course, there will be people who object. There will be people who will
say that this is a revival of racial science. Perhaps so. I would argue,
however, that even if this is a revival of racial science, we
should engage in it for it does not follow that it is a revival of racist
science. Indeed, I would argue, that it is just the opposite.
NATURE OF NORMAL HUMAN VARIETY [3.15.05]
A Talk with Armand Leroi
REALITY CLUB: James J. ODonnell, Andrew Brown,
Tim D. White, Alun Anderson, Nicholas Humphrey respond to Armand Leroi
the early '90s, I was visiting Cambridge and went out to dinner with
the late Stephen
Jay Gould. During a long evening of conversation we talked about his
race, racial racial differences, racial equality, including his well-known
writings on the use and misuse of IQ tests and other such measures.
came away from the conversation with the distinct sense that he believed
there were some things better left unsaid, some areas of investigation that
were out of bounds if he wanted to have a just society. Nothing strange
were, and still are, consistent with the daily
our important newspapers and magazines.
Leroi, a biologist at Imperial College, feels differently. He loves what
he calls "the problem of normal human variety".
uniquely among modern scientific problems, he says, "it
can apprehend as we walk down the street. We live in an age now
where the deepest scientific problems are buried away from our immediate
perception. They concern the origin of the universe. They concern
the relationships of subatomic particles. They concern the nature and
structure of the human genome. Nobody can see these things without large
bits of expensive equipment. But when I consider the problem of human
I feel as Aristotle must have felt when he first walked down
to the shore at Lesvos for the first time. The world is new again. What
is more, it is a problem that we can now solve, a question we can now
answer. And I think we should.
"Of course, there will be people who object. There will be people
who will say that this is a revival of racial science". Leroi
argues that "there will always be people
who wish to construct socially unjust
racial differences. And though it is true that science
can be bent to evil ends, it is more often the case that injustice creeps
in through the cracks of our ignorance than anything else. It is to finally
off those cracks that we should be studying the genetic basis of human
is a Reader in Evolutionary Developmental Biology at Imperial College,
London. He is the author of Mutants: On Genetic Variety and the
Human Body, winner of The Guardian First Book Award, 2004.
NATURE OF NORMAL HUMAN VARIETY
The question that interests me, as it does so many other people, is
how to go about making a human being. It's a very difficult problem.
Roughly what it boils down to is this: We have 30,000 genes or so. It's
often said that the genome is something in the nature of a book. It
has words, a grammar, a syntax, and of course those words have meaning.
The only problem is that we don't know actually what that meaning
is. So the question is, how do we decipher that? And turning that question
around, looking at it from the point of the human body, what do those
genes mean to the construction of the human body?
Of course, I myself don't actually work on humans. They're
just too inconvenient. I work on worms. This worm is Caenorhabditis
elegans, for which Brenner, Sulston, and Horvitz just won the Nobel
Prize. And the reason why I and a thousand other scientists work on
this worm is that, for all of their marvelous properties, it's
easy to keep thousands of them in petri dishes, and it's very
easy to find mutants. And that's the critical thing. We find mutants
that interrupt particular genes, and that tells us what those genes
do and what they mean to the body of a worm.
Developmental biologists have been doing this for a long time—once
a field has its Nobel, you can be sure it's reasonably mature. What
people haven't really done, however, is to do this for the human body.
The reason is obvious: you just can't go out and generate mutants in
humans. For humans you've got to go out and find those mutants. But
they're out there. There are thousands upon thousands of mutants out
there—no, more, millions—no, actually billions. This is
because we are all mutants. That's one thing you don't expect but which
happens to be statistically true. Each of us carries mutations that
interrupt particular genes. So if you can just find who is a mutant
for a particular gene, and examine what they look like, you can actually
then work out what those genes do.
This raises a question: what exactly is a mutant? Worm biologists and
fly biologists—geneticists generally, working on model organisms—use the word "mutant" in a very particular way.
In worms and flies there is an arbitrarily defined strain which we call
the "wild type". But in humans there is no arbitrary wild
type. So can you, in fact, speak of mutant humans?
You can, but the definition of what is a mutant in humans is necessarily
more roundabout. This is because we have such an extraordinary amount
of natural variation in our species. If you go around the world you
see tall people, short people, red-haired people, brown-haired people,
people with curly hair, people with no hair, and so on. Given all this
variation, what exactly, and who exactly, is a mutant? It's an
important question, because to say something is a mutant is in fact
to make an invidious distinction. This is to say something is not just
different but actually abnormal in some fashion. Yet despite the fact
that there is so much variation in our species, it is actually possible
to speak in a coherent way of mutations and of mutants in humans.
