Documentos de Académico
Documentos de Profesional
Documentos de Cultura
652
AMERICAN ANTHROPOLOGIST
SEPTEMBER
1998
African ancestry, or to lump them together for some purposes with their parent populations in Africa as constituting a "Negroid" group.
Proponents of the race concept acknowledge that racial
classifications can be used to discriminate against people.
But because such classifications reflect certain facts of human biology, they can also be used justly and fairly to
serve benign ends. For example, doctors need to be alerted
to the elevated probability of sickle-cell disease in patients
of equatorial African ancestry. Forensic anthropologists
may be asked by the police to provide racial identifications to help in solving crimessay, to determine whether
a skeleton found in the woods could be that of an African
American murder victim. Because there are some skeletal
traits that occur more frequently among some North
American ethnic groups than among others, it is sometimes possible to answer such questions with a fair degree
of confidence. And because racially defined ethnic groupings are real and important elements in American culture,
we often need to recognize such groupings in investigating the interaction between culture and biology. For instance, if we wish to determine whether Black children
have been systematically exposed to higher environmental lead levels than Whites, we need to structure our
sample in terms of race (Schell 1997).
CARTMILL
653
were pretty well correlated with geography (with Negroids restricted to Africa, Caucasoids to western Eurasia,
Mongoloids to eastern Asia, and so on) until the era of
European colonialism, when massive population movements both voluntary (like the colonization of South Africa by Dutch settlers) and forced (like the initial colonization of the Americas by enslaved Africans, or of Australia
by deported English convicts) brought different races together in various parts of the world and produced racially
mixed populations that are not easy to classify. This artificially induced and unnatural commingling of different
races has muddied the original picture, but enough of the
human species remains in a relatively pristine condition to
enable us to reconstruct the original situation by studying
"primitive isolates" today in uncolonized parts of the
world like Amazonia, the Ituri rain forest, and Lapland. Or
so the story goes.
It is true that human populations in some parts of the
world were more uniform and distinctive a thousand years
ago than they are at present. But populations like those of
modem North America, with high levels of phenotypic
variability maintained partly by migration and gene flow
from elsewhere, are not a new phenomenon. Similar
populations have inhabited northern and southern Africa
and much of western, central, and southern Asia for centuries or millennia. It would have been just as futile an exercise to try to apply racial typologies to the highly variable
people of Egypt or India four thousand years ago as it is to
do so in the United States today. In such populations, "racial" types are polymorphic, like ABO blood-group phenotypes. It makes no more sense to classify the individuals
comprising these populations into racial categories based
on epidermal pigmentation, hair texture, or nose, lip, and
eyelid shape (the traits that loom largest in our racial typologies, probably because they are all visible in people's
faces) than it would to separate them into races on the basis
of their ABO phenotypes. In fact, it makes even less sense,
since ABO phenotypes are discrete, whereas "racial"
types in such populations are highly variable and intergrade imperceptibly with each other.
There are of course things to be learned about human
adaptations by trying to reconstruct the past distributions
of human phenotypes. Skin pigmentation furnishes a familiar example. The darkness of human skin in the Old
World appears to be inversely correlated with distance
from the equator. Populations that deviate from this general pattern can plausibly be interpreted as recent migrants
from higher to lower latitudes, or vice versa (Brace 1996).
The pattern probably reflects a long history of low-level
natural selection favoring dark skin in areas of high yearround insolation. We might not discern this pattern so easily if we used modern data uncritically and pretended not
to know that the presence of large numbers of pinkskinned people in the Transvaal and of black-skinned people in Canada is a relatively recent phenomenon.
654
AMERICAN ANTHROPOLOGIST
But there are also things we can learn about human adaptations by looking at modem phenotype distributions.
European colonies were established throughout most of
the world during the eighteenth and nineteenth centuries.
In temperate-zone areasAustralia, New Zealand, North
America, southern South America, and South Africa
these colonies tended to expand and supplant or supplement the indigenous peoples; in tropical areas, they generally made less of an impact. Crosby (1986) suggests that
this differential success reflects European cultural adaptation to temperate-zone ecologies. However, it might also
reflect patterns of selection against European phenotypes
in equatorial climates. Perhaps pink-skinned variants of
Homo sapiens do not thrive in areas of high year-round insolation. To test this thesis, we need to look at recentpopulations, not at our reconstructions of populations that lived
a thousand years ago. Both past and present distributions
provide biological data on human adaptations and microevolution. It is a mistake to think that today' s human populations are somehow unreal, unnatural, or corrupted, or
that modem technology has freed them from selection
pressures.
No matter whether we look at past or at present-day
populations, the use of racial categories in structuring our
samples hinders, not helps, our efforts to describe human
variation and explain its causes. If we want to frame or test
hypotheses about the adaptive significance of skin color,
the appropriate question to ask is not, "Do Negroids do
better than Caucasoids in some environments?" but
rather, "Do dark-skinned people do better than lightskinned people in some environments?" Since there is
considerable variation within, and overlap between, "Negroids" and "Caucasoids" with respect to skin color, analyzing the variation in terms of racial categories serves
only to blur the question and introduce irrelevant variables
into the data.
Similar criticisms apply to any attempt to use racial
categories in describing or analyzing human genetic variation. Since there are thousands of separate, independently assorting variable loci in the human genome, it
is highly unlikely a priori that variation at any particular
locus will covary with any other. We would therefore expect on theoretical grounds that a descriptive classification based on a small number of "racial" traits would be of
little use in summarizing the variation that occurs at any or
all of the other variable loci in the human genome. Empirical studies bear out this expectation. Even if we try to
backdate the evidence to A.D. 1500 by restricting our data
base to supposedly isolated aboriginal populations, the
geographic patterns of variation at most loci do not fit our
racial typologies. In the case of the ABO system, for example, the A allele reaches its highest frequencies in
southern Australia, Europe, northeastern Asia, and the
Arctic fringes of North America; the B allele in central
Asia and West Africa; and the O allele in the New World,
SEPTEMBER
1998
CARTMILL
655
160
140 -
a
120 -
/
a6
Total humarwanaSon
C3--Tolal AJPA articles per year
Year
E S 2 ! S E : 2 ~ I D " '
2 2 2 2 S & S S S S S S S S S S S
Figure 1. Annual publication rates in the American Journal ofPhysical Anthropology, 1965-1996. White squares signify total scientific
articles, and black diamonds, total articles on modem human variation.
found in these 810 papers: "Typical Mongoloid and Negroid collectives," "American Japanese males," "(American) subjects primarily of African origin," "Full-blooded
Papago Indians," and "Australians of pure Aboriginal ancestry." Some papers could have been scored either way,
and a different observer would no doubt come up with
somewhat different definitions and numbers, but the standards used here were at any rate applied fairly consistently.
In the sampled years, the number of articles about modern human variation published annually in the AJPA rose
almost every year from 1965 to 1983. Since then, it has declined steadily. This rise and fall partly reflects a parallel
but more erratic fluctuation in the number of articles on all
subjects published annually in the AJPA (Figure 1). But
even when we correct for this by expressing the number of
articles on human variation as a percentage of total AJPA
articles published in each sampled year, there still appears
to have been a long-term decline in the percentage of the
AJPA devoted to the topic of modern human variation
(Figure 2). This might be interpreted as betokening a decreasing interest in the whole subject of "race" broadly defined. However, it might also reflect the proliferation of
new journals that provide alternative publication venues
for articles on human biology and genetics.
Of the 810 articles on modern human variation published in the AJPA during the years sampled, 40.5% utilized racial categories. The annual tally of racial-category
articles as a percentage of all human-variation articles
fluctuates widely around a near-identical mean of
10
r-
t*.
r-
t**
Year
656
AMERICAN ANTHROPOLOGIST
neither the proponents nor the opponents of racial classification have any grounds for thinking that history is on
their side.
SEPTEMBER
1998
mals. To live, we need to breathe, assimilate food, and excrete wastes. If we stop doing any of these things, we die.
Eventually we die anyway, no matter what we do. Like
other sorts of animals, we also face particular, speciesspecific biological constraints. Salmon can breathe water
but cannot learn to play the piano. The reverse is true of
most human beings. Cultural innovations may someday
enable us to evade our present biological constraints, but
at the moment we are stuck with them.
People also face environmental constraints on their
lives. Although many of these are beyond human control,
a lot of them are imposed on us by other people. Mostof us
could bericher,wiser, kinder, more accomplished, healthier, and happier than we are if only we had spent our lives
in different environments. Almost every sort of human potential is limited by both environmental and biological
factors. I cannot learn everything there is to leam because
my brain and my lifespan are finite. This is a fact of human
biology, which would be true in any environment currently attainable. But by the time I die, I will have learned
even less than I might have given my brain and lifespan,
because of my choices and because of the constraints
placed on me by my environment. This is true for all people, no matter what their hereditary capacities are.
I hope that we can all agree that these are simple, obvious truths about the human condition. (They are also truths
about the salmon condition, the horse condition, and so
on.) It follows that it makes no sense to ask whether a particular capacity is in principle limited chiefly by heredity
or by environment. Everything is in principle 100% limited by both heredity and environment. The life of a concert pianist must begin with a fertilized human ovum; an
opossum ovum will not suffice, because of its biological
limitations. On the other hand, no matter what sort of a human ovum we start with, it cannot develop into a concert
pianist in most environmentssay, in Europe in 10,000
B.C., or in the womb of an opossum, or in a 10% solution of
formaldehyde. Asking whether piano-playing skill is primarily determined by heredity or by environment is therefore meaningless. The relative importance of heredity and
environment in producing the observed differences between people in this or any other trait depends on the relative variabilities of the two factors in any particular situation. If all people were raised in identical environments,
any differences among them not due to their own choices
would obviously be due to heredity. If they were genetically identical, all such differences would be due to environmental factors.
In the world as it is today, it seems clear that some of the
differences between people (say, the differences between
an infant with Tay-Sachs disease and its parents) are almost entirely determined by genetic factors. Others (say,
the differences between political liberals and conservatives) are as far as we know determined almost entirely by
environment and individual choices. Yet others (e.g.,
CARTMILL
piano-playing ability) are probably determined by combinations of, or interactions among environment, heredity,
and choice. Although scientists often try to estimate the
relative contributions of these factors to the observed variation in various human traits, we need to remember that
such estimates are themselves dependent on environmental variables. Some people believe that they can
evaluate the relative contributions of environmental and
genetic factors to a trait by comparing that trait's variation
among identical twins (who are clones of each other) with
its variation among fraternal twins (who are genetically
different). But even this approach does not really allow us
to factor out environmental influences altogether, because
the environment determines the extent to which a given
trait is influencedby geneticfactors. For example, in acultural context where (say) redheaded people were stereotyped as stupid and ineducable and were accordingly neglected by their teachers, we would expect identical twins
to resemble each other more closely in their educational
attainments than fraternal twins, simply because identical
twins are more likely to have the same hair color than fraternal twins are. In such an environment, success in school
might be causally linked with genetic factors that would
not affect educational attainment in other environments.
