. 2
( 8)


This reading of the status of being a girl or a boy would allow the
alien anthropologists to integrate another incident into their analysis,
one occurring among members of San Francisco™s transgendered com-
munity. Matt Califia-Rice was born and brought up as a girl. In his
twenties, however, he sought to move from a female assignment to a
male one and began to take testosterone. Shortly thereafter he had his
breasts removed. He also began to live with a former lover who had
also been born and raised as a girl and was now also living as a man,
taking testosterone and contemplating chest surgery. Reconstituted
as a couple, the two now wanted a baby. Since Califia-Rice™s partner
had already had a hysterectomy for health reasons, Califia-Rice him-
self decided to become pregnant. He stopped taking testosterone and
had himself impregnated. He and his partner, both living as men, one
of whom was now pregnant, resided in San Francisco. Yet, ˜˜within
that world they were a scandal . . . Real men,™™ one transgendered male
said, ˜˜don™t have babies.™™41
The connection of sex and gender with interests and activities
means that certain interests and activities pursued by the wrong sex
and gender necessarily raise concerns. Hence, if one has changed one™s
identity from that of a woman to that of a man, one simply ought not
to perform certain actions, even if one™s body is capable of performing
them and even if one wants to perform them. The transgendered
community includes those who have had sex reassignment surgery,
those who are contemplating some part of it, and those who plan to
remain pre-operative. Some in the transgendered community ˜˜take
hormones to change their secondary sex characteristics; some do not;
many dress and live as close to the traditional definition of male and
female as possible; others are androgynous.™™42 In this community,
then, anatomy is of less interest than it was for Bruce Reimer™s original
doctors. Moreover, differences in life-styles, interests, and activities
are met with more tolerance. Nevertheless, even within this

Mary McNamara, ˜˜Era of the Gender Crosser,™™ Los Angeles Times, February 27,
2001, Section A, p. 20.
Ibid., p. 20.
32 A F T E R I D E N T I T Y

community, tolerance does not always extend to men who have
babies. If one intends to bear a child and one has not been a man
since birth, then one should not claim to be a man now. The trans-
gendered community does not ask for commitment to all the activ-
ities or aspects of anatomy of one sex assignment or gender identity:
one can evidently decide whether or not to have breasts. Still, if one is
to be a man refraining from certain other activities, such as child-
bearing, appear to be de rigueur.

On one reading, status as a girl is more tightly woven with a specific
complex of interests and proclivities than status as a boy. Brian Reimer
was a quiet, gentle child who disliked fighting except with girls. While
he therefore seemed to deviate from some of the activities normative for
the male gender, his gender status was not at risk. In contrast, because
Brenda was rowdy, assertive, and dominant, her status as a girl was at
risk. Men can be men, it seems, however they act. Women can be
women only if they act in an appropriate way. But if status as a girl is
more tightly woven with interests and proclivities than status as a boy,
the latter is more tightly woven with appendages and capacities.
Because Bruce could not successfully perform the activities, both sex-
ual and non-sexual, that boys are supposed to be able to perform, he
could not be a boy. Indeed, although one can retain one™s status as a girl
even if one loses one™s clitoris, one cannot remain a boy if one loses
one™s penis. Likewise, one cannot retain one™s status as a man if one is
pregnant although one can remain a woman if one is not.
How, then, might our alien anthropologists articulate an under-
standing of sex and gender that takes account of these differences?
They might decide that a certain complex of appendages, interests,
activities, and proclivities is especially mandatory when it is neces-
sary for heterosexual intercourse and when an anatomy has been
re-manufactured after birth. Hence, because a penis is necessary to
vaginal penetration whereas a clitoris is not, the lack of a penis has a
greater impact on male sex and gender status than the lack of a clitoris
T H E T R A G E D Y O F D A V I D R E I M E R 33

does on female sex and gender status. Moreover, because Brian
retained the sex and gender assignment he had received at birth, his
failure to conform completely to gender norms could escape scrutiny,
whereas because Brenda™s sex and gender had been reassigned she
could not. Similarly, because Matt Califia-Rice was not born with
the anatomy of a man, his attention to gender norms had to be espe-
cially scrupulous and his waywardness in seeking to bear a child
placed his sex reassignment and gender re-identity at risk.
Nevertheless, the anthropologists who offered this account of
sex and gender attributions and identities would find themselves
stymied by legal decisions in various American states and in a variety
of foreign countries. Take two somewhat recent cases. In Littleton v.
Prange, the Fourth Circuit Court of Appeals in Texas ruled against
Christie Littleton in her suit against her late husband™s physician for
the medical treatment that she claimed led to his death.43 Christie
Littleton was once a man but received sex reassignment surgery prior
to her marriage. Her husband, Jonathan Littleton, was aware of the
surgery. Nor did the lawsuit reveal other troubles within the marriage;
it lasted seven years until Jonathan Littleton™s death. Christie
Littleton was heterosexual according to the standards of her new
anatomy and there was no reason to think that she did not embrace
the roles, behaviors, and interests linked to it. Yet, the court ruled that
she could not recover damages for the wrongful death of her husband
because her marriage to him violated the Texas Family Code prohibit-
ing same-sex unions. ˜˜Biologically,™™ the court wrote, ˜˜a post-operative
female transsexual is still a male.™™ The Littleton marriage was thus a
marriage between two men and had never been legal.
In a subsequent case, a Kansas court came to the same con-
clusion. In this instance, another ˜˜post-operative female transsexual,™™
J™Noel Ball, tried to defend her right to inherit from the estate of her
deceased husband, Marshall Gardiner. The Kansas Supreme Court,
however, denied that the marriage was legal because, despite her