Roughly, the reason you can do this is as follows: If you look at the
coding sequence, more precisely, the protein sequence produced by any
given gene, it's the case that for most genes nearly everybody
has the same version. True, there are some genes that are polymorphic — variable — and these are genes that give us all our natural
diversity. But they actually constitute a very small fraction of the
genome. Most people have the same functional version of a gene. Given
that fact, you can define a mutant as somebody who has a rare variant
of a gene, moreover, a variant that harms him in some fashion. And if
you look at it that way, it's clear that we all carry rare variants
that do us harm in some way, and that we are in fact all mutants.
We can even put some numbers on this. One of the really surprising results
in recent years, which comes from the comparison of the genomes of different
species, is that every newborn child carries three novel deleterious
mutations, that is, mutations that its parents didn't have. Not
only that, but each child inherits at least some of the mutations that
its parents have as well. It's estimated then — and of course
this is just an estimate — that every newly conceived person has
something like 300 mutations that affect its health for the worse in
Of course, that number doesn't tell us a whole lot. We not only
need to know about the number of mutations that we have, but also the
distribution of their effects. This is because some mutations have very
severe effects. They are the mutations that cause the big known inherited
diseases — and about 10,000 such diseases have been identified
so far. But there must be many, many more mutations that do us harm,
but only subtly so. These are the mutations that give us weak eyes,
bad backs and the like. These are mutations that we know very little
about but that statistically speaking must be there. It's actually
in these mutations that a lot of human health lies—or rather
the absence of human health lies. At least it does once you have gotten
rid of the contagious diseases.
When I speak of mutations that do somebody harm what I really mean is
not so much that they just affect physiological health; what I really
mean is they affect the Darwinian fitness, the probability that they
will reproduce. It's an evolutionary definition. It's the
kind of definition that can encompass an enormous range of impairments,
and the kinds of impairments that you see that are caused by mutations
are at times of a degree and of a form that you really just cannot conceive
If you go to teratology museums—literally "monstrosity
museums"—in places such as Amsterdam and Philadelphia,
you can see rows of babies in bottles. These infants, usually stillborn,
are deformed in ways that are truly hideous, that really represent the
kinds of monstrosities that you might expect from Greek myth. I mean
this quite literally. They include children born with a single eye in
the middle of their forehead, and who look exactly like the monsters
of Greek myth—Polyphemus in The Odyssey, for example. Indeed,
it's sometimes suggested that the monsters of Greek myth were
inspired by deformed children, and this seems to be a fairly remarkable
correspondence, at least with some of them.
These infants, when you see them, are truly horrific. But very quickly,
after you look at them, a sort of intellectual fascination takes over
because it's clear that these children tell us something very
deep about how the human body is built. Take, for instance, these children
with a single eye in the middle of their foreheads. The syndrome is
called, appropriately, Cyclopia. Cyclopia is caused by a deficiency
in a gene called Sonic hedgehog. Sonic hedgehog is named after a fruit
fly gene which when mutated causes bristles to sprout all over the fruit
fly larva, hence "hedgehog". When the gene was found in
mammals, some wit called it Sonic hedgehog after the video game character.
If you get rid of this gene, bad things happen. You lose your arms beneath
the elbow and legs beneath the knee. The face collapses in on itself,
such that you get a single eye in the middle of the forehead and the
rest of the face collapses into a long, trunk-like proboscis. The forebrain,
which is normally divided such that we have a left and a right brain—the left and right cerebral hemispheres—is fused into
a single unitary structure. Indeed the technical name for this syndrome
is called Holoprosencephaly.
Now all this is very horrible, and actually that's just an initial
list of things that can go wrong in infants that have no Sonic hedgehog.
But what's really interesting about it is that by looking at infants
of this sort you can reverse-engineer and ask what Sonic hedgehog does
in the embryo. Instantly it tells you that one of the things that Sonic
hedgehog does is to keep our eyes apart because if you don't have
the gene the face collapses. It also separates the left and and right
sides of our brains. And it's needed for the formation of our
arms and legs. In fact, it is one of the most ubiquitous and powerful
molecules in the making of our bodies.