Because the degree to which any trait is genetically conditioned depends on environmental circumstances, there
is no such thing as "heritability" in the abstract. To quote
Weizmann et al. (1996:192-193), "Heritabilities depend
on the specific genetic composition of the population and
the environmental circumstances experienced by that
population.... [They] cannot be generalized to other
populations or other environmental conditions."
What is true of heritability is also true of fitness. The
theory of natural selection entails that within any species,
some genetic variants are more fit than othersthat is,
there are nonrandom factors that make certain variants
more likely than others to leave copies in the gene pools of
succeeding generations. Not all evolutionary change is
driven by natural selection, and it is not always easy to distinguish variants favored by selection from those that
prosper due to mere coincidence. For example, average
human skin pigmentation may well have decreased from
the seventeenth through the nineteenth centuries A.D. as a
side effect of the great expansions and emigrations of
light-skinned European populations during the era of colonialism. At the moment, human pigmentation may be on
the upswing again due to higher population growth rates in
the tropical countries of Africa, Asia, and the Americas,
where average skin color is darker than it is in Europe.
There is no reason to suspect that these historical fluctuations reflect changing patterns of natural selection on human skin color.
Genetic variants favored by natural selection can be
properly described as biologically superior to others. But
such variants are superior only in relation to a specific en-
657
vironmental context, including the species itself, its population structure, and its relationship to and interactions
with all aspects of its circumstances. Again, there is no
such thing as generalized fitness in the abstract. For example, lizards bom without limbs are generally at a disadvantage, but there have been situations in the past where this
was not the casewhich is why there are snakes and limbless lizards in the world today. Likewise, people who are
born without limbs are generally at a disadvantage; but we
can imagine or create environments where they are just as
fit as anyone else, or even more so.
Our culture leads us to regard mental abilities as the
most important markers of human status. Bothour cultural
traditions and our own professions as scholars and teachers encourage us to lump all mental abilities together as a
single variable called "intelligence," to equate high "intelligence" with biological superiority, and to feel that people with exceptional mental abilities somehow deserve to
be at the top of the heap. When a man with a crippled body
becomes a great astrophysicist, we are awed and inspired
by his example. When an illiterate, inarticulate, and unreflective man becomes a great boxer, we are less impressed. If the stupid prizefighter makes ten times as much
money in the course of a year as the crippled astrophysicist, we regard it as a scandal. Because intellectuals tend to
value other skillful manipulators of symbols more highly
than they value skillful boxers, gardeners, hunters, or masons, most of the published debate concerning the supposed biological superiority of certain human populations
has centered around the issue of congenital average differences in "intelligence" between "races." While nobody
gets very excited if scientists suggest that Swedes are on
the average taller than Japanese for genetic reasons, everybody gets hot under the collar whenever someone
claims that Swedes are on the average smarter than Nigerians for genetic reasons. The difference between the two
responses is due in part to the fact that "intelligence" is a
notoriously dubious variable, which is far less clearly defined and less easily quantified than "height." But at a
deeper level, the difference reflects the different cultural
values that we attach to stature and IQ.
However we choose to define or subdivide "intelligence," it is an unpleasant fact that some genetic variants
make their possessors stupider than other people: that is,
they result in impaired mental abilities in all currently attainable human environments. Some of these genes are
known to be significantly more common in some human
populations and ethnic groups than in others. These two
facts suggest (but do not prove) that human populations
and ethnic groups may well differ congenitally in average
mental potential at birth. This conclusion sounds shocking. However, even if it is true, it turns out to be far more
innocuous and less interesting than either racists or egalitarians assume.
658
AMERICAN* ANTHROPOLOGIST
SEPTEMBER
1998
represents a "gene for intelligence." In other environments, it does not. Our science and technology enable us to
create favorable environments in which phenylketonuria
becomes a relatively harmless genetic variant. Someday,
we may be able to create environments in which TaySachs homozygotes, or other sorts of so-called "congenital mental defectives," are not handicapped. As we alter
our environments in pursuit of such aims, the contribution
of genetic variation to variation in human mental abilities
will decline accordingly.
I suspect that the question of interethnic differences in
average mental abilities attracts more attention than it deserves because some of the people who write about it and
ought to know better are not really thinking about heredity
in terms of particulate Mendelian inheritance. Rather,
they are thinking about it in terms of our folk concepts of
"blood" heredity. Suppose for the sake of argument that,
say, the inhabitants of Ireland make lower average scores
on IQ tests than the inhabitants of Scotland. Average genetic differences between the two populations might well
be contributing to that difference in test results. B ut even if
all this were true, it would not imply that Irish "blood"represents some sort of a hereditary taint that dooms all those
of Irish descent to some degree of congenital thick-headedness. It might simply mean that, say, the phenylketonuria mutation occurs in higher frequencies in Ireland
than in Scotland. Any genetically conditioned difference
in average mental abilities between two human groups
will fit this general description.
As the theory of natural selection would lead us to expect, all genetic variants known to yield gross mental deficiencies in the present range of human environments occur in quite small percentages in every ethnic group, local
population, or "race." No doubt there are undiscovered
variants that produce smaller and subtler deficiencies in
various environments. Such genetic variants will be less
heavily selected against than the seriously deleterious
variants that produce phenylketonuria or Tay-Sachs disease, and therefore may occur in higher frequencies; but
they are equally unlikely to occur uniformly throughout
any ethnic group. A possible exception may occur in cases
where certain phenotypes are culturally interpreted as
markers of ascribed membership in a stigmatized group.
In cultural settings where (say) left-handed or darkskinned or obese children are held to be congenitally stupid, we might expect such children to be differentially neglected, discriminated against, and taught to think poorly
of themselves. In these environments, but not in others,
genetic variants promoting right-handedness, light skin,
or slenderness may turn out to be "genes for intelligence."
Again, it should be emphasized that there is no such thing
as a "gene for intelligence" outside a particular environmental context, and that culture always affects the interaction between genes and environment in our species.
CARTMILL
659
image. But social facts are not necessarily part of the biological landscape. In multiethnic regional populations,
races are merely ethnic groups linked to vague, inconsistent, and stereotypical ideal phenotypes. Growing awareness of the meaninglessness of racial taxonomy is currently leading increasing numbers of U.S. citizens to
refuse to classify themselves racially, or to allow themselves to be so classified by others (Fish 1995). In the long
run, we would probably be better off if we all followed
their example.
Notes
Acknowledgments. I am grateful to George Aimelagos, Kayc
Brown, Jon Marks, Dan Schmitt, Ted Steegman, and three
anonymous referees for their helpful comments on earlier drafts
of this paper.
1. For a quick survey of the polar positions among physical
anthropologists on the subject of race, see the recent issue of
Evolutionary Anthropology in which two distinguished physical anthropologists were asked to provide separate reviews of
the same four books dealing with issues of race and human genetic diversity (Aimelagos 1995; Harpending 1995). Each reviewer praised the same books that the other condemned.
References Cited
AAPA (American Association of Physical Anthropologists)
1996 AAPA Statement on Biological Aspects of Race.
American Journal of Physical Anthropology 101:569
570.
Ampola, M.G.
1982 Metabolic Diseases in Pediatric Practice. Boston: Little, Brown.
Aimelagos, G.
1995 Race, Reason, and Rationale. Evolutionary Anthropology 4:103-109.
Aronson.S.M.
1964 Epidemiology. In Tay-Sachs Disease. B. W. Volk.ed.
Pp. 118-153. New York: Grune and Stratton.
Barkan, E.
1992 The Retreat of Scientific Racism: Changing Concepts
of Race in Britain and the United States between the World
Wars. Cambridge: Cambridge University Press.
Blakey, Michael L.
1994 Passing the Buck: Naturalism and Individualism as
Anthropological Expressions of Euro-American Denial. In
Race. S. Gregory and R. Sanjek, eds. Pp. 270-284. New
Brunswick, NJ: Rutgers University Press.
19% Skull Doctors Revisited: Intrinsic Social and Political
Bias in the History of American Physical Anthropology,
with Special Reference to the Work of Ales Hrdlicka. /n
Race and Other Misadventures: Essays in Honor of Ashley
Montagu in His Ninetieth Year. L. T. Reynolds and L. Lieberman,eds. Pp. 64-95. Dix Hills, NY: General Hall.
Brace, C.L.
1964 A Non-Racial Approach toward the Understanding of
Human Diversity. In The Concept of Race. M. F. A. Montagu, ed. Pp. 103-152. New York: Free Press.
660
AMERICAN ANTHROPOLOGIST
SEPTEMBER 1998
RACHEL CASPARI
HE EVENTSSURROUNDING
THEPUBLICATION
of
COPYRIGHT ? 2003,
AMERICAN ANTHROPOLOGICALASSOCIATION
66
addressdenouncingthebook aroundthetimeof itspublicationat theAAAAnnualMeetingin Chicago on November 16, 1962. The publishedversion(Washburn1963) is
much less harsh,focusingon the limiteduse of race as a
valid object of studyand the lack of scientificsupportfor
Public denunciationof
any claims of racial inferiority.
Coon's ideas seemed necessary;segregationists
were alto
their
them
bolster
There
werea
readyusing
arguments.
of
from
the
scientific
variety responses
community.Statementson racewereissuedbyboththeAAAand theAmerican Associationof PhysicalAnthropologists
(AAPA).Several edited volumes appeared throughoutthe 1960s
critiquingthe race concept.In 1966, MargaretMead and
Theodosius Dobzhanskyorganizedan AmericanAssociation forthe Advancementof Science (AAAS)symposium
meantto deliverthe scientificvoice againsta popularracism based on "misinformation"
and "evil myths"about
race.As embodiedby itsorganizers,
the symposiumrepresentedan alliancebetweenBoasian culturalanthropology
and evolutionarybiology,includingdiverseperspectives
fromwithinanthropology,
genetics,ethnology,psychology, and sociology.With few exceptions,most anthropologists had become opposed to hereditarianclaims
about race and intelligence,
and manywerenow skeptical
of the race conceptitself.What became clearby the mid1960s was thatrace was no longera unifyingconceptin
mainstreamphysicalanthropology,
justas it had ceased to
be a unifyingconcept foranthropologyas a whole since
Boas's workon race a halfcenturyearlier.In physicalanthropology,race was now a divisiveconcept. Although
Washburnhad publishedhis ideas about the new anthropologyearlier,this periodmarkeda turningpoint in the
on the
discipline,withgreaterinstitutionalintrospection
race concept.Some have even arguedthat it markedthe
demiseof theraceconcept.
