowing that it will appear in The New York Times the next day.

Science, especially the science of the human mind, which is a fallible, halting, slow process, depends not on individual discoveries which seldom have a huge, long-term impact, but on the accumulation of dozens or hundreds of studies which all point in a given direction or not.

The way that I think scientists proceed or ought to proceed is they look at meta- analyses and literature reviews and assessments of a large literature that begin to emerge years after the first discovery.  The way that journalism works is reporting individual discoveries, and I think that's a built in bias in science journalism that inevitably lead to the kind of misinformation that you've alluded to.

Science journalists are not going to get their stories published if they simply look at -- in large part.  There are exceptions -- at, say, a review paper in a review journal that looks at a meta- analysis of ten years of research.  I mean, that does happen, but far more often stories that you read about are based on one discovery that was published in Science or Nature or the New  England Journal of Medicine the previous day.

PROF. GEORGE:  Can I follow up, Leon?

CHAIRMAN KASS:  Go ahead.

PROF. GEORGE:  On that, yeah, to what extent can responsibility be ascribed to research scientists themselves who are involved in these episodes?  Is it simply that they publish their research and then the journalists get hold of it and there it goes, or are there incentives for research scientists to sometimes -- I don't know if perhaps grants or what -- but are there incentives that would lead people perhaps to hype discoveries that aren't really verified?

DR. PINKER: No, there is certainly that, and it would be highly misleading of me to say that this is a problem that comes from science journalism because clearly it's also part of the incentive structure among the scientists themselves.  That's absolutely true.

CHAIRMAN KASS:  Alfonso.

PROF. BLACKBURN:  I just had one word.  It's called "ego," Robby.

(Laughter.)

PROF. GEORGE:  We have that in the humanities and social sciences as well.

CHAIRMAN KASS:  Alfonso Gómez-Lobo, please.

DR. GÓMEZ-LOBO:  This was a very clear and persuasive presentation.  I think I've persuaded myself that I'm not going to see genetically modified babies or designer babies at least in my immediate family in a short time.

But I'm really wondering, what is the notion of human nature with which you're operating here?  Because, of course, the expression has represented many different concepts in history, and some of the traits that apparently could be modified, I mean, if we knew more about genes, et cetera, would be considered not really essential to human beings.  It would be considered accidental.

You know, human nature would not be changed if we are a couple of inches taller or something like that.  So I'm curious about that because, of course, in the writings that you gave us, again, the argument is very convincing, but part of it it's because at least to me it's unclear exactly what you mean by human nature as such.

DR. PINKER: Yes, that's a completely legitimate question.  I would characterize human nature as a set of emotions, motives, and cognitive abilities shared throughout the species by all neurologically normal individuals with quantitative variation, but much less qualitative variation across individuals.

To be concrete, every neurologically normal child learns a language upon exposure to it, but we also know that vocabulary size and verbal fluency vary quantitatively along a Bell curve across individuals.

Also, an important addendum to that is that what is universal in human nature is certainly not a set of behaviors because we know from National Geographic and Anthropology 101 that there's enormous variation from culture to culture in sexuality, in child rearing and religion and virtually every other trait.  And we know that those differences don't come from genetic differences among peoples because of the experiment known as immigration; that a child coming from one culture to another will pretty much or entirely show no genetic carryovers from the culture in which his ancestors belonged.

So whatever human nature consists of would be abstract abilities or motives that would translate themselves into actual behavior in radically different ways depending on the environment and the social circumstances.

Again, to come back to language as a touchstone, children clearly aren't born with genes for English or Swahili or Japanese.  They conceivably could be born with genes that predispose them to acquiring words with a sound meaning pairing, phrases with subjects and objects and nouns and verbs.  The abstract universal grammar that my colleague Noam Chomsky made famous, which doesn't correspond to any language that you actually use.

Similarly, in the domains of the emotions and motives, there isn't any particular behavior that is universal.  It's not the case that, for example, men are universally polygamous polygynous or monogamous.  That varies among individuals and among cultures.

Nonetheless, it may be true that the underlying desires are much more universal than the overt behavior.  We all remember President Carter who committed adultery in he heart many times.  As far as we know, he didn't commit it in reality even once.  This is, I think, a feature of psychology, namely, fantasy that may be much more uniform that actual behavior.

So human nature can't be equated with human behavior.  It refers to desires, tendencies, abstract abilities rather than to concrete acts.

CHAIRMAN KASS:  Alfonso, please.

DR. GÓMEZ-LOBO:  A quick question.  Couldn't we then characterize human nature as a set of capabilities, abilities, potentialities?  Since they can be realized in such different ways, one could say there's a potentiality for learning the language and for learning abstract predicates.

Now, whether they're expressed in Spanish or in English would really depend on the culture, but then my question would be:  what stage of development do human beings have that human nature?

DR. PINKER: I'm sorry.  At what stage of development, do you mean in the ontogeny of the individual, that is, childhood, or do you mean in cultural evolution and history?

DR. GÓMEZ-LOBO:  No, I mean in the individual because that's what's been and will continue to be a matter of dispute.

DR. PINKER: Well, the answer will be different for different aspects of our psychology.  Research on the minds of infants have shown that infants show many more human specific commutabilities that we formerly appreciated.

The whole idea of babies was that the world of the infant was a blooming, buzzing confusion, a famous phrase from Williams James; that a newborn basically saw the world as a kaleidoscope of fluctuating pixels and had to learn even that there was such a thing as an object.

More recently  clever techniques has shown that there are some, many abilities that seem to come on line very, very early in life.  Children from the day they're born lock onto human faces.  They recognize the sound of their mother's voice, the smell of their mother.  As soon as their visual systems are mature, they pay attention to objects, expect them not to disappear without a trace, pay attention to humans and their interactions, pay attention to speech, and so on.