And other, more subtle mutations, tell more about it. For example, just
as having too little Sonic hedgehog causes the face to collapse in upon
itself, having too much causes it to expand. I was recently in San Francisco,
in Jill Helms's lab at the University of California, San Francisco,
where she's got a jar containing the head of a pig. Or is it two
pigs? It's just not clear since the jar contains a pig with two
faces, two snouts, two tongues, two throats, and three eyes. It's
not a Siamese twin pig; it's just a pig with two faces. Chickens
and pigs with two faces crop up periodically, as indeed do humans with
two faces, or nearly two faces. There's a syndrome in which you
have eyes that are very widely spaced from each other, and in which
the nose becomes duplicated. You have two noses side by side in two
varying degrees of development.
The gene for this syndrome has recently been cloned, and guess what?
It turns out to be the gene that controls Sonic hedgehog, and that,
in fact, switches it off. People with the syndrome have too much Sonic
hedgehog just as infants with Cyclopia have too little. So by looking
at a range of these kinds of syndromes you can put together a very complete
picture of how a gene like Sonic hedgehog controls one particular feature
of us, the width of our faces. It's a very mundane thing that
you'd hardly think about, but that seems to be controlled by this
There are many other disorders that equally informative. The star at
the Mütter Museum at the College of Physicians of Philadelphia
is Harry Eastlack, a man who had a disease called Fibrodysplasia Ossificans
Progressiva. It's a disorder in which supernumerary bones form.
The Mütter Museum has his skeleton, which he donated at the time
of his death when he was in his forties. The skeleton is essentially
not one man's skeleton—it is, as it were, one skeleton
encased in another. What happens in this disorder is that wherever you
get a bruise or a wound, instead of normal cells moving in to regenerate
the skin and the flesh and heal the wound, bone forms. So every bruise
turns to bone. The kids are born relatively normal, but as they go through
life bone accretes all over them such that they can no longer move.
They become rigid, locked into place. You can cut it away, of course,
but as soon as you make an incision, and that incision heals, more bone
forms. So it's a vicious circle. We don't know which gene
is mutated in this syndrome. But it's almost certainly got something
to do with bone morphogenetic protein, a protein that is, as the name
suggests, normally involved in making bone in infants. It's just
that in most of us this gene switches off. In these people this protein
keeps on being produced throughout life, especially when there's
a wound. It's another marvelous instance of how a given mutation
can tell us something important about how bones are formed. FOP is a
very rare disorder, and the reason why the gene hasn't been cloned
is because to identify genes, to clone genes, you need to have big pedigrees.
At least it helps. But these people just never have children.
People sometimes ask what developmental biology is good for. We can
identify genes that are responsible for making this or that part of
the human body. But in humans, of course, there's a very pressing
question: namely, how can you fix the deleterious consequences of these
mutations? It's one thing to go into a clinical genetics ward,
a pediatric ward, and study kids who are seriously deformed, and say,
"This is just terribly interesting. Your son is highly informative
about the function of the Jagged-2 gene." — but that's
not a great deal of comfort to the parents who actually have to deal
with raising children who are variously deformed and may die or, at
the very least, have to undergo a great deal of surgery. That, of course,
is the problem. The molecular biology is beautiful, but when it actually
comes to curing people, you just have surgery — which is little
more than a rather sophisticated form of butchery.
The great promise, of course—and it's been a promise for
years now and will remain so for some time—is that by learning
about what genes do and how organs and tissues are constructed we can
reconstruct them as we wish. By working out the program we can take
cells, put them in a test-tube, and rebuild tissues. You don't
have cartilage in your larynx? We can build it for your child, and we
can fix it. You don't have a breast? We can rebuild that, too.
And so on. This is a whole new area called tissue engineering. There
are big institutes devoted to it now, where engineers, materials scientists,
and molecular biologists are all working together. So far, it must be
said, it's more institution building and propaganda than real
results, but it'll happen. That is when the justification for
this whole science will ultimately come.
There's no doubt that when you see some of these children who
are so terribly deformed, it's very difficult. It's shocking,
it's heartbreaking, and if you spend any time with them, whether
they're alive in pediatric wards, or whether they're just
babies in bottles, it takes a real psychological toll. I certainly don't
ever get completely hardened to it. But what is also true is that intellectual
fascination of seeing what has gone wrong in these unusual bodies takes
over. This is especially the case when your eye is attuned to perceiving
the differences in detail, once you see that it's not just arbitrary
deformity, once you understand what you are really looking at are the
outcomes of the laws that regulate and make the human body. When this
happens, deformity acquires a real beauty. It's a beauty that
emerges from answering one of the oldest questions in biology: namely,
how are we put together?
that's intellectual beauty. What of human physical beauty? This
is something that interests me greatly. I'm not interested in
the general aesthetic question here, but ourselves. Some people say
that beauty is uninteresting and that it's just a matter of taste.