Severalfactorsinfluencedchangingviews about race
withinphysicalanthropologyduringthis time. First,social factorspromptedscientiststo challengeassumptions
ofbiologicaldeterminism
and intellectualinferiority
associatedwiththe race concept.The Holocaustin the 1930s
and 1940s and the controversy
school desegsurrounding
regationin the early1960s may have been the most importantexamples.Anothercomponentof social pressure
resultedfromthe relationshipbetweenanthropologyand
interestin raceand racialinequality,an ingovernmental
terestthat had promotedthe "racialization"of U.S. anthropologyin the firstplace. Second,the race conceptitselfwas challengedby thepopulationalprinciplesespoused
in the modernsynthesis;evolutionaryideas were incompatiblewith the essentialistfoundationsof the race conviewsofpopulationand clines,based
cept,and alternative
frompopulationgenetics,led
largelyon understandings
manyscholarsto considerrace an invalidtool forunderstandingbiologicalvariation.Finally,theevolutionofU.S.
fromitsemergenceas a subfieldto
physicalanthropology,
the presentday, has been influencedby its relationship
withthe restof anthropology-specifically,
four-field
an-
that
thropologyas embodiedby the AAA.It is interesting
as earlyas 1894, a quartercenturypriorto the emergence
of physicalanthropologyas a truesubdiscipline,Boas began to challengethe race concept. By the time physical
anthropologyclearlyemergedin the 1920s,Boas's followers held some of the mostpowerfulpositionswithinU.S.
anthropologyand were a dominant voice in the AAA.
Therefore,the racial physicalanthropologythat was rejectedin the 1960s developedwithina broaderanthropologicalcontextthathad been grapplingwiththe raceconcept foryears;partsofthatcommunityalreadyquestioned
overthe
race,and the AAAhad been involvedin struggles
issue of race betweenanthropologyand governmentpolicies and funding,as well as strugglesbetweenanthropologyand othersciences.The rejectionofrace in the 1960s
was not so new;itwas a partoftheheritageof physicalanthropologywithinU.S. anthropology.
This historysuggeststhatthe raceconcepthas no remaininglegacyin physicalanthropology.What actually
changed?Is the race conceptreallydead? What elements
of the race conceptstillpersistand influencephysicalanthropology
today?In thisarticle,I addressthesequestions,
themwithinthecontextofthe scientific
and
investigating
social influenceson mainstreamphysicalanthropology
thatwere a major forcein the evolutionof the race concept. I arguethatsome elementsof the raceconceptwere
in factrejected,but thatothersremain,subtlyinfluencing
ourviewsofwhatwe todayterm"populations."
THEATTRIBUTES
OF THERACECONCEPT
The race concept that was examined and rejectedby so
manyin the 1960s includesassumptionsabout the cause
and natureofgeographicand otherkindsofvariation.The
historybehind these assumptionshas helped createthe
conceptthatwe grapplewithtoday.Althoughforthe last
100 years the race concept has been thoughtabout in
terms,its most fundamentalelements
quasi-evolutionary
areessentialism,
These
clades,and biologicaldeterminism.
attributesare clearlyrelated,and all of them have informedthe theoriesabout human variationin physical
The raceconcepthas changed,yettheseatanthropology.
tributesof racehave not all changedtogether.While bioand its social implicationshave been
logicaldeterminism
questioned since the inceptionof the field,essentialism
and the concomitantrenderingof races as clades have
been less amenableto change.
Essentialism
The racesdefinedby the Westernrace conceptwerecodifiedby Linnaeus(1758) and by the definitive10thedition
ofSystemae
Naturae,in whichhe describedfivesubspecies
of humans,listingforeach typeboth the morphological
and behavioralcharacteristics
thatwereconsidereda part
of the essenceof the category.These wereimplicitly(and
explicitly)understoodto be partof theintrinsicbiologyof
the race. European prejudiceswere clearlyincorporated
68
strictimmigration.
In a studyof over 18,000 immigrants,
he found changes in head formthat underminedthe
dogmaof thestabilityof racialtypesand the Europeanfocus on head shape as a major indicatorof race. He could
not explainthe causes of change,althoughhe considered
themin some way "environmental"
and, as an empiricist,
that
what
was
argued
importantwas the documentation
of thechangeitself.Throughhis workon racialquestions,
Boas challengedboth biologicaldeterminism
and the nature of racial categories,two criticalcomponentsof the
race concept.These challengeswere a centralpartof anthropologicalthinkingin the U.S. beforeHooton started
producingPh.D.sin physicalanthropology.
"Racializing" PhysicalAnthropology
Both governmentaland privateforcespromotedthe renaissanceofthe "scienceof race"in U.S. anthropology
durWorld
War
counter
to
the
traditions
develI, running
ing
However,therewere
opingin muchof U.S. anthropology.
also tensionsbetween Boasians and other,at the time,
smallerfactionsof the AAAwho weresympathetic
to the
anthropologistsassociated with Washingtoninstitutions
thatuntilHooton
(Stocking1968). It mustbe remembered
startedproducingPh.D.s at Harvardin 1925,therewas no
specifictrainingin physicalanthropologyin the United
States.Only six U.S. Ph.D.s in physicalanthropologyhad
been awardedpriorto 1925-five of thesefromHarvard,
trainedby specialistsin otherdisciplines.Ales HrdliEka,
the founderofAmerican
in
Journal
ofPhysicalAnthropology
1918, and AAPAin 1928, had no studentsin his position
at theNationalMuseumofNaturalHistory.
Duringthe second decade of the 20th century,many
scholarswho claimedto representphysicalanthropology
were actually eugenicistsfrom other disciplines (that
claimed scientificsuperiority
to anthropology),
and some
wereverypowerfulin theU.S. scientific
politicalstructure.
Theseincludedmembersofthe GaltonSociety,whichwas
dedicatedto the studyof racialanthropology,
such as the
Fairfield
Osborn
(then
paleontologistHenry
presidentof
the AmericanMuseumof NaturalHistory),and the biologistsRaymondPearland JohnC. Merriman(head of the
NationalResearchCouncil [NRC]).Manyin the anthropologicalcommunitysaw themas a threat;theywereracial
deterministswith a political agenda, and the BoasiandominatedAAAdid not acceptthemas anthropologists.
Therewas clearlya need forphysicalanthropologists
trainedwithinanthropology.
This dearthbecame veryapwhen
the
National
Research
Councilsoughtto form
parent
an anthropology
whichwas to deal withphysicommittee,
cal anthropologyand eugenics.Aside fromHrdliekaand
Boas, therewerefewphysicalanthropologists
recognized
by the AAAto serve(Stocking1968). Madison Grantand
Charles Davenport,ardentracistsand eugenicists,founders of the Galton Societywith strongpoliticalagendas
and tiesto Washington,servedon the originalcommittee.
Whilethe AAArefusedto recognizeGrantor Davenportas
70
thereremainedenormouspressureon ananthropologists,
to
thropology "racialize,"both from the government,
which had become increasinglyinterestedin restricting
immigrationon racialgrounds,and fromthe eugenicinterestscontrollingothermajor fundingsources.Members
of the Galton Societyincluded the heads of institutions
thathad been (or potentiallycould be) importantsources
of anthropologicalfunding.By the 1920s, fundingincreasedforstudiesof race and racialpsychology.U.S. anthropologistsrespondedto this fundingincreaseand to
criticismsthat they neglected biology and the racial
makeupofthe U.S. byexpressingmoreinterestin physical
anthropology.Ironically,severalof Boas's students(e.g.,
Mead, Herskovitz,
Klineberg)werefundedby NRC fellowships in the biological sciences forworkthat supported
the culturalbasis forracial differences,
and Boas himself
these
sources
for
his
own
work
on race. Other
exploited
studentsof Boas, such as AlfredE. Kroeber(and Roland
Dixon), as well as more conservativenon-Boasian elements of the anthropologicalcommunity,also became
moreinterested
in physicalanthropology,
placingthe race
concept and eugenicsat the focus of the emergingnew
physicalanthropology.
Hooton
EarnestHooton was one of the mostinfluentialfiguresin
physicalanthropology(Giles 1999). He was a professorat
Harvardfrom1913 until his death in 1954 and was responsible fortrainingvirtuallyan entireacademic field,
spawningseveralgenerationsof studentswhen fewother
universities
offeredphysicalanthropologyas partof their
curricula.Hooton's Ph.D. (in 1911 fromWisconsin)was in
classics.He had littlebackgroundin anthropology,
and it
took some time to get his programoffthe ground,but
startingin 1926, a floodof Ph.D.s in physicalanthropology emergedfromHarvard.Withina fewyears,physical
and
anthropologywas a majorpartof U.S. anthropology,
this was reflectedin AAAmembershipand the developmentoftheAAPA.
Race studiescame to be the focusof Hooton's career,
but he formedhis ideas about race and physicalanthropologyin generalafterhe came to Harvard.His workwas
typologicaland manifested,like Haeckel's,as an "evolutionarypolygenism"(Wolpoffand Caspari1997). Hooton's
Harvardcolleague, Dixon, influencedhis views on race;
Hooton's 1931 classification
is verysimilarto Dixon's 1923
classification.
The polygenismof Hooton and Dixon was complicated,groundedin the beliefin once-pureracesthathad
separateevolutionaryhistories.Likemanyotherscientists
with fundamentallypolygenistideas, they understood
thatpresenthumanvariationcould not be accommodated
withina fewracial types.Hooton thoughtthat the complexity of human variation could be accounted for
betweenonce-pureprimaryracial
throughinterbreeding
groupsthatrelatively
recentlyunderwenta secondaryrace
72
Genetics,Populations,and Clines
The need to confrontpublicmisrepresentations
ofscience
thatwereactivelyused in the 1960s fosteredalliancesbetween various elements that had foughtracism before
with some that had not-the architectsof the modern
synthesis,"mainstream"anthropologistsas represented
as represented
by the AAA,and physicalanthropologists
AAPA.
the
The
1966
AAAS
by
Mead-Dobzhanskysymposium represented
thisalliance,as did a numberofvolumes
on the studyofraceproducedat the time(e.g.,Mead et al.