So even though there's an enormous amount of learning that takes place, the learning abilities themselves seem to be up and running quite early in development.

CHAIRMAN KASS:  Gil Meilaender.

PROF. MEILAENDER:  Let's say that we're persuaded that the kind of enhancement that you were talking about is certainly not inevitable and maybe just not going to happen because it turns out to be very complicated and difficult in the variety of ways that you demonstrated.

What do we conclude from that about proceeding with the project?  In other words, is there some reason one ought not try it?

And one possible reason, for instance, would be that maybe it will just be a lot of wasted effort.  On the other hand, I could just say I see all of the obstacles there, but, boy, this would be wonderful if we could do it, and I'd like to, you know, take my shot at it, or is there some other reason?

In other words, what do we conclude from this depiction of human nature about the project of human enhancement?  Is there any reason not to try it from what you've told us?

DR. PINKER: Certainly there's no reason not to have a better understanding of the genetics of personality and intellect and the process of neural development.  I consider that to be possibly the great frontier of science in the 20th century. 

How a one dimensional genome results in an organ like the brain with the ability to see and think and feel and plan has got to be the most exciting and the most challenging scientific question facing us, perhaps the most exciting scientific question of all time.

We want to learn more about it.  There will be practical applications above and beyond enhancement.  For example, if we knew the genetic basis of disorders like schizophrenia, we would know more about the actual molecular pathway from gene to brain to behavior, offering the possibility of non-genetic enhancement, such as drugs that could interfere with the process that leads to schizophrenia.

Also, to answer the intellectual puzzle of what makes us what we are.  The more detail in which we know it, I think the more enriched we will be as a scientific community and as a species.

In terms of actual enhancement, I think the main ethical impediment is going to be the possibility of harm to the unborn child.  For as long as that is a considerable possibility, as long as the chances are well above zero that a child could be harmed by genetic enhancement, I think most other questions will remain moot.  My hunch is that that's going to be the biggest impediment to getting there from here.

PROF. MEILAENDER:  But assume that the harm issue were somehow put aside, just for the sake of argument.  Would you then have any hesitation about that kind of endeavor?

DR. PINKER: If it  way, say, one of the you can't be too rich or too thin traits like IQ; I mean, if there were a magic gene that was guaranteed to have no side effects, that could make children smarter, then I would say it's an extreme hypothetical.  I would say if that existed, then I would not have any problems with it, but I want to make it clear that that ethical sentiment of mine is separate from the factual arguments that I've been making so far.

I have not heard any good arguments, arguments that I consider sound, that this would be a bad thing if we would ever reach that state.

CHAIRMAN KASS:  Bill Hurlbut, Rebecca.

DR. HURLBUT:  So I welcome your statements about the difficulty of genetic engineering, but what I want to ask you is assume that genomics will give us a great deal of information about the construction of the organism and, therefore, a lot of power to understand proteomics and, therefore, to intervene with designer drugs at various states.  Is it your sense that human nature may be amenable to some kind of improvement by alteration through pharmaceutical agents which would not be as dangerous?

And specifically, I want to ask you:  do you think that we might make moral improvements?  And a corollary question is:  do you think that there are genetic and, therefore, biochemical differences in human moral nature?

DR. PINKER: Okay.  Several questions.  It might be possible to have pharmaceutical interventions that have a consistent beneficial effect, although, again, there I would urge people to have a skeptical eye on such claims.

We now know that the effects of Prozac, for example, are real, but were certainly over- hyped from the way they were portrayed ten or 12 years ago.  Just to give one example of how most things will have costs as well as benefits, Prozac in many cases diminishes libido.  So should we put it in the drinking water?  Would people take it on, you know, a prophylactic basis to feel better about themselves if they knew it would nullify their sex drive?

DR. HURLBUT:  Is that a moral improvement, by the way?

(Laughter.)

DR. PINKER: Morally, the question of whether we should eliminate all of the rough spots and pain of the human condition, the depression, the anxiety and so on, I'll give you an analogy of physical pain.

There is a syndrome studied by one of my undergraduate teachers, Ronald Melzack, in which some people are born without the ability to feel pain, and first you might think, "Wow, what a great thing.  You know, you'd stub your toe and you'd walk away without, you know, swearing and feeling the agony and so on."

In fact, this is a bad thing.  The people with that syndrome generally die in their early 20s.  The reason is that they don't have the feedback signals that tell them when they're damaging their body, and they suffer from massive inflammation of the joints simply from not shifting their weight when it gets uncomfortable, something that's second nature to the rest of us that feel pain.

That is going to be true of many of the negative psychological emotions that we feel.  The ability to feel sad is the other side of the coin of the ability to feel love and commitment.  If you didn't feel sad when you child died, could you have really loved your child?  If you can't feel anxious, I'm sure I don't have to remind anyone in this room that anxiety gets us to do many things that otherwise we would not have done.

On the other hand, getting back to the touchstone of pain, it's also not the case that if you have a toothache you should stay off the aspirin because pain is a good thing.

Pain, like negative psychological emotions is a mechanism that has a function.  On the other hand, it's in many cases a clumsy, over- reactive mechanism, and once we recognize what these negative emotions ought to be doing in order for us to lead better lives, there's no reason, I think, for people to suffer simply because on average in the species, the mechanism is there for a purpose.

So I don't think there would be a sound argument for preventing people who are depressed or anxious or irritable or hyperactive from doing something that would lead to an increase in their well-being simply because it's unnatural or because the mechanism had a function, as long as we realize that reducing these negative emotions to zero, as with reducing pain to zero, would not be a good thing either.

DR. HURLBUT:  Can I follow up on that?  If there are values to pain, probably there are differences in pain thresholds between individuals.  Now, translate that into moral instincts, moral awareness, moral sentiments.  Is it in your thought possible that not only do human beings individually vary one to one, but the different small environments of evolutionary adaptation that have produced externally evident morphological differences between human groups' geographic origins might also correlate with differences in moral understanding?