I don't think so. I would say, and there are others who would
certainly agree with me, that we have a general psychological program
from which stems a universal notion of beauty. Incidentally, this idea
that we all perceive certain features to be beautiful is one that Darwin
would have disagreed with. Darwin believed that the perception of beauty
was particular to particular peoples in particular times and places.
He was probably wrong, or at least he was only partly right. I won't
attempt to justify that answer, but I think it to be true. These days,
the general thinking tends to be that there's a universal notion
of beauty which is true for people around the world. And the question
is, what is that and what drives it?
Many people think that beauty is a certificate of health; this is an
idea that comes out of sociobiology. But it is more obvious than than
that. It's simply the idea that beautiful people are healthy people
and we search for healthy mates. And that's probably true. Or
at least it was. But is it still? In the past, health was primarily
a matter of environmental conditions—your exposure to contagious
diseases and the amount of food that you had when you were growing up.
Rich people had better environments, hence the positive association
between beauty and wealth. But what of modern economically egalitarian
societies such as Holland? In such societies, does the ancient association
still obtain? If the variance in beauty is due to the variance in the
quality of the rearing environment then it must be the case that the
Dutch — who all eat much the same good food, live in much the
same well-designed houses, and have access to much the same excellent
health-care — must all be equivalently beautiful. But is this
so? The answer is, of course, no. Among the Dutch you can find good-looking
and not so good-looking people. And the question is then, why?
I would argue that the reason for this is that there is and will always
be variance in beauty is because there is variance in mutational load.
What is beauty fundamentally about? I would argue — and this
is really just a postulate at this time, but it is one that interests
a great deal — that the fundamental reason why some of us are
more beautiful than others is because of those deleterious mutations
that we all carry We may carry 300 deleterious mutations on average,
but there is of course a variance associated with that. Not everybody
has 300. Some people have more, some people have fewer. If this is
statistically it must be true — then someone in the world has the fewest
mutations of all. Someone in the world is the least
mutant human of all. Indeed, we can actually calculate, making some
assumptions about the shape of the distribution, how many mutations
that person has — and it turns out to be 191 versus the average
of 300. This, to my mind, is surprisingly many. I would suggest that
if we could find that person, he or she would be a good candidate for
being the most beautiful person in the world. At least she would be,
assuming she did not grow up in some impoverished underdeveloped nation.
Which, statistically, she will have done since most people do.
one more thing that I should like to know about, and that is the nature
of normal human variety. There are tens of thousands of geneticists
around the world, all of whom are busy identifying the genes that cause
human disorders of one sort or another. Historically they began with
the really easy ones, the big congenital disorders, especially those
that allow people to survive and produce children and so have big pedigrees
that allow them to map the genes. Now the emphasis has shifted to studying
the genetic basis of more subtle and more complicated kinds of disorders,
things like diabetes and cancer, that have lots of genes that underlie
them each of which has a small effect. This is a much more difficult
task, but people are doing it because these are the inherited diseases
that affect millions.
But there's one aspect of human inheritance that people are resolutely
ignoring. And that is normal human variety. Or, to put it more crisply:
race. If we look around the world we find that people look very different
from each other. These differences are manifestly genetic. They must
be. That's why people's kids look like them. Yet we know nothing about
that variety. We don't know what the differences are between white skin
and black skin, European skin versus African skin. What I mean is we
don't know what the genetic basis of that is. This is actually amazing.
I mean, here's a trait, trivial as it may be, about which wars have
been fought, which is one of the great fault lines in society, around
which people construct their identities as nothing else. And yet we
haven't the foggiest idea what the genetic basis of this is. It's amazing.
Why is that?
The reason is two-fold. The first—which is not such a trivial
problem—is that skin colour is not controlled by one gene. If
it were only one gene we'd know it. It's many genes—more than
three but certainly fewer than 30. It's a difficult problem, although,
frankly, it's not such a difficult problem that if geneticists really
wanted to solve it, they couldn't. It'd be easy enough to do if they
put a fraction of the effort that went into discovering the BRCA1, the
breast cancer gene. I'm not saying they should; finding the breast cancer
gene is more important than discovering the genetic basis of white and
black skin, but still, it's not a technically impossible thing to do.