1968; Montagu 1964) thatbroughttogetherworkfroma
varietyof disciplines.One way of attackingracismwas
(and is) througha focus on the inadequacy of the race
conceptforexplaininghumanvariation.Studiesof clines,
the geographicdistributionof individualmorphological
and genetic traits,were introducedand population replaced race as a focus of study.This was by no means
purelypolitical;it was a consequenceof the evolutionary
approachofthenew physicalanthropology.
As C. LoringBrace pointed out in 1964, races,and
even populations,are inadequateforthe studyof human
variation.Instead,he advocatedforthe studyof individual traits-thestudyoftheirdistribution
and the selection
thatcauses theirvariation.The studyof clinescame to replace race as a focus of analysis for many researchers.
FrankLivingstone,
in his 1962 articleon the nonexistence
of human races,eloquentlylays out whyraceor any subspecifictaxonomyis misleading:
aredifferThecausesofintraspecific
variation
biological
variation
andtoapplythe
entfrom
thoseofinterspecific
is notonly
termsubspecies
to anypartofsuchvariation
theexplanaorimpossible
buttendsto obscure
arbitrary
tionofthatvariation.
[1962:279]
He was a strongproponentof nonracialclinal studies,arguing,"thereare no races,onlyclines"(Livingstone1962:
279).
Others,however(e.g., Brues 1972), acceptedthe importanceof clines but arguedthat the biologyof groups
themselveswas also a valid targetof inquiry.With the
populational thinkingof the modern synthesis,which
formedthe basis of Washburn'snew physicalanthropology,populationsreplacedrace as the unit of study.What
did this mean? How did the studyof populationsdiffer
fromrace? Mayr himselfsuggeststhat the populational
74
and physicalanthropologistsno longersupportthe notion that races are subspecies.The importanceof gene
flowand the fluidityof the speciesis recognizedeven by
forensicanthropologists
whose continueduse of typological and raciallycharged termsmakes them appear less
populationalthan theyoftenreallyare. However,in spite
of the rejectionof racesas subspecies,and a reluctanceto
use the term race,populations are oftenthoughtof in
much the same way that raceswere in the earlierliterature. Essentialismcontinues to influence conceptions
about humangroups,and thisis exemplified
by the use of
treesas metaphorsforhuman populationrelationshipsin
studiesof morphologicaland genetichumanvariation.Inclades are an enduringlegacyof racial anthrotraspecific
pologyand continueto informour thinkingabout populations.The raceconceptmaybe rejectedby anthropology,
but itsunderlyingracialthinking
persists.Physicalanthrono
races.
pologists longerstudy
Populationsare now studied, butnot all approachesto the studyofpopulationsare
populational.
RACHELCASPARI Department
of
ofAnthropology,
University
MI 48109-1382
Michigan,AnnArbor,
NOTES
I wouldliketothank
JimCalcagnoforinviting
Acknowledgments.
me to participate
in thissymposium
and inspiring
me to think
aboutraceandphysical
in thecontext
oftheAAA.I
anthropology
alsoappreciate
thehelpful
ofJimCalcagno,
comments
Katarzyna
and an anonymous
reviewer.
I especially
thankFran
Kaszycka,
Mascia-Lees
andtheeditorial
staff
oftheAAfortheir
excellent
sugandsupport.
gestions
1. Ultimately,
raceis a taxonomy
ofpeople.Itcategorizes
people
basedon socialfactors,
evenwhenthosefactors
arebelievedto
"natural
Becauseall taxonomies
represent
categories."
dependon
essentialism
is a critical
ofracialthinkessentializing,
component
is a product
ofthehuing.Today,somearguethatessentializing
manmind,suggesting
racialthinking
mayhavea psychological
tothisview,humanscreate
taxonomies
of
component.
According
thebiological,
worldin similar
social,and physical
ways,cross(Atran1990,1994;Hirschfeld
1996,1998).Thesetaxculturally
onomies
areknowledge
structures
thatallowinferences
tobemade
(beyondthe information
given)about constituent
categories.
Somecategories
aremoreinferentially
richthanothers,
allowing
toform
without
basis.Thesecategories
have
stereotypes
empirical
beentermed
"natural
tobe part
becausetheyarebelieved
kinds,"
ofthenatural
nothumanconstructions.
world-"real,"
"Natural
kinds"areproduced
mechdifferent
through
cognitive
anismsthatarespecific
to particular
domains(basedon different
mentalmodules).
"Natural
kinds"
thatreflect
thebiological
world
have been termed"livingkinds."Peoplelearn"livingkinds"
different
thanthoseused to learn
through
cognitive
processes
aboutinanimatethingsor the processesthatrelateto them.
Hirschfeld
haveargued
to
(1996,1998)andothers
that,inaddition
a cognitive
domainthatgoverns
humanshavea
"livingkinds,"
domainthatallowsthemto easilylearn"humankinds"
separate
andthecomplete
setoftraits
thatmakeuptheessence
ofa particular
kinds"
kind.These"human
aresocialcategories
thatareparticularly
to a culture;
ofthatculimportant
theyarethought
bymembers
turetobe intrinsic
toa person's
Justas biological
categoidentity.
riescarryinformation
abouttheessenceof a species(or a dog
abouttheessenceofa
breed),"humankinds"carryinformation
typeofperson-what
theyaresupposedto looklike,thinklike,
andactlike.Thefactthatmanymembers
ofa category
do notconformto thestereotype
does notdispelthestereotype.
"Human
kinds"aregroups
whosemembers
arebelieved
tosharesomefunda-
REFERENCESCITED
Armelagos,
GeorgeJ.,DavidS. Carlson,and DennisP. Van Gerven
ofSkeletal
1982 TheTheoretical
Foundations
andDevelopment
InHistory
ofAmerican
Physical
Biology.
Anthropology
ed.Pp.305-328.London:Academic
1930-1980.FrankSpencer,
Press.
Atran,Scott
1990 Cognitive
Foundation
ofNaturalHistory.
NewYork:CamPress.
bridgeUniversity
1994 CoreDomainsvs.Scientific
Theories:
EvidencefromSysInMappingtheMind.L.
tematics
andItza-Maya
FolkBiology.
Hirschfeld
andS.Gelman,eds.Pp.316-340.Cambridge:
CamPress.
bridgeUniversity
Barkan,Elazar
1988 Mobilizing
Scientists
againstNaziRacism,1933-1939.In
G.
Bones,Bodies,Behavior.
Essayson Biological
Anthropology.
W. StockingJr.,
ed.Pp.180-205.Madison:University
ofWisconsinPress.
Boas,Franz
1894 HumanFaculty
as Determined
ofthe
byRace.Proceedings
American
Association
fortheAdvancement
ofScience
43:301-327.
1974 TheShapingofAmerican
1883-1911:A
Anthropology,
FranzBoasReader.NewYork:BasicBooks.
Boyd,WilliamClouser
1950 GeneticsandtheRacesofMan.Boston:Little,
Brown.
Brace,C. Loring
towards
theUnderstanding
ofHu1964 ANonracialApproach
InTheConceptofRace.M. F.AshleyMontagu,
manDiversity.
ed.
Pp.103-152.NewYork:FreePress.
1982 TheRootsoftheRaceConceptinPhysical
In
Anthropology.
AHistory
ofAmerican
1930-1980.Frank
Physical
Anthropology
ed.Pp.11-29.London:AcademicPress.
Spencer,
Brues,Alice
1972 ModelsofClinesandRaces.American
ofPhysical
AnJournal
37:389-399.
thropology
Caspari,Rachel,and MilfordH. Wolpoff
1996 Coon andWeidenreich.
HumanEvolution11:203.
Cole, Douglas
1999 FranzBoas:TheEarlyYears,1858-1906.Seattle:University
ofWashington
Press.
Coon,CarletonS.
1939 TheRacesofEurope.NewYork:MacMillan.
1950 HumanRacesinRelation
toEnvironment
andCulture,
with
totheInfluence
ofCultureuponGenetic
SpecialReference
ColdSpring
HarborSymposium
ChangesinHumanEvolution.
on Quantitative
Biology15:247-258.
1962 TheOriginofRaces.NewYork:Knopf.
1965 TheLivingRacesofMan.NewYork:Knopf.
1981 Adventures
andDiscoveries.
TheAutobiography
ofCarleton
S. Coon. Englewood
Cliffs,
NJ:Prentice-Hall.
Coon,C. S., S. M. Garn,andJ.B. Birdsell
1950 Races:AStudyoftheProblems
ofRaceFormation
inMan.
IL:C. C. Thomas.
Springfield,
Dixon,RolandB.
1923 TheRacialHistory
ofMan.NewYork:Scribner.
75
Theodosius
Dobzhansky,
1944 On Speciesand RacesofLivingandFossilMan.American
ofPhysical
2:251-265.
Journal
Anthropology
1962 MankindEvolving:
oftheHumanSpecies.
TheEvolution
NewHaven,CT:YaleUniversity
Press.
1963 Possibility
ThatHomosapiens
EvolvedIndependently
5
TimesIsVanishingly
Small.Current
4:360-366.
Anthropology
Garn,StanleyM.
1957 RaceandEvolution.
American
59:218-224.
Anthropologist
1962 HumanRaces.Rev.edition.Springfield,
IL:C. C. Thomas.
Garrett,
HenryE.
1961 TheEqualitarian
inBiologyandMediDogma.Perspectives
cine4:480-484.
Gasman,Daniel
1971 TheScientific
ofNationalSocialism:
SocialDarwinOrigins
isminErnst
HaeckelandtheGermanMonistLeague.NewYork:
Elsevier.
Gates,R.Ruggles
1944 Phylogeny
andClassification
ofHominidsandAnthropoids.
ofPhysical
2:279-292.
AmericanJournal
Anthropology
George,WesleyC.
1962 TheBiologyoftheRaceProblem.
DC: National
Washington,
PutnamLetters
Committee.
Giles,Eugene
1999 Hooton,Earnest
Albert.In
American
NationalBiography,
vol. 11.Pp.147-149.John
A.GaratyandMarkC. Carnes,eds.
NewYork:OxfordUniversity
Press.