DR. PINKER: Well, let me first answer a slightly different question where I think we know more, and that is differences among individuals within a racial group.  That is, you take two Caucasians.  There is good reason to believe that some moral traits have a partly heritable basis.  There's good reason to believe, for example, that psychopathy, which comprises callousness to people and inability to empathize, has a partial genetic basis like all psychological traits.  It's only statistical, not absolute.

So the answer to the question of could there be variation in moral sentiments, I think the answer is very likely that there is among individuals within an ethnic or racial group.

Whether ethnic or racial groups on average differ in moral sentiments is, I don't have to remind you, a politically fraught question.  I would say at present there's no reason to believe that such differences exist.  It doesn't mean that they can't exist in principle.  It means there are no data at present that would lead one to conclude that they exist, and it's a separate question from whether individuals within a group differ.

We know just from genetic variation that there are far more genetic differences between two individuals within an ethnic group than there are between the average of one ethnic group and the average of another ethnic group by a very large factor, a factor of at least ten.

CHAIRMAN KASS:  Rebecca Dresser.

PROF. DRESSER:  Two questions.  I wonder if you have the same skepticism about the ability to enhance physical characteristics in, you know, embryos.

And the other, do you think there's something in human nature that makes people want to change human nature?

I get very frustrated with this hype because I am somewhat skeptical, but why do the newspapers carry all of this?  And why is there this ongoing fascination?

DR. PINKER: Yes, yes.  Well, for physical enhancement it will, I think -- it won't be as easy, again, as many of the pronouncements in the press would lead you to believe.  Remember during the energy crisis in the 1970s you'd often see the ad for the 200 mile per gallon carburetor where you just unscrew your old carburetor, put in the new carburetor, and you would go from 20 miles a gallon to 200 miles per gallon.

Now, there's reason to be skeptical of that invention simply because a car engine is such a complex system, and there's such incentives to making it better that if that were physically possible it would have been thought of a long time ago.

Likewise with the human body, natural selection tends towards optima.  We know that there are tradeoffs in the design of the human body, a simple example being the fact that males are on average physically stronger and faster than females, but also die younger, and those are probably related, namely, that there are different points along a tradeoff.

I suspect there is a possibility, having said that, of many genes with very small effects that conceivably could add up to improvement simply because we know that there are differences among individuals.

I think that it's much more likely for simple one dimensional traits like height or muscle mass than for a complex system, such as functioning of the heart, which probably depend on combinations of hundreds or thousands of genes as opposed to something like height, which is a one dimensional trait that could be under the control of a small number of them.

In terms of the second question, I guess I don't know enough intellectual history to know whether this is really a feature of human nature or whether it's a sort of post enlightenment, Western concept and whether fatalism, you know, there's nothing new under the sun; empires rise, empires fall; time is a cycle, and so on, which of these is more dominant in the history of human thought.

It certainly is a feature of our culture to believe that we can change anything we don't like.

CHAIRMAN KASS:  Mike Gazzaniga.

DR. GAZZANIGA:  Steve, sometimes there's the feeling that ethicists are chasing rainbows that are  generated by the popular press, is one way of saying what you've been saying.  If were to just ask you freshly, with you view of the nature of human nature and with the technological advances in neuroscience and biology that are now occurring, what would you see as the great ethical questions of the next 20 or 30 years?

DR. PINKER: I would say that one of them was just raised, namely, as we know more about effects of genes on personality and behavior, I think we will have the possibility to answer questions, such as on average do different ethnic groups differ in distributions of genes that have effects on psychology.  Should our attitude be don't go there because no good can come from studying these differences?

The reaction to the book, The Bell Curve, that came out ten years ago would suggest that by and large we're not ready for such discoveries.

On the other hand, is more knowledge always a good thing?  And could it be inevitable that such discoveries will arise as a byproduct of ethnic and racial differences in medical treatment?

For example, if there are average racial differences in the effects of or abundance of testosterone, a fact that we may need to know in order to study the demographics and treatment of prostate cancer, for example, well, testosterone also has an effect on behavior.  What will we do with the discovery of differences if such discoveries are in the cards?

I consider that to be a potentially inflammatory area of research, certainly ethically fraught, and I can't say that I'm certain where I stand on that issue.

I think drugs such as Ritalin, which would be given to certain segments of the population but not others, will certainly raise issues of equity, who has access to them, and the flip side of that, the allaying fears that they will be used as a method of social control of sapping boyhood, of sedating disaffected  inner city youth, all of the issues that have come up with connection with Ritalin may come up with other drugs.

The question of moral responsibility in the criminal justice system, in general people who commit heinous crimes must have something different in their brains from people who wouldn't.  Otherwise they wouldn't have committed those crimes.

We're going to be better and better able to discover them whether there are differences in genes or differences in cerebral metabolism or brain anatomy.  I think we'll need to have very clear guidelines for insanity defenses, diminished capacity, and that whole suite of legal issues as we reach the point where for a large percentage of malefactors we'll be able to say this is what's different about them compared to you and me.

So those would be three.

CHAIRMAN KASS:  Could I follow up on this because I was also in the queue?

I mean, the subject on which you were invited to speak is massive, and you've chosen in the formal part of the presentation to speak about human nature vis-a-vis possibilities for genetic alteration of it, and let's set that aside.  I, for one, don't have any reason to dissent from the presentation.

But I guess three things.  One has to do with the question of human self-understanding through the progress of, on the one hand, genetics and, on the other hand, neuroscience.

In this last remark about moral responsibility and culpability, would you speculate on how an increasingly biologically based account of who we are, whether it be in terms of genes or be in terms of brains, is going to affect how human beings understand who and what they are, that is to say what their human nature is?