But of course the fundamental reason why people don't do it is because
it's race genetics. It's because of the long and sorry history of
and racial differences. And indeed, more than that, the whole
thrust of genetics since the war has been to argue that races don't
exist and that they are just social constructs. This is very much the
Harvard School — Dick Lewontin for example has been one of the
big proponents of this point of view. The late Stephen Jay Gould
After the Second World War, when the enormities of Nazi science really
hit home — which were in turn the consequence of a much larger
racial science, not just in Germany, but everywhere—all right-thinking
scientists made a resolute effort to ensure that science would not be
bent to such evil purposes again. They were determined that science
would never again be used to make invidious discriminations among people.
The immediate result of this was the UNESCO Declaration on Race in 1950,
fronted by Ashley Montagu and backed up by geneticists such as Theodosius
Dobzhansky which affirmed the equality of races. Then, in the 1960s,
Dick Lewontin and others discovered that gel electrophoresis could be
used to survey genetic variation among proteins. These studies showed
that humans have a huge amount of concealed genetic variation. What
is more, most of that genetic variation existed within continents or
even countries rather than among them. UNESCO said races were equal;
the new genetics said they didn't exist. Finally, moving a few decades
on, the Out-of-Africa hypothesis of the origin of Homo sapiens comes
to the fore, and multi-regionalism falls from fashion as it becomes
clear that humans are not only a single species — something which
we've known since Linnaeus' day—but a single species that has
only diverged into sub-populations very recently.
The result of this history — which has been partly driven by data,
and partly by ideology — is that these days anthropologists and
geneticists overwhelmingly emphasise the similarities among people from
different parts of the world at the expense of the differences. From
a political point of view I have no doubt that's a fine thing. But I
suggest that it's time that we grew up. I would like to suggest that
actually by emphasizing the similarities but ignoring the differences,
we are turning away from one of the most beautiful problems that modern
biology has left: namely, what is the genetic basis of the normal variety
of differences between us? What gives a Han Chinese child the curve
of her eye? The curve I read once described by an eminent Sinologist
as the purest of all curves. What is the source of that curve? And what
gives a Solomon Islander his black-verging-on-purple skin? Or what makes
Actually, the last is the one thing we do know. It turns out that red
hair is due to a mutation in a gene called MC1R, melanocortin receptor
1, which controls the production of p eumelanin, black pigment, versus
red pigment, phaeomelanin. Rather marvellously, it also turns out that
mutations in MCIR also cause red hair in red setters, Scottish cattle,
and red foxes. But we don't know what causes brown eyes versus blue
eyes versus green eyes. We know very little about the variation in normal
human height. We don't know why some girls have big breasts and some
of them have small breasts. These are important questions — or
at least jolly interesting ones—and we just don't know their answers.
The reason I love the problem of normal human variety is because, almost
uniquely among modern scientific problems, it is a problem that we can
apprehend as we walk down the street. We live in an age now where the
deepest scientific problems are buried away from our immediate perception.
They concern the origin of the universe. They concern the relationships
of subatomic particles. They concern the nature and structure of the
human genome. Nobody can see these things without large bits of expensive
equipment. But when I consider the problem of human variety I feel as
Aristotle must have felt when he first walked down to the shore at Lesvos
for the first time. The world is new again. What is more, it is a problem
that we can now solve, a question we can now answer. And I think we
Of course, there will be people who object. There will be people who
will say that this is a revival of racial science. Perhaps so. I would
argue, however, that even if this is a revival of racial science,
we should engage in it for it does not follow that it is a revival of
racist science. Indeed, I would argue, that it is just the
opposite. How shall I put it? If you want to prove, what most of us
believe, that skin colour does not give the measure of a man, that it
tells nothing about his abilities or temperament—then surely the
best way is to learn about the genetics of skin colour and the genetics
of cognitive ability and demonstrate that they have nothing to do with
each other? The point is that there will always be people who wish to
construct socially unjust theories about racial differences. And though
it is true that science can be bent to evil ends, it is more often the
case that injustice creeps in through the cracks of our ignorance than
anything else. It is to finally close off those cracks that we should
be studying the genetic basis of human variety.