Haeckel,Ernst
1905 TheWondersofLife.NewYork:Harper.
Harding,RosalindM.
2000 Diversity,
NotDivergence.
GeneticHistory
ofModernHumans.8thCEPHAnnualConference,
Facultede MedicineLariboisiereSaint-Louis,
Paris,May,24-26,2000.TrendsinGenetics
16:381.
Hawks,John,KeithHunley,Sang-HeeLee,and MilfordH. Wolpoff
2000 Bottlenecks
andPleistocene
HumanEvolution.
MolecularBiologyandEvolution17:2-22.
LawrenceA.
Hirschfeld,
1996 RaceintheMaking:Cognition,
Culture
andtheChildsConstruction
ofHumanKinds.Cambridge,
MA:MITPress.
1998 Natural
Race,EssenceandTaxonomiesofHuAssumptions:
manKinds.SocialResearch
65(2):331-350.
LawrenceA.,and SusanA. Gelman,eds.
Hirschfeld,
1994 MappingtheMind:DomainSpecificity
inCognitionand
Culture.
NewYork:University
ofCambridge
Press.
Hooton,EarnestA.
1931 Upfrom
theApe.NewYork:MacMillan.
1936 PlainStatements
aboutRace.Science83:511-513.
1937 Apes,Men,andMorons.NewYork:G. P.Putnam.
1939 Twilight
ofMan.NewYork:G. P.Putnam.
Howells,WilliamW.
1942 FossilManandtheOriginofRaces.American
Anthropologist44:182-193.
1959 MankindintheMaking.GardenCity,NJ:Doubleday.
1993 Getting
Here.TheStory
ofHumanEvolution.
Washington,
DC: CompassPress.
S.
Hulse,Frederick
1962 Raceas an Evolutionary
Episode.American
Anthropologist
64:929-945.
C. Haddon
Huxley,JulianS.,and Alfred
1935 WeEuropeans:
ASurvey
of"Racial"Problems.
London:
Jonathan
Cape.
Jackson,
JohnP.,Jr.
2001 "InWaysUnacademical":
TheReception
ofCarletonS.
Coon'sTheOrigin
oftheHistory
ofBiology
Journal
ofRaces.
34:247-285.
Keith,Arthur
1936 History
fromCaves.ANewTheoryoftheOriginofModem
RacesofMankind.London:British
Association
PubSpeleological
lication.
RichardC.
Lewontin,
1972 TheApportionment
ofHumanDiversity.
BiolEvolutionary
ogy6:381-398.
76
Leonard
Lieberman,
2001 How"Caucasoids"GotSuchBigHeadsandWhyThey
FromMortontoRushton.
Shrank:
Current
Anthropology
42:69-95.
Linnaeus,Carolus
1758 Systemae
Naturae.10thedition.Stockholm:
Laurentii
Salvii.
F. B.
Livingstone,
1962 On theNon-Existence
ofHumanRaces.Current
Anthropology3:279-381.
Mayr,Ernst
1982 TheGrowth
ofBiological
and
Evolution,
Thought:Diversity,
MA:BelknapPressofHarvard
Inheritance.
UniverCambridge,
sityPress.
1991 One LongArgument:
CharlesDarwinandtheGenesisof
ModernEvolutionary
MA:HarvardUniverThought.Cambridge,
sityPress.
TheodosiusDobzhansky,EthelTobach,and
Mead,Margaret,
RobertE. Light,eds.
1968 ScienceandtheConceptofRace.NewYork:ColumbiaUniPress.
versity
Montagu,Ashley
1942 Man'sMostDangerousMyth:TheFallacyofRace.NewYork:
Press.
ColumbiaUniversity
Montagu,Ashley,ed.
1964 TheConceptofRace.NewYork:FreePress.
Morgan,LewisHenry
MA:Harvard
Cambridge,
Society.
University
1964[1877] Ancient
Press.
Putnam,Carleton
1962 RaceandReason:AYankeeView.Washington,
DC: Public
Affairs
Press.
Relethford,
J.H.
1998 GeneticsofModemHumanOrigins
andDiversity.
Annual
27:1-23.
ReviewofAnthropology
1 of 4
R.C. Lewontin, Alexander Agassiz Professor Emeritus of Zoology at Harvard University, has written a number of books and
articles on evolution and human variation, including Biology as Ideology: The Doctrine of DNA and The Triple Helix: Gene,
Organism, and Environment
Over the last thirty five years a major change has taken place in our biological understanding of the concept
of human race, largely as a consequence of an immense increase in our knowledge of human genetics. As
a biological rather than a social construct, race has ceased to be seen as a fundamental reality
characterizing the human species. Nevertheless, there appear from time to time claims that racial
categories represent not arbitrary socially and historically defined groups but objective biological divisions
based on genetic differences. The most recent widely noticed rebirth of such claims is an essay by Armand
Marie Leroi on the Op-Ed page of The New York Times (March 14, 2005), an essay that illustrates both the
classical confusions about the reality of racial categories and the more recent erroneous conclusions about
the relevance of such racial identifications for medical practice.
There are four facts about human variation upon which there is universal agreement. First, the human
species as a whole has immense genetic variation from individual to individual. Any two unrelated human
beings differ by about 3 million distinct DNA variants.
Second, by far the largest amount of that variation, about 85%, is among individuals within local national
or linguistic populations, within the French, within the Kikuyu, within the Japanese. There is diversity
from population to population in how much genetic variation each contains, depending upon how much
immigration into the population has occurred from a variety of other groups and also on the size of the
population. The United States, with a very large population whose ancestors came from all over the earth
including the original inhabitants of the New World, is genetically very variable whereas small populations
of local Amazonian tribes are less genetically variable, although they are by no means genetically uniform.
Despite the differences in amount of genetic variation within local populations, the finding that on the
average 85% of all human genetic variation is within local populations has been a remarkably consistent
result of independent studies carried out over twenty-five years using data from both proteins and DNA.
Of the remaining 15% of human variation, between a quarter and a half is between local populations within
classically defined human races, between the French and the Ukrainians, between the Kikuyu and the
Ewe, between the Japanese and the Koreans. The remaining variation, about 6% to 10% of the total human
variation is between the classically defined geographical races that we think of in an everyday sense as
identified by skin color, hair form, and nose shape. This imprecision in assigning the proportion of
variation assigned to differences among population within races as compared to variation among races,
2 of 4
arises precisely because there is no objective way to assign the various human populations to clear-cut
races. Into which race do the Hindi and Urdu speakers of the Indian sub-continent fall? Should they be
grouped with Europeans or with Asians or should a separate race be assigned to them? Are the Lapps of
Finland and the Hazari of Afghanistan really Europeans or Asians? What about Indonesians and
Melanesians? Different biologists have made different assignments and the number of races assigned by
anthropologists and geneticists has varied from 3 to 30.
Third, a small number of genetic traits, such as skin color, hair form, nose shape (traits for which the genes
have not actually been identified) and a relatively few proteins like the Rh blood type, vary together so that
many populations with very dark skin color will also have dark tightly curled hair, broad noses and a high
frequency of the Rh blood type R0. Those who, like Leroi, argue for the objective reality of racial divisions
claim that when such covariation is taken into account, clear-cut racial divisions will appear and that these
divisions will correspond largely to the classical division of the world into Whites, Blacks, Yellows, Reds
and Browns. It is indeed possible to combine the information from covarying traits into weighted averages
that take account of the traits' covariation (technically known as "principal components" of variation).
When this has been done, however, the results have not borne out the claims for racial divisions. The
geographical maps of principal component values constructed by Cavalli, Menozzi and Piazza in their
famous The History and Geography of Human Genes show continuous variation over the whole world
with no sharp boundaries and with no greater similarity occurring between Western and Eastern
Europeans than between Europeans and Africans! Thus, the classically defined races do not appear from
an unprejudiced description of human variation. Only the Australian Aborigines appear as a unique group.
A clustering of populations that does correspond to classical continental "races" can be acheived by using a
special class of non-functional DNA, microsatellites. By selecting among microsatellites, it is possible to
find a set that will cluster together African populations, European populations, and Asian populations, etc.
These selected microsatellite DNA markers are not typical of genes, however, but have been chosen
precisely because they are "maximally informative" about group differences. Thus, they tell us what we
already knew about the differences between populations of the classical "races" from skin color, face shape,
and hair form. They have the added advantage of allowing us to make good estimates of the amount of
intermixture that has occurred between populations as a result of migrations and conquests.
The every-day socially defined geographical races do identify groups of populations that are somewhat
more closely similar to each other genetically. Most important from the standpoint of the biological
meaning of these racial categories, however, most human genetic variation does not show such "race"
clustering. For the vast majority of human genetic variations, classical racial categories as defined by a
combination of geography, skin color, nose and hair shape, an occasional blood type or selected
microsatellites make no useful prediction of genetic differences. This failure of the clustering of local
populations into biologically meaningful "races" based on a few clear genetic differences is not confined to
the human species. Zoologists long ago gave up the category of "race" for dividing up groups of animal
populations within a species, because so many of these races turned out to be based on only one or two
genes so that two animals born in the same litter could belong to different "races."
In his article, Leroi is inconsistent and shifting in his notion of race. Sometimes it corresponds to the
classical social definitions of major races, but elsewhere he makes race coincident with a small local
3 of 4
group such as the Negritos or Inuit. In this shifting concept of race he goes back to the varying use of the
term in the 19th century. Then people spoke of the Scots race, the Irish race and the race of
Englishmen. Indeed race could stand for a family group defined by male inheritance, as in the
description of the last male in a family line as the last of his race. This inconsistent usage arises from the
fact that there is no clear criterion of how much difference between groups of genetically related
individuals should correspond to the category race. If it had turned out that groups of related populations
were clearly different in the great majority of their genes from other groups, then racial categories would be
clear and unambiguous and they would have great predictive power for as yet unstudied characters. But
that is not the way it has turned out, at least for the human species.
The fourth and last fact about genetic differences between groups is that these differences are in the
process of breaking down because of the very large amount of migration and intergroup mating that was
always true episodically in the history of the human species but is now more widespread than ever. The
result is that individuals identified by themselves or others as belonging to one race, based on the small
number of visible characters used in classical race definitions, are likely to have ancestry that is a mixture
of these groups, a fact that has considerable significance for the medical uses of race identification.