I think this is partly not unrelated to where Alfonso was going earlier and where you yourself have also, I think, written.

Let me leave it at that.  I've got a couple more, but let's start with that one.

DR. PINKER: Yeah.

CHAIRMAN KASS:  I mean simply on the question of freedom and responsibility or the character and object of desire, things of that sort.

DR. PINKER: Yeah.  There certainly will be changes.  The idea of humans as possessing some immaterial essence that categorically distinguishes them from animals, I think, is going to come under -- is going to become less and less credible, and there will be, I think, a crisis among the religious faiths that depend critically on the assumption that there is some nonmaterial essence.

I mean, this is intellectual development that certainly began, well, probably began hundreds of years ago, but was acute, for example, in the writings of Dostoyevsky and other 19th Century authors.

I think there's going to be a rethinking of ethical issues, such as responsibility and justice and equality, not that it will evaporate, not that Nietzschean fear that we'll have a total eclipse of all human values once people realize that the human mind is a product of the brain, which in turn is shaped by genetics.  It's not that our values will go out the window.

On the contrary, I think they will focus our ethical discussions on what we most value, what we want moral guidelines to do.  Let me be concrete because I was very abstract.

In the case of moral responsibility, there is the ancient antimony between free will and determinism that has kept philosophers employed for millennia.  It keeps college students debating until the wee hours of the morning in their dorm rooms.

I think there's actually a more useful and practical way of couching that issue, namely, once we find that the mass murderer has a defect, we find a red pixel in his brain, should we get him off the hook?

The practical question is:  what are the effects going to be of our policies for holding people responsible?  Holding people responsible is basically a long-term deterrence policy.  If you hold people responsible, that in itself is an environmental cause of behavior that we hope and, indeed, diminishes the probability of harmful behavior occurring.

If someone thinks that they will be thrown in jail for holding up the liquor store, they'll be less likely to hold up the liquor store.

The question is:  in adopting policies of that sort, which of those policies will have the predictable effect of reducing harmful behavior without causing unnecessary, spiteful punishment of people who could not have been deterred to start with?

The reason we don't throw five year olds in jail is that we think that a policy of throwing five year olds in jail will have no effect on the future behavior of five year olds.  That's also why we don't punish animals or put them in jail or try to shame them.  It would be futile to expect that that will lead to a change in behavior.

Whereas for the vast majority of adults, saying that we will hold you responsible we expect will decrease the probability of harmful behavior.

Most questions on insanity defense, diminished capacity, and so on, I think, are more fruitfully reconceptualized not in terms of the metaphysical concept of free will, namely, was the behavior caused or not in some metaphysical sense, which is probably unanswerable, but rather what are going to be the effects of those policies.

If we had a schizophrenic with a certain brain condition, would not have been deterred from committing harmful act regardless of the punishment that we put into effect, then subjecting him to criminal punishment would simply be inflicting harm without satisfying the goal of reducing harmful behavior.

I think that's an example of how a pressing ethical issue will be reconceptualized by realizing that behavior is caused by the brain rather than it simply being eliminated as some people fear.

PROF. MEILAENDER:  The most effective way of stopping certain behavior would be periodically to frame certain people for having done it and punish them publicly, if we could somehow satisfy ourselves that that would be the most effective way of stopping it, would that be the right thing to do?

DR. PINKER: No, because I think the --

PROF. MEILAENDER:  But then you think that issues of dessert somehow enter in?

DR. PINKER: Yes.

PROF. MEILAENDER:  And responsibility?

DR. PINKER: Yes.  I think that the policies in the criminal justice system trade off between having a deterrent structure that reduces harmful behavior while causing the least amount of preventable harm or suffering.

PROF. MEILAENDER:  But while also punishing only people who are somehow responsible and guilty?

DR. PINKER: Yes.  And I actually think that the concept of dessert, in addition to the concept of deterrence, they're not -- I don't think they're completely independent because if you probe, if you try to dissect our intuitions about just desserts, they very often, in fact, perhaps even always, act as a kind of long term deterrent policy aimed, I think, at preventing people from gaining the system by acting in just the way that would allow them to escape the net of criminal punishment.

Let me be concrete.  Why do we track down elderly Nazis in Paraguay even though the chances of them perpetrating another Holocaust is zero?  There's no deterrent effect of that policy.  Nonetheless, most of us believe that this is the right thing to do, that it's inherently unjust to let them die in their beds without facing justice.

Well, in part, it's that even if it has no deterrent, specific deterrent effect on that individual, it would have a general deterrent effect in that future perpetrators of atrocities would have to think twice if there was such an implacable desire for justice, for hunting down malefactors; that even if it wouldn't be worth the while of a society to track them down for that particular case, the concept of just desserts would force potential malefactors from thinking twice knowing that there is this desire on the part of society at large to track them down.

And the concept of just desserts, even though there are thought experiments that one could come up with that would pit it against deterrence in specific cases, I think, has the effect of implementing a cheater proof policy of deterrence in general over the long run.

PROF. MEILAENDER:  Once more, but our motive for tracking them down is in order to deter  the future evil doers.  There's a kind of a gap that grows up between this implacable desire for justice that we want some people to believe is important and the motivate that spurs us to track them down, which is not an implacable desire for justice and the thought that we should catch them and punish them if we can, but rather simply that if we don't do this, future generations will not be deterred from similar horrific acts.

Do I have you right?

If so, I think there's a real theoretical problem.

DR. PINKER: I would add the proviso that there's a bit of a paradox here, but the fact that we have this almost irreducible, implacable desire for justice itself serves over the long run as a deterrent, namely, if we have an implacable desire to bring people to justice no matter how much it costs, no matter how trivial the gains in deterrence, that itself makes the credibility of the implicit deterrent that much stronger.