Littleton v. Prange 9 SW 3d 223.
34 A F T E R I D E N T I T Y

sex-reassignment surgery, J™Noel was not a woman. ˜The words ˜˜sex,™
˜˜male™™ and ˜˜female™™ in everyday understanding do not encompass
transsexuals. The plain, ordinary meaning of ˜persons of the opposite
sex™ contemplates a biological man and a biological woman and not
persons who are experiencing gender dysphoria.™™44
What are our alien anthropologists to make of these two cases?
The difficulty they might have in integrating them into their overall
account of sex and gender identities is that both diverge from the ideas
that they might have elicited from the David Reimer and Matt Califia-
Rice cases. Christie Littleton and J™Noel Ball both had re-manufactured
anatomies and therefore, according to the account of male and female
status we have constructed, they would have had to adhere carefully to
the demands of their new identity. In contrast to David Reimer and Matt
Califia-Rice, however, there is no evidence that they did not do so; they
both married men and seemed to have helped to make their marriages a
success. Nevertheless, neither was accorded the status of women by the
courts. Why not? In its presentation of the Littleton case, the Texas court
asked, ˜˜Can a physician change the gender of a person with a scalpel,
drugs and counseling, or is a person™s gender immutably fixed by our
Creator at birth?™™45 The question was rhetorical for the court had no
hesitation in picking the second option: one™s identity as a man or
woman is determined by a person™s chromosomes and therefore remains
indifferent to the results of any genital reconstructive surgery. Yet, if
chromosomes determine gender status, this finding undermines the
basis for just the sort of surgery that David Reimer had as an infant.
Indeed, if we were to follow the Kansas and Texas courts, the Reimers™
attempt to make up for Bruce™s circumcision accident would have been
legally doomed from the start. He could never have been made into a girl
because his chromosomes dictated that legally he would always be a boy.
Operations that medical practitioners often perform on inter-
sexed infants also fail to adhere to the logic of the Texas and Kansas

In re Estate of Marshall G. Gardner 273 Kan. 191; 2002.
Littleton v. Prange, pp. 230“231.
T H E T R A G E D Y O F D A V I D R E I M E R 35

courts.46 Instances of intersexuality include Congenital Adrenal
Hyperplasia (CAH), in which infants are born with what appear to be
either very small penises or large clitorises;47 Turner Syndrome, in
which individuals possess female genitalia but unformed gonads;48
5-alpha-reductase deficiency, in which infants are born with what
appear to be mostly female characteristics but during adolescence
develop male external genitalia, deep voices and hair according to
˜˜male™™ patterns;49 and Klinefelder Syndrome, in which gonads do not
develop ˜˜properly.™™50 Infants can also be born with hypospadias, in
which the urethral opening is not at the tip of the penis.51 In other
cases clitorises may be considered too large (over 0.9 cm) while penises
may sometimes be seen as too small (under 2.5 cm). While some
pediatricians and intersexuals criticize surgical interventions on
infants, in its 2000 report, the American Academy of Pediatrics
(AAP) continued to recommend them.52 The point of such surgeries
is to fashion anatomies that would conform to prevailing norms for
male and female genital appearances and allow for standard male and
female activities. Part of the rationale for this kind of surgery is to help
parents: if they are not to be ambivalent in helping their children
acquire an appropriate gender identity they need to be comfortable
with their genital appearance and the supposition is that they can be
comfortable only if that appearance falls within standard parameters.
Hence, operations on infants with large clitorises reduce them; oper-
ations on infants with hypospadias draw skin from other parts of the
body to create tubes so that the children can urinate standing up; and
infants with micropenises are sometimes castrated and brought up as

See New York Times, September 19, 2004.
See Anne Fausto-Sterling, Sexing the Body, p. 52.
Fausto-Sterling, Sexing the Body, p. 52.
Fausto-Sterling, Sexing the Body, p. 109.
Fausto-Sterling, Sexing the Body, p. 52.
Domurat Dreger, Hermaphrodites, p. 39. Also, see Geoffrey Cowley, ˜˜Gender
Limbo,™™ Newsweek, May 19, 1997, pp. 64“67 on the case of Heidi, an infant with
one X and one Y chromosome and a small penis with the urethral opening at the
base. S/he was castrated and brought up as a girl.
See Committee on Genetics, ˜˜Evaluation of the Newborn With Developmental
Anomalies of the External Genitalia,™™ Pediatrics, 106, 2000, pp. 138“142.
36 A F T E R I D E N T I T Y

girls since their original penises are thought too small to allow for
vaginal penetration.53 Since surgeons often assign the sex that is most
likely to result in a successful surgery,54 and since it is easier to create
a vagina than a penis, most intersexed infants end up as girls.55
On the one hand, these cases seem to re-confirm the standards
for sex assignment and gender rearing that we expected our alien
anthropologists would infer from Bruce Reimer™s reassignment as
Brenda. One condition for being a boy is being able to urinate standing
up. Accordingly, if one is born with hypospadias either one™s urethra
must be refashioned so that one can do so or one must be brought up as
a girl. Similarly, one condition for being a girl is appropriately delicate
genitalia (or, in some cultures, none at all). Consequently, if one is
born with a clitoris that is too large, it must be reduced so that one™s
status as a girl cannot be questioned. Although assessments of the
success of operations are mixed, it is clear that they can lead to less
than perfect results, as well as to losses of sensation, painful scarring,
repeated infections, and