A common claim, repeated by Leroi, is that racial categories are of considerable medical use, especially in
diagnostic testing because some genetic disorders are very common in ancestral racial populations. For
example sickle cell anemia is common among West Africans, who were brought as slaves to the New
World, and Tay-Sachs disease is common among Ashkenazi Jews. So, it is argued, racial information can
be a useful diagnostic indicator. Certainly classical race contains some medically relevant information in
some cases, as for example white as opposed to African American if the contrast is between Finland and
West Africa, but not if it is a contrast between a white Mediterranean and an Asian Indian. There is a
confusion here between race and ancestry. Sickle cell anemia is in high frequency not only in West Africans
but also in some white Middle Eastern and Indian populations. Moreover, a person with, say, one African
great-grandparent, but who is identified by herself and others as white has a one in eight chance of
inheriting a sickle-cell mutation carried by that ancestor. There are, in addition, a number of other simply
inherited hemoglobin abnormalities, the thalassemias, that are in high frequency in some places in the
Mediterranean (Sardinia), Arabia and southeast Asia. The highest frequency known for a thalassemia
(80%) is in Nepal, but it is rare in most of Asia. The categorization of individuals simply as white or
Afro-American or Asian will result in a failure to test for such abnormal hemoglobins because these
abnormalities do not characterize the identified race of the patient. Even group identities below the level
of the conventional races are misleading. Two of my incontrovertibly WASP grandchildren have a single
Ashenazi Jewish great-grandparent and so have a one in eight chance of inheriting a Tay-Sachs
abnormality carried by that ancestor. For purposes of medical testing we do not want to know whether a
person is Hispanic but rather whether that persons family came from a Caribbean country such as Cuba,
that had a large influx of West African slaves, or one in which there was a great deal of intermixture with
native American tribes as in Chile and Mexico, or one in which there was only a negligible population of
non-Europeans. Racial identification simply does not do the work needed. What we ought to ask on
medical questionnaires is not racial identification, but ancestry. Do you know of any ancestors who were
(Ashkenazi Jews, or from West Africa, from certain regions of the Mediterranean, from Japan)? Once
again, racial categorization is a bad predictor of biology.
4 of 4
There has been an interesting dialectic between the notion of human races and the use of race as a general
biological category. Historically, the concept of race was imported into biology, and not only the biology of
the human species, from social practice. The consciousness that human beings come in distinct varieties
led, in the history of biology, to the construction of race as a subgrouping within species. For a long time
the category race was a standard taxonomic level. But the use of race in a general biological context
then reinforced its application to humans. After all, lots of animal and plant species are divided into races,
so why not Homo sapiens? Yet the classification of animal and plant species into named races was at all
times an ill-defined and idiosyncratic practice. There was no clear criterion of what constituted a race of
animals or plants that could be applied over species in general. The growing realization in the middle of the
twentieth century that most species had some genetic differentiation from local population to local
population led finally to the abandonment in biology of any hope that a uniform criterion of race could be
constructed. Yet biologists were loathe to abandon the idea of race entirely. In an attempt to hold on to the
concept while make it objective and generalizable, Th. Dobzhansky, the leading biologist in the study of the
genetics of natural populations, introduced the geographical race, which he defined as any population
that differed genetically in any way from any other population of the species. But as genetics developed and
it became possible to characterize the genetic differences between individuals and populations it became
apparent, that every population of every species in fact differs genetically to some degree from every other
population. Thus, every population is a separate geographic race and it was realized that nothing was
added by the racial category. The consequence of this realization was the abandonment of race as a
biological category during the last quarter of the twentieth century, an abandonment that spread into
anthropology and human biology. However, that abandonment was never complete in the case of the
human species. There has been a constant pressure from social and political practice and the coincidence
of racial, cultural and social class divisions reinforcing the social reality of race, to maintain race as a
human classification. If it were admitted that the category of race is a purely social construct, however, it
would have a weakened legitimacy. Thus, there have been repeated attempts to reassert the objective
biological reality of human racial categories despite the evidence to the contrary.
1 of 10
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1893020/
PMCID: PMC1893020
Department of Human Genetics, University of Utah Health Sciences Center, Salt Lake City, Utah 84112, Department of Anthropology, University of
Utah, Salt Lake City, Utah 84112, McDermott Center for Human Growth and Development, University of Texas Southwestern Medical Center, Dallas,
Texas 75390 and Department of Biological Sciences, Louisiana State University, Baton Rouge, Louisiana 70803
1Corresponding author: Department of Human Genetics, Eccles Institute of Human Genetics, University of Utah, 15 N. 2030 E., Room 7225, Salt Lake
City, UT 84112-5330. E-mail: lbj@genetics.utah.edu
Communicating editor: L. E
Abstract
Go to:
The proportion of human genetic variation due to differences between populations is modest, and individuals
from different populations can be genetically more similar than individuals from the same population. Yet
sufficient genetic data can permit accurate classification of individuals into populations. Both findings can be
obtained from the same data set, using the same number of polymorphic loci. This article explains why. Our
analysis focuses on the frequency, , with which a pair of random individuals from two different populations is
genetically more similar than a pair of individuals randomly selected from any single population. We compare
to the error rates of several classification methods, using data sets that vary in number of loci, average allele
frequency, populations sampled, and polymorphism ascertainment strategy. We demonstrate that classification
methods achieve higher discriminatory power than because of their use of aggregate properties of populations.
The number of loci analyzed is the most critical variable: with 100 polymorphisms, accurate classification is
possible, but remains sizable, even when using populations as distinct as sub-Saharan Africans and Europeans.
Phenotypes controlled by a dozen or fewer loci can therefore be expected to show substantial overlap between
human populations. This provides empirical justification for caution when using population labels in biomedical
settings, with broad implications for personalized medicine, pharmacogenetics, and the meaning of race.
DISCUSSIONS of genetic differences between major human populations have long been dominated by two facts:
(a) Such differences account for only a small fraction of variance in allele frequencies, but nonetheless (b)
multilocus statistics assign most individuals to the correct population. This is widely understood to reflect the
increased discriminatory power of multilocus statistics. Yet B
et al. (2004) showed, using multilocus
statistics and nearly 400 polymorphic loci, that (c) pairs of individuals from different populations are often more
similar than pairs from the same population. If multilocus statistics are so powerful, then how are we to
understand this finding?
All three of the claims listed above appear in disputes over the significance of human population variation and
race. In particular, the A
A
A
(1997, p. 1) stated that data also show
that any two individuals within a particular population are as different genetically as any two people selected from
5/14/2016 11:46 PM
2 of 10
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1893020/
any two populations in the world (subsequently amended to about as different). Similarly, educational material
distributed by the H
G
P
(2001, p. 812) states that two random individuals from any one
group are almost as different [genetically] as any two random individuals from the entire world. Previously, one
might have judged these statements to be essentially correct for single-locus characters, but not for multilocus
ones. However, the finding of B
et al. (2004) suggests that an empirical investigation of these claims is
warranted.
In what follows, we use several collections of loci genotyped in various human populations to examine the
relationship between claims a, b, and c above. These data sets vary in the numbers of polymorphic loci
genotyped, population sampling strategies, polymorphism ascertainment methods, and average allele frequencies.
To assess claim c, we define as the frequency with which a pair of individuals from different populations is
genetically more similar than a pair from the same population. We show that claim c, the observation of high ,
holds with small collections of loci. It holds even with hundreds of loci, especially if the populations sampled
have not been isolated from each other for long. It breaks down, however, with data sets comprising thousands of
loci genotyped in geographically distinct populations: In such cases, becomes zero. Classification methods
similarly yield high error rates with few loci and almost no errors with thousands of loci. Unlike , however,
classification statistics make use of aggregate properties of populations, so they can approach 100% accuracy
with as few as 100 loci.
MATERIALS AND METHODS
Go to:
data sets were used. Loci or individuals with >10% missing data were not included in any data
set (loci were pruned first and then individuals). The first data set (insertions) consists of 175 polymorphic
transposable element insertion loci (100 Alu and 75 L1) previously genotyped in 259 individuals. The population
sample consists of 104 individuals from sub-Saharan Africa, 54 East Asians, 61 individuals of northern European
ancestry, and 40 individuals from Andhra Pradesh, India (W
et al. 2005; W
et al. 2006). The
second data set (microarray) consists of 9922 biallelic single-nucleotide polymorphism (SNP) loci genotyped in
278 individuals (55 Africans, 42 African Americans, 40 Native Americans, 22 Indians, 20 East Asians, 62
Europeans, 18 HispanoLatinos from Puerto Rico, and 19 individuals from New Guinea). This data set is derived
from that of S
et al. (2005). The third data set (resequenced) is derived from the 10 ENCODE regions of
the HapMap project, release 16c.1 of phase I, June 2005 (I
H M C
2005). These
regions were resequenced in 48 individuals to identify SNPs without ascertainment bias in favor of loci with
common polymorphisms. These SNPs were then genotyped in 209 unrelated individuals: 60 Yoruba in Ibadan,
Nigeria (YRI); 60 Utah residents with ancestry from northern and western Europe (CEU, from the CEPH
diversity panel); and 89 Japanese in Tokyo, Japan, plus Han Chinese in Beijing, China (CHB + JPT). Our subset
consists of 14,258 SNPs. All markers in all three data sets are biallelic. The proportions of missing genotypes are
2.4, 2.1, and 0.36%, respectively.
Data subsampling: To
examine the effect of population sampling (i.e., the effects of comparing relatively isolated
populations vs. more closely related or admixed ones), two subsets were constructed from each of the insertions
and microarray main data sets: one consisting of the entire data set, with all its labeled populations, and another
consisting of East Asian, European, and sub-Saharan African population groups only. The resequenced data set
consists only of the latter three population groups.
To investigate the effect of allele frequency, these five data subsets were subdivided according to three further
treatments: loci with common polymorphisms (with minor allele frequency, MAF, > 0.1); loci with rare
polymorphisms (MAF < 0.1); and all polymorphic loci, regardless of frequency. Henceforth we refer to these
classes of loci as rare polymorphisms, common polymorphisms, or all polymorphisms. For this classification,
allele frequencies were computed across the entire sample in the parent data set. To investigate the effect of
incrementally increasing the number of loci used, loci from each of these 15 data subsets were sampled (without
5/14/2016 11:46 PM
3 of 10
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1893020/
replacement) to produce 200 independent data sets with numbers of loci varying in 21 steps on a logarithmic scale
from 10 to the maximum.