So there is an autonomous, I think, moral and psychological imperative to see justice done.  I don't believe that people literally calculate the deterrent value of pursuing justice, but paradoxically it is that irreducible desire for justice that over the long run makes it effective for the same reason that someone who issues any kind of threat is that much more credible if he has implacable, rational reasons for carrying out the threat.  That makes it much harder to call his bluff.

And a society or a criminal justice system with the concept of just desserts is harder to -- it's harder to call its bluff or to game the system.

PROF. MEILAENDER:  Of course, I don't doubt that they're connected in that way, that a system in which justice is rendered will have deterrent effects.  The question is:  what's first order and second order here?

The question is:  what are motivations for tracking down that Nazi in Argentina is?

And I would have thought that our motivation for tracking him down is, our first order of motivation, is that he's an evildoer and justice requires that we punish him if we can.

Our first order of motivation is not that we should track him down so that future generations may be deterred.

DR. PINKER: I think the first makes the second more likely, that is, it's the very autonomy of our intuitions of justice that also make it effective as a deterrent.

So I think I agree that they're conceptually separate, but I think that they are also  ultimately related.

DR. KRAUTHAMMER:  But are they fictional?

PROF. SANDEL:  They're an illusion, and in fact, if they heard your analysis, they wouldn't be so implacable.  They'd just be free riders and justice could fall pray to a collective action problem.

They think they're implacable because they're under the illusion that they're trying to get this old Nazi because he ought to be punished because he deserves punishment.

And you then describe, well, that's actually a functional illusion that gives the implacable character to their pursuit and that has these desirable things in the long run.

But if they heard and grasped the truth that you're offering, then they would be rid of their illusion.  The guilty deserve punishment, and therefore, they should go running after him, right?

DR. PINKER: Well, not necessarily because the fact that I can explain, say, the ultimate long-term, perhaps even evolutionary rationale for a deep seated intuition, namely, bad people must be punished, doesn't mean that that sentiment is any easier for me to give up.

It may be that here's the reason why.

PROF. SANDEL:  But on reflection it should be given up.

DR. PINKER: Well, even if so --

PROF. SANDEL:  But shouldn't it?  I want to know what you think.  Never mind about the whole evolutionary thing.

DR. PINKER: Right.  On reflection, there are a lot of things that should be given up that we won't give up because of the way we're built, and I think the desire for justice, even if I can tell you why my brain has this concept of just desserts and I can say that in another planet, another evolutionary history my brain may not have had that intuition, the fact is it does have that intuition, and that intuition is --

PROF. SANDEL:  But that intuition is so biologically brute that even listening to your deconstruction of it won't disabuse me of it?

DR. PINKER: Yes, I think that's right.  I'll give you an example.  In many --

CHAIRMAN KASS:  Also, the existence of your immaterial soul, Michael.  Don't worry.

DR. PINKER: Well, among the justifications for criminal punishment, one -- of course, opinions vary.  Some people do disabuse themselves of intuitions whose rationale are then laid bare, but my understanding is that many judicial theorists say that some degree of retribution is a legitimate function of criminal punishment; that we allow more and more victims of crimes have a say, and that the widespread intuition that somehow the universe is knocked out of balance if evil doers are not punished above and beyond the practical effects is often recognized as a legitimate function simply because we think that it is part of human nature, that people will be enraged, will seek private vengeance unless society satisfies this desire.

So I don't think it's that easy to eradicate, and it's probably a good thing that it isn't that easy to eradicate, although it is, I think, a good thing to be aware of it simply because we can then -- to get back to the question that Professor Meilaender raised, if it was only deterrence that we wanted criminal justice to accomplish, then we would do things like frame a few people just to keep everyone else on their toes or many other things that we would find horrific.

Realizing what the goals are of a system of deterrence, we can calibrate the desire for deterrence against the other desire not to inflict unnecessary harm, and realize what we're doing when we impose these policies.  At least that's what I would argue.

CHAIRMAN KASS:  Dan Foster.

DR. FOSTER:  You know, I thought earlier that you really were holding to the conclusion that, you know, once we knew the genes and everything would follow.   We'd get to something.

But it sound a lot to me like your deep intuition is very much like what somebody else might call an essence, that this desire for justice, let's say, that's so deeply intuitive in you, that you are not sure where it ever came from.

And I think that maybe some people might be skeptical about the view that there's really no fundamental difference in the nature of an animal or of a human, since I don't know what the exact difference is between the genes in a chimp and the humans.  Maybe five percent.  It's pretty low though as far as we can tell.  I mean, it's pretty low.

But one doesn't, I think, normally sense that a chimpanzee or certainly lower species have an intuitive drive for justice.  I mean, G.K. Chesterton used to always say that there were certain fundamental differences between animals and humans, that the human animal, for example, creates.  She makes art and music, not nests to live in, and no other animal species as far as we know does that.

He said the human animal differs from all other animals in another respect, that they have a deep, intrinsic sense of guilt, that we know intuitively the difference between right and wrong.

Now, one would argue, you know, that my dog knows the difference between right and wrong because I've got a Pavlovian response, you know, to him if he doesn't behave.  You know, the human intrinsically knows  the sense of joy, in a way.  Animals don't laugh as far as we know.

And finally -- I mean,  the hyena might -- but finally, he said that the human animal is very peculiar in that from the beginning, even from the earliest graves that we see, somehow worships, has a sense that there's something that might be intrinsically different from what genes do.

I mean, it seems like to me that you've shifted a little bit from your earlier sort of a genetic determinism, if that's not an old fashioned word, to a description of yourself and others that says that there is something different about the essence of this particular species.

I don't know whether that's -- but just listening to you.

I have a friend at my medical school who won a Nobel Prize in medicine and who happens to be Jewish, and he usually argues very intensely down to the quarks about determinism.  And so I didn't say, "Well, it's too bad that Hitler got those bad quarks and genes and so forth and he ought to go free," because it was all determined along those lines.