Dissimilarity fraction
: Let
This method is inherently limited to dividing individuals into just two clusters using only biallelic loci, so we
limit our definitions to that situation. Consider individuals sampled from two populations, A and B, and
genotyped at many biallelic loci. At each locus, we identify the allele whose frequency is higher in population A
and assign it a value of 0. The other allele (more frequent in B than in A) is assigned a value of 1. Let qij represent
the genotype of individual i at locus j, defined as the average of the assigned values of the two alleles carried by
that individual at that locus. Now define qi as the average of qij over all loci j (so qi is a polygenic quantitative
genetic trait). Given these definitions, if populations A and B are typified by even slightly different allele
frequencies at many loci, then qi will usually be smaller for a member of population A than for a member of
population B. Thus the value of the trait qi indicates membership in one population or the other, so we call qi the
population trait value of individual i.
Individuals are assigned to population A or B depending on whether their population trait value qi falls below or
above some dividing criterion qC, respectively. In the case of just two populations, these assignments are
5/14/2016 11:46 PM
4 of 10
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1893020/
compared to the known origins of the individuals, and the proportion misclassified is reported as CT. The
classification criterion qC is chosen as follows. Let
be the mean of qi taken over all individuals in population
A, and define
similarly for population B. If the distributions of qi for individuals from the two populations are
symmetric with equal variance, then letting qC = ( + )/2 minimizes misclassification (cf. R
et al. 2002;
E
2003). To better account for unequal variances, we generalize slightly and solve for a criterion qC such
that r(qC) = s(qC) and
< qC <
where r and s are normal probability density functions with means and
variances estimated from the distributions of qi for populations A and B, respectively.
To extend this inherently pairwise approach to more than two populations, assignments for each individual are
initially computed with reference to each possible pair of populations. The values (0 or 1) assigned to particular
alleles, the criterion qC, and all qi are calculated anew for each pair of populations. Individuals are finally
assigned to a population only if they were assigned to it in all pairwise comparisons involving that population.
The proportion of individuals misclassified (or not classified, since this method can fail to classify individuals) is
reported as CT. For comparison, a single-locus classification error rate is computed by using this method to
classify individuals using each locus singly and then averaging the results over all loci.
RESULTS
Go to:
The statistics
CC, and CT are closely related by design. To illustrate the relationships
between them, the distributions of the genetic measures that underlie them are shown in Figure 1. For simplicity,
only two populations (Europeans and sub-Saharan Africans) and 50 typical loci randomly chosen from the
insertions data set are used. The distributions of pairwise genetic distances for within- and between-population
pairs of individuals (Figure 1A) overlap considerably even for these geographically isolated populations. The
dissimilarity fraction,
is 20%, indicating that between-population pairs are more similar than within-population
pairs one-fifth of the time. In contrast, the distributions of individuals' distances to the centroids of their own or
different populations (Figure 1B) show much less overlap, resulting in CC = 4.2%. The population trait value
distributions for Africans and Europeans overlap for just three individuals, yielding CT = 1.8%. Classifications
using model-based methods such as Structure (P
et al. 2000) achieve 90% accuracy or better using the
same data (B
et al. 2003; W
et al. 2006).
Distributions of distances:
F
1.
Frequency distributions of the underlying genetic measures used to
compute
CC, and CT, for a subset of 50 loci genotyped in 104
sub-Saharan African and 61 European individuals of the insertions data set.
The measures shown are (A) 13,530 pairwise genetic ...
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1893020/figure/fig1/
The variances of the distributions are much greater for the individual-to-individual comparisons (Figure 1A) than
for the centroid-to-individual comparisons (Figure 1B). The distribution means are nearly identical, however, so
the distributions overlap more in Figure 1A than in 1B, and thus > CC. The difference in variances is due to the
fact that each genetic distance to a centroid (each datum in Figure 1B) is equivalent to the average of a sizable
subset of pairwise genetic distances represented in Figure 1A (see
). That averaging step
eliminates considerable variation and produces the narrower distributions of Figure 1B.
5/14/2016 11:46 PM
5 of 10
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1893020/
For example, an African individual x with qx = 0.52 will be more similar to a European y with qy = 0.60 than to
another African z with qz = 0.4. Yet that individual x will still be closer to the population mean trait value for
Africans (qA 0.48, the African centroid) than to the mean value of Europeans (qB 0.68). It follows that many
individuals like this one will be correctly classified (yielding low CC and CT) even though they are often more
similar to individuals of the other population than to members of their own population (yielding high ).
To empirically and quantitatively understand the relationships and contrasts between and the misclassification
rates CC and CT, we examine three primary factors that influence them: the number of polymorphic loci used, the
allele frequencies at those loci, and the degree of differentiation between the populations examined.
data sets, labeled insertions, microarray, and resequenced, were used, and 15 subsets
were constructed from these to examine the effects of different data collection strategies (see
). Table 1 lists the 15 data subsets and reports
CC, and CT (each computed over all loci in each data
subset) as well as the expected value of CT when only a single locus is used. Table 1 also gives values of five
descriptive statistics for each data subset: the proportion of genetic variance explained by interpopulation
differences (FST); the observed proportions of heterozygotes (% het); the absolute differences in allele
frequencies between population pairs (averaged), ; the fraction of polymorphisms that are rare (MAF < 0.1) in at
least one population and at the same time common (MAF > 0.1) in another population (% rare and common); and
the fraction of loci that are monomorphic in at least one population and common polymorphisms in another (%
fixed and common). The values observed are typical of human population genetic data sets (N 1973; D
et
al. 1994; I
H M C
2005; S
et al. 2005; W
et al. 2006).
TABLE 1
Data set descriptions and summary results
Figure 2 shows the dependency of
CC, and CT on the number of loci for
each of the 15 data subsets listed in Table 1. As the number of included loci increases,
CC and CT decrease.
This is the expected behavior for CC (S
and C
1986; M
et al. 2002; C
et al.
2003) and CT (R
et al. 2002; E
2003). However, does not decrease nearly as rapidly. Figure 2A
shows the results for a diverse sample of individuals genotyped at 175 insertion loci, a number that is typical of
many studies of human genetic diversity published during the last decade. The downward trend in is apparent,
but even with the full data set it remains at 15% (with all four population groups; Table 1). Across all data sets
and using <100 polymorphisms, generally exceeds 10% (Figure 2). With <100 loci, then, it will often be the
case that two individuals from different populations are more similar to one another than are two individuals from
the same population.
Dependency of
on number of loci:
F
2.
Behavior of the dissimilarity fraction ( ) and error rates of the centroid
(CC) and population trait (CT) classification methods (red, blue, and
green lines, respectively) for each of 15 data subsets (see Table 1 and ...
The power of large numbers of common polymorphisms is most apparent in the microarray data set, comparing
the European, East Asian, and sub-Saharan African population groups (Figure 2C). approaches zero (median
0.12%) with 1000 polymorphisms. This implies that, when enough loci are considered, individuals from these
population groups will always be genetically most similar to members of their own group. In general, CC and CT
decrease more rapidly and to lower values than
rare polymorphism subsets defy this trend by converging toward high values of
5/14/2016 11:46 PM
6 of 10
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1893020/
as loci are added. This is largely because the frequencies of rare polymorphisms are necessarily quite similar
across populations, whereas higher-frequency polymorphisms have the potential to differ more. For example, the
frequency of an allele with an overall MAF of 5% can differ by at most = 10% between two populations (absent
in one, at 10% frequency in another). This situation yields > 0 and very poor classification accuracy, since most
between-population pairs are identical but some within-population pairs differ. In contrast, an allele with an
overall frequency of 50% across two populations could be fixed in one and absent in the other, resulting in = 0
and allowing perfect classification. It is these frequency differences that allow populations to be distinguished, so
the data sets with lower (and thus generally lower FST) have lower classification power.
The sensitivity of these statistics to allele frequencies explains some differences between the data sets. The
microarray data set exhibits strong ascertainment bias for common polymorphisms, and it is with this data set that
drops most rapidly and to its lowest values (Figure 2, C and D). The insertions data set exhibits a weaker
ascertainment bias and includes more rare polymorphisms, so remains higher (Figure 2, A and B). Similarly,
CC and CT drop more rapidly for the microarray data set than for the insertion data set. The resequenced data set
polymorphisms were ascertained by resequencing a sizable panel of individuals from the genotyped populations
and thus include many rare polymorphisms, but this is partially offset by the equally large number of common
polymorphisms (Figure 2E). The classification methods are less affected by the inclusion of rare polymorphisms.
contrast two choices: sets of populations that have been relatively isolated from
each other by geographic distance and barriers since the earliest migrations of modern humans out of Africa and
sets that include populations that were founded more recently, are geographically closer to one another and
therefore more likely to exchange migrants, or have recently experienced a large genetic influx from another
population in the set. Sampling only from the more distinct populations yields lower -values, as expected.
Figure 2, A, C, and E, shows the results of using only the three most distinct population groups (Europeans, East
Asians, and sub-Saharan Africans). Figure 2, B and D, expands the samples used in Figure 2, A and C, to include
recently founded and/or geographically intermediate populations (Indians in the insertions data set and New
Guineans, South Asians, and Native Americans in the microarray data set) and admixed populations (i.e., those
that have recently received many migrants from different populations, such as the African American and
HispanoLatino groups in the microarray data set). With just 175 loci, choosing to sample distinct populations vs.
more closely related ones makes only a modest difference (insertions data set, compare Figure 2A to 2B; Table 1).
The effect of population sampling becomes more pronounced when 1000 loci are available. In the microarray
data set, drops to zero at 1000 loci if only distinct populations are sampled. With geographically intermediate
and admixed populations added, however, reaches an asymptotic value of 3.1%, CC remains well above zero,
and even CT does not reach zero (microarray data, Figure 2, C and D; Table 1).
also appears to reach a nonzero asymptotic value in the resequenced data set, instead of continuing to trend
downward as would be expected given the distinct populations used. This may be due to the fact that many of the
polymorphisms in that data set are physically linked and therefore nonindependent. Overall, the responses of the
two classification methods to data set composition variables are qualitatively similar to the behavior of (
Figure 2). The most apparent difference is that the misclassification rates (CC and CT) decrease much more
rapidly, and to lower values, than does as the number of loci considered increases.