And then immediately he usually sort of backs off.  "Well, I didn't really mean it that way."

But it sounded to me, and I'd like for you just to comment about that, just to focus on the justice issue here, that you were describing when Michael was pressing you about yourself and not the literature and so forth, that you were saying -- I think I heard you say that there's something deeply intuitive in me that makes me drive for justice, and I want to know what the difference is between your genes and the chimp's genes that give you an intrinsic drive for justice that maybe the chimp has, but I don't know.  I mean, I don't know, but at least the studies haven't shown that so far.

DR. PINKER: Yeah.  Well, the genetic differences between humans and chimps are small as a proportion of the genome calculated on, you know, a base pair by base pair basis, but because DNA is basically a computational system, small differences in the sheer information content can make a big difference in the final product.

So just an analogy, if you were to take a text file on your computer and change one bit in every byte, the result wouldn't be 12 percent different.  The result would be 100 percent different because a single change can result in a protein product that has a radically different effect.

So even though genomically we're very similar to chimps and in terms of the phenotype there are more similarities perhaps than we'd want, including by the way laughter; chimps definitely do laugh, but this is not to deny that there aren't significant differences between humans and chimps, just as there are significant differences in any pair of species.

So the analogy that I often use is that an elephant has a trunk which is as far as we know unique among animals.  It's the only animal that has a trunk.

Humans also have a number of unique traits:  language, that is, grammatical combinatorial language; probably moral sentiments, such as guilt, shame, trust.  One can debate whether there are rudiments of them in chimps, but there's no denying that what you find in humans is very, very different from what you see in any other primate.

I don't consider this to be an evolutionary paradox for the same reason that I don't think that the fact that the elephant has a trunk and its relatives don't is a paradox, namely, that evolution creates divergence.  It can lead to the development of traits, including mental traits that are found in one species but not its relatives, and I think there are quite intelligible reasons for thinking that in the case of the evolution of homo sapiens things like language and the moral sentiments and technological know-how, such as tool making, developed in the last six to eight million years in a much greater extent in our branch of the family tree than in chimps.

And I think that the sense of some kind of primitive sense of justice, of just desserts, might be something that really is universal in humans and probably absent or rudimentary in chimpanzees, as you said.

DR. FOSTER:  Thank you very much. 

We'll be talking all night if we continue this.  So I yield in response.

CHAIRMAN KASS:  He's got the best subject in the world.

DR. MAY:  It may be the same point that Dan was interested in, but it seems to me finally you have to say that Eichmann is a victim for the larger social good served by punishment even though biologically considered he doesn't deserve it.  In that sense we have framed him, but for a good purpose.

DR. PINKER: Well, there's a lot of debate on the specifics of the Eichmann case, namely, whether he really was the faceless bureaucrat that he conveniently portrayed himself to be.

PROF. SANDEL:  That would make no difference to this question.  That would be irrelevant to this question

DR. PINKER: Well, the way I would try to make that question tractable is if we have consistent policy for what to do with Eichmanns and we held to it steadfastly and it was announced beforehand, what would be the effect on future Eichmanns?

That's not identical to the question that you raised, but I think it's more tractable than the question that you raised, and I think my hunch is that if you actually worked it out, you would end up in a very similar position as the one that you would arrive at if you reasoned in terms of the raw intuition of just desserts.

I think that the intuition of just desserts, no matter how passionately held, can be examined.  We can say, "Well, let's lay out the thought experiments.  What would you think if?"

And my sense  is that it would be a -- in the cases that I've thought through, such as insanity defense, punishment of animals and children and the brain damaged and so on, you end up with very similar answers to the one of what's the best long-term general deterrence policy balanced against the moral harm of inflicting suffering that has no beneficial --

PROF. SANDEL:  Does that mean you agree with Bill's premise that on your account, every instance of punishment is a case of framing?

DR. PINKER: No, I don't.  No.

PROF. SANDEL:  Some are and some aren't?

DR. PINKER: Well, in an ideal system none of them would be in the sense that we would not inflict punishment on someone that we had excellent reason to believe did not commit the act and hence could not have been deterred by such a policy in the future, namely, innocent people who are frame.

There are other people in those shoes out there, and a policy that would net them in isn't going to prevent them from doing evil because they didn't do evil and they never wanted to do evil, and so that would be a moral harm inflicting unnecessary suffering.

So that's why we --

PROF. SANDEL:  That would all depend on the perceptions and beliefs of the onlookers.  It would have nothing to do with the guilt or innocence of the person.

CHAIRMAN KASS:  But, Michael, you --

DR. PINKER: I'm not sure I follow that because if the person didn't commit the harmful act and would have no impact --

PROF. SANDEL:  But everybody else believed they did, then for long-term deterrence reasons you do want to punish that person.

DR. PINKER: Well, if long-term deterrence was your only goal in the criminal justice system.  If you had two goals, namely, long-term deterrence and prevention of unnecessary suffering, then you have the tradeoff of under what conditions will you tolerate unnecessary suffering in order to get an increment in deterrence.

And I believe that in the case of deliberately framing an innocent suspect we would say that the first outweighs the second.

PROF. SANDEL:  It depends what the consequences would be.  There's the famous hypothetical of a heinous crime that's been committed.  The entire town is outraged.  They're about to go on a rampage against the neighborhood from where they think the criminal came.  The sheriff doesn't know who committed the crime, but to prevent this terrible cost and rampage goes and takes the town drunk from jail, announces that he's the criminal and hangs him.

DR. PINKER: Yes.  That would be the utilitarian calculus of preventing the -- and your question is why?

I mean, I assume that the assumption is that we all consider that to be a bad thing to do, and the question is why.  Is that what you're asking?

PROF. SANDEL:  Yeah.

DR. PINKER: It's a good question.  I would concede that it's the weak point of this analysis, and I haven't thought that case through in enough detail to answer you, but I suspect it would be state that as a policy, and that is not just what does the sheriff do on the spur of the moment, but you write it down.  What should sheriffs do in general?  What should law enforcement officers do in general?

Look at the policy and see how much unnecessary suffering does it cause as opposed to the alternatives.  I'm not going to bluff and say that it will come out the same way, but I have a hunch that it would.  But granted it's a valid objection.

CHAIRMAN KASS:  Paul McHugh.

DR. McHUGH:  I enjoyed your talk very much, and the theme of your talk was that things were more complicated than we thought and interfering with things at the genetic level and at the neuronal level was going to be less likely to produce a better world and a better person than we imagined.

This same kind of thing is turning up here.  It seems to me when I hear you discussing with Michael and Dan and others, you seem to function on the idea that things work in an A to B position.  We do things because we know the consequences, and that you and our understanding of human nature might be able to recognize those consequences.

But it seems to leave out what Charles Saunders Peirce used to talk about thirdness, what the symbol of all of this is, and that human beings have a capability, therefore, of seeing something and having it have a deep and penetrating meaning to them that they ultimately make more and more explicit as they become more and more developed.

And so, for example, in this area of punishment, you talk about deterrence and you talk about retribution.  I've always thought about it in reprobative  terms rather than retributive terms, that is, this act we cannot tolerate amongst human beings, not as to whether there's going to be any more of them or not, but really because what it means.

And we see that punishment work in a good way and a bad way in relationship to notorious cases.

I feel that the execution of Eichmann was very just, very true, and if I could have been one of the guys tracking him down, I would have loved it.  I would carry it as a badge.

But, you know, the execution of the Rosenbergs was thought in some way to perhaps be going to deter.  It didn't deter anybody, and in fact, we now know that, in fact, there was a very cruel and vicious element to putting the Rosenbergs to that test, that the government probably thought that they would crack them rather than have them die.

So those two things are events in which punishment was delivered.  One of them -- both of them might have been thought of as having rebributive elements, but reprobatively they are miles apart.

So my question to you fundamentally is this.  Do you not give to human nature the capacity to move from their implicit things, to make things progressively more explicit, and in the process progress not simply in the direction of being more effective, but also to being more good?

DR. PINKER: Oh, absolutely, and if I've tried in my own writing to change the common conception of human nature, it would be to emphasize that one of the features of human nature is a combinatorial apparatus that can generate new combinations of ideas, and again, I'll use language as my touchstone.

We're equipped not with a finite list of sentences that like, you know, a Sesame Street doll where you punch a button and one of a dozen sentences is selected at random and comes out verbatim.  What we're equipped with is a set of grammatical rules that are assembled, nouns and verbs in new combinations that allow us to express new thoughts.

Similarly, in thought, which obviously feeds language, we have the ability to multiply a fixed stock of ideas to come up with unlimited combinations.  We can have theories of the origin of the universe.  We can have new political theories.  That's why we're in the business that we're in of exploring ideas and making discoveries.

So they bear the stamp, I would say, of particularly human ways of thinking.  We conceive of things in particular ways that a Martian might not.

On the other hand, that doesn't mean that there's a finite number of human-to- think- about thoughts or that we're doomed by our neurology to recycle the same ideas over and over again for the same reason we're not doomed to regurgitate the same sentences over and over again.

DR. McHUGH:  So you would, therefore, agree that there might be determinism, but you don't believe in fatalism.  Would that be a useful way of describing yourself?

DR. PINKER: I don't believe in determinism.  I mean, determinism is a word that has many meanings, and it's often used more as an epithet, I think, than to -- I mean, I don't know of anyone who claims to be a determinist.  I know a lot of people who are called determinists.

If you use determinism in the mathematician sense of an event happening with Probability 1, then I am absolutely not a determinist and not for any philosophical reasons, but for an empirical reason.

The identical twins raised together correlate only, say, .4 to .8.  That technically refutes determinism in its actual sense, but I certainly do believe that the genome leads humans to think and feel in characteristic ways, but because the brain is so complex, because it has multiple systems and a number of them have the ability to crank out new combinations, infinite combinations of ideas, the idea of a fixed human nature doesn't mean that there's a fixed repertoire of behavior or thoughts.

And I think we have the ability to, well, I'd like to think and I think we do have the ability to learn the lessons of history, to be persuaded by argumentation to see things in new ways, and in fact, again, this isn't just kind of a sappy sentiment, but things have changed which would be impossible if we were genetically fixed.

The rates of violence have gone down in the last couple of hundred years in the society.  Concepts that were thought to be inevitable, such as slavery, subjugation of women, inevitability of blood feuds, for example, all have greatly diminished.

So the notion of human nature doesn't mean that society will never change or ideas will never change.

DR. KRAUTHAMMER:  Can I come in here a second?

CHAIRMAN KASS:  Go ahead, Charles.

DR. KRAUTHAMMER:  But what I hear is a critique of determinism on the basis that the old determinism was very one to one and very unsophisticated.  It's interesting we are not discussing here bioethics but the rather interesting world of evolutionary psychology, and I think that's what some of us are rather seized by in this discussion and provoked by.

You said a little earlier that you thought one of the challenges or one of the developments in the future is going to be a decrease in what I believe you called essentialism, a self-perception by human of their exceptional nature.

If you take away that essentialism, then you end up with either the theory of justice, for example, that is based entirely on its evolutionary advantages, meaning deterring bad effects and bad behavior, and leaves no space for anything else other than illusions.

So my question is that if you believe in evolutionary psychology, biology as an explanation for our current human nature and you believe that it really excludes any essentialism, which is some sort of archaic, perhaps superstitious notion about human nature, then what's left?