DISCUSSION
Go to:
It has long been appreciated that differences between human populations account for only a small fraction of the
total variance in allele frequencies (typically presented as FST values of 1015%; L
1972; N and
R
1972; L
1980; B
et al. 1997; J
et al. 2000; W
et al. 2003;
I
H M C
2005; R
et al. 2005). Such observations triggered controversy
from the outset. Some geneticists concluded the differences were negligible (L
1972); others disagreed
(M
1978). Despite the limited data, it soon became apparent that even a modest number of loci should allow
5/14/2016 11:46 PM
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1893020/
7 of 10
1978; S
et al. 1982).
Thus the answer to the question How often is a pair of individuals from one population genetically more
dissimilar than two individuals chosen from two different populations? depends on the number of
polymorphisms used to define that dissimilarity and the populations being compared. The answer, can be read
from Figure 2. Given 10 loci, three distinct populations, and the full spectrum of polymorphisms (Figure 2E), the
answer is
0.3, or nearly one-third of the time. With 100 loci, the answer is 20% of the time and even using
1000 loci,
10%. However, if genetic similarity is measured over many thousands of loci, the answer becomes
never when individuals are sampled from geographically separated populations.
On the other hand, if the entire world population were analyzed, the inclusion of many closely related and
admixed populations would increase
This is illustrated by the fact that and the classification error rates, CC
and CT, all remain greater than zero when such populations are analyzed, despite the use of >10,000
polymorphisms (Table 1, microarray data set; Figure 2D). In a similar vein, R
et al. (2002) and S
and P
(2004) have suggested that highly accurate classification of individuals from continuously sampled
(and therefore closely related) populations may be impossible. However, those studies lacked the statistical power
required to answer that question (see R
et al. 2005).
How can the observations of accurate classifiability be reconciled with high between-population similarities
among individuals? Classification methods typically make use of aggregate properties of populations, not just
properties of individuals or even of pairs of individuals. For instance, the centroid classification method computes
the distances between individuals and population centroids and then clusters individuals around the nearest
centroid. The population trait method relies on information about the frequencies of each allele in each population
to compute individual trait values and on the means and variances of the trait distributions to classify individuals.
The Structure classification algorithm (P
et al. 2000) also relies on aggregate properties of populations,
such as HardyWeinberg and linkage equilibrium. In contrast, the pairwise distances used to compute make no
use of population-level information and are strongly affected by the high level of within-groups variation typical
of human populations. This accounts for the difference in behavior between and the classification results.
5/14/2016 11:46 PM
8 of 10
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1893020/
Since an individual's geographic ancestry can often be inferred from his or her genetic makeup, knowledge of
one's population of origin should allow some inferences about individual genotypes. To the extent that
phenotypically important genetic variation resembles the variation studied here, we may extrapolate from
genotypic to phenotypic patterns. Resequencing studies of gene-coding regions show patterns similar to those
seen here (e.g., S
et al. 2001), and many common disease-associated alleles are not unusually
differentiated across populations (L
et al. 2006). Thus it may be possible to infer something about an
individual's phenotype from knowledge of his or her ancestry.
However, consider a hypothetical phenotype of biomedical interest that is determined primarily by a dozen
additive loci of equal effect whose worldwide distributions resemble those in the insertion data set (e.g., with =
0.15; Table 1). Given these assumptions, the genetic distance used in computing and CC is equivalent to a
phenotypic distance, so Figure 2 can be used to analyze this hypothetical trait. Figure 2A shows that a trait
determined by 12 such loci will typically yield = 0.31 (0.200.41) and CC = 0.14 (0.0540.29; medians and
90% ranges). About one-third of the time ( = 0.31) an individual will be phenotypically more similar to
someone from another population than to another member of the same population. Similarly, individuals will be
more similar to the average or typical phenotype of another population than to the average phenotype in their
own population with a probability of 14% (CC = 0.14). It follows that variation in such a trait will often be
discordant with population labels.
The population groups in this example are quite distinct from one another: Europeans, sub-Saharan Africans, and
East Asians. Many factors will further weaken the correlation between an individual's phenotype and their
geographic ancestry. These include considering more closely related or admixed populations, studying
phenotypes influenced by fewer loci, unevenly distributed effects across loci, nonadditive effects, developmental
and environmental effects, and uncertainties about individuals' ancestry and actual populations of origin. The
typical frequencies of alleles that influence a phenotype are also relevant, as our results show that rare
polymorphisms yield high values of
CC, and CT, even when many such polymorphisms are studied. This
implies that complex phenotypes influenced primarily by rare alleles may correspond poorly with population
labels and other population-typical traits (in contrast to some Mendelian diseases). However, the typical
frequencies of alleles responsible for common complex diseases remain unknown. A final complication arises
when racial classifications are used as proxies for geographic ancestry. Although many concepts of race are
correlated with geographic ancestry, the two are not interchangeable, and relying on racial classifications will
reduce predictive power still further.
The fact that, given enough genetic data, individuals can be correctly assigned to their populations of origin is
compatible with the observation that most human genetic variation is found within populations, not between
them. It is also compatible with our finding that, even when the most distinct populations are considered and
hundreds of loci are used, individuals are frequently more similar to members of other populations than to
members of their own population. Thus, caution should be used when using geographic or genetic ancestry to
make inferences about individual phenotypes.
Acknowledgments
Go to:
References
Go to:
We thank Jinchuan Xing, Michael Bamshad, Dennis O'Rourke, and Thomas Doak for thoughtful comments. This
work was supported by National Science Foundation grants BCS-0218338 (M.A.B.), BCS-0218370 (L.B.J.), and
EPS-0346411 (M.A.B.); by National Institutes of Health grant GM-59290 (L.B.J. and M.A.B.); by the Louisiana
Board of Regents Millennium Trust Health Excellence Fund HEF (2000-05)-05 (M.A.B.), (2000-05)-01
(M.A.B.), and (2001-06)-02 (M.A.B.); and by the Intramural Research Program of the National Institute of
Diabetes and Digestive and Kidney Diseases, National Institutes of Health.
5/14/2016 11:46 PM
9 of 10
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1893020/
American Anthropological Association, 1997. Response to OMB directive 15: race and ethnic standards for
federal statistics and administrative reporting (original statement at http://web.archive.org
/web/19990507115624/http://www.ameranthassn.org/ombnews.htm; amended 2000 statement at
http://www.aaanet.org/gvt/ombdraft.htm).
Bamshad, M., S. Wooding, B. A. Salisbury and J. C. Stephens, 2004. Deconstructing the relationship between
genetics and race. Nat. Rev. Genet. 5: 598609. [PubMed]
Bamshad, M. J., S. Wooding, W. S. Watkins, C. T. Ostler, M. A. Batzer et al., 2003. Human population genetic
structure and inference of group membership. Am. J. Hum. Genet. 72: 578589. [PMC free article] [PubMed]
Barbujani, G., A. Magagni, E. Minch and L. L. Cavalli-Sforza, 1997. An apportionment of human DNA diversity.
Proc. Natl. Acad. Sci. USA 94: 45164519. [PMC free article] [PubMed]
Bowcock, A. M., A. Ruiz-Linares, J. Tomfohrde, E. Minch, J. R. Kidd et al., 1994. High resolution of human
evolutionary trees with polymorphic microsatellites. Nature 368: 455457. [PubMed]
Campbell, D., P. Duchesne and L. Bernatchez, 2003. AFLP utility for population assignment studies: analytical
investigation and empirical comparison with microsatellites. Mol. Ecol. 12: 19791991. [PubMed]
Chakraborty, R., and L. Jin, 1993. A unified approach to study hypervariable polymorphisms: statistical
considerations of determining relatedness and population distances. Exper. Suppl. 67: 153175. [PubMed]
Cornuet, J. M., S. Piry, G. Luikart, A. Estoup and M. Solignac, 1999. New methods employing multilocus
genotypes to select or exclude populations as origins of individuals. Genetics 153: 19892000.
[PMC free article] [PubMed]
Dean, M., J. C. Stephens, C. Winkler, D. A. Lomb, M. Ramsburg et al., 1994. Polymorphic admixture typing in
human ethnic populations. Am. J. Hum. Genet. 55: 788808. [PMC free article] [PubMed]
Edwards, A. W., 2003. Human genetic diversity: Lewontin's fallacy. BioEssays 25: 798801. [PubMed]
Human Genome Project, 2001. The human genome. Nature 409: 812.
International HapMap Consortium, 2005. A haplotype map of the human genome. Nature 437: 12991320.
[PMC free article] [PubMed]
Jorde, L. B., W. S. Watkins, M. J. Bamshad, M. E. Dixon, C. E. Ricker et al., 2000. The distribution of human
genetic diversity: a comparison of mitochondrial, autosomal, and Y-chromosome data. Am. J. Hum. Genet.
66: 979988. [PMC free article] [PubMed]
Lao, O., K. van Duijn, P. Kersbergen, P. de Knijff and M. Kayser, 2006. Proportioning whole-genome singlenucleotide-polymorphism diversity for the identification of geographic population structure and genetic
ancestry. Am. J. Hum. Genet. 78: 680690. [PMC free article] [PubMed]
Latter, B., 1980. Genetic differences within and between populations of the major human subgroups. Am. Nat.
116: 220237.
Lewontin, R. C., 1972. The apportionment of human diversity. Evol. Biol. 6: 381398.
Lohmueller, K. E., M. M. Mauney, D. Reich and J. M. Braverman, 2006. Variants associated with common
disease are not unusually differentiated in frequency across populations. Am. J. Hum. Genet. 78: 130136.
[PMC free article] [PubMed]
Manel, S., P. Berthier and G. Luikart, 2002. Detecting wildlife poaching: identifying the origin of individuals with
Bayesian assignment tests and multilocus genotypes. Conserv. Biol. 16: 650659.
Mitton, J. B., 1977. Genetic differentiation of races of man as judged by single-locus and multilocus analyses.
Am. Nat. 111: 203212.
Mitton, J. B., 1978. Measurement of differentiation: reply to Lewontin, Powell, and Taylor. Am. Nat. 112:
11421144.
Mountain, J. L., and L. L. Cavalli-Sforza, 1997. Multilocus genotypes, a tree of individuals, and human
evolutionary history. Am. J. Hum. Genet. 61: 705718. [PMC free article] [PubMed]
Mountain, J. L., and U. Ramakrishnan, 2005. Impact of human population history on distributions of
individual-level genetic distance. Hum. Genomics 2: 419. [PMC free article] [PubMed]
Nei, M., 1973. Analysis of gene diversity in subdivided populations. Proc. Natl. Acad. Sci. USA 70: 33213323.
5/14/2016 11:46 PM
10 of 10
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1893020/
5/14/2016 11:46 PM