And what I hear is what's left is a notion of criminal justice, for example, that leaves no room whatsoever for the notion of real guilt, of real agency, and that's what I think we find rather shocking.

Am I correctly explaining your understanding of nonessentialism?

DR. PINKER: Yes and no.   This is what I call the fear of nihilism, that a materialist, Darwinist view of human mind will expose all of our values to be in some sense shams, that they are just means to the end of some practical function, like propagating genes, deterring violence and so on.

DR. KRAUTHAMMER:  Well, you gave us a pretty good example of that with criminal justice.

DR. PINKER: Yes.  Well, let me give two answers to that.  One of them is this is what Daniel Dennett calls the idea of Darwinism, is universal asset.  It just eats through any possible container and dissolves everything we hold precious.

First of all, the fact that we can understand our own sentiments doesn't mean that they're shams simply because they are our sentiments.  So let me give you an example.

There's reason to believe that our aesthetic judgments are evolutionary adaptations.  We like particular landscapes, particular faces because there are rational reasons why we should have evolved in that way.

Does that mean that nothing is beautiful and that there's no point in looking at attractive landscapes or faces?  Well, no, because that is the way we're put together.  The fact that we understand why we're put together that way doesn't mean that here in our own skin those sentiments are any less real to us.

If cosmically you can say there's nothing particularly beautiful about the Rocky Mountains as opposed to a New Jersey oil refinery, I don't know how to answer that question, and I don't really care about that question.  The mere fact that I am wired together, wired to like the Rocky Mountains better is good enough reason to indulge that.

Also, there are some cases in which I think we can actually step outside our skin and at least entertain the possibility that some of our perceptions and values do pick up on an external abstract reality.  Again, I'll be concrete and I'll give an example.

There's good reason to think that our sense of number is an evolutionary adaptation, that there's good reason for an organism to be able to tell the difference between one and two and three and have the elementary concept of addition, but it doesn't mean that one and one equals two is a hallucination or a fiction.

It's in the nature of reality that any organism that can grasp the concept of number is forced to come up with certain conclusions.

Likewise there are cases in ethics, at least so some moral philosophers argue, and I'm not prepared to disagree, where there is a reality to some moral judgments, and the fact that our moral sense may be an evolutionary adaptation of the brain doesn't mean that the things that it thinks about are figments.

DR. KRAUTHAMMER:  All right.  Let me test you on that.  If you say evolution urges or -- I'm sorry.

CHAIRMAN KASS:  No, go ahead.  One last round because we're well over.

DR. KRAUTHAMMER:  I'm sorry.

It's been advantageous for us to understand that one and one is two, and that is giving us an intuition into a platonic universe in which one and one, in fact, are two, and we understand that.

So by analogy, evolution has given us the intuition that we ought to hunt down a Nazi.  Does that give us any intuition or does it not give us the intuition that there's a platonic universe in which that Nazi is evil and ought to be hanged regardless?

DR. PINKER: Well, let me put it maybe.  Let me put it that way.  It's not a question that I can answer that has, you know, challenged the best minds for millennia.

The way I would put it is there are certain core moral intuitions that I think could be argued to have a basis in reality, such as the fact that no particular person can argue that he occupies a privileged position in the universe whose well-being can trump the well-being of anyone else simply because that's a logically untenable argument as soon as one enters into rational discourse at all.

For the same reason that I can't say that this spot in the universe is privileged because I happen to be occupying it, I can't say that my interests are privileged over yours as long as I'm willing to enter the discussion at all.

That isn't an arbitrary figment, but it's in the nature of that kind of discourse in the same way that one and one equals two is a necessary consequence of thinking mathematically to begin with, which is a reason that I think different moral traditions end up with some notion of reciprocity or golden rule or categorical imperative over and over again.  It's a kind of forced move in perhaps a platonic nature of relationships among ideas from which one perhaps could deduce that Nazi war criminals ought to be hanged from a chain of intermediate steps.  Given that I don't want to be the victim of genocide, how can I tolerate it if other people are the victims?

Then if I want to reduce it over the long run, given the presence of other agents trying to gain the system, what is the most effective way of universalizing my own desires, and so on?

I could imagine a chain of steps that would lead to that as a theorem from axioms that might have some kind of universal warrant.  So I mean, that's the best that "little me" could do in grappling with these cosmic questions.

I don't have an answer to them.  I think that we can make progress in scrutinizing them and not prematurely satisfying ourselves that some intuition simply ought to be accepted as the nature of things without penetrating that intuition and asking why might we have it.

I think there's only good that can come from scrutinizing those intuitions as opposed to taking them as givens, and the reason that I feel emboldened to say that is that we know that people can have absolute certainty in certain intuitions, which upon reflection they can be argued out of or externally we recognize to be horrendous, such as slavery, ownership of women, other things that seem self-evident in past centuries, but where I think as these things get scrutinized they are revealed to be inconsistent with other beliefs or untenable.

Because we know that that kind of moral progress can take place when intuitions are scrutinized, it is important to scrutinize our own intuitions.  It doesn't mean that we will end up in a state of nihilism where all morality is a fiction, but I think and hope and would argue that it would lead to a case where our ethical system is more human, more effective, and more defensible.

CHAIRMAN KASS:  There are lots of people at this table who would be eager to continue this for hours.  We're already 15 minutes over, and since I was next in line with a long list, I will squelch myself, express my thanks to you for a very interesting and forthcoming and provocative conversation.

We'll have a 15 minute break in which the people who want to sort out the question of whether the differences between Dr. Pinker and the rest of us is owing to the fact that we're just differently wired or he actually has discovered the immaterial truth on this subject.

But we'll take a break, and let's make it a little shorter.  Five minutes to four so that we won't have to finish too late.

Thank you very, very much.

(Applause.)

(Whereupon, the foregoing matter went off the record at 3:47 p.m. and went back on the record at 4:11 p.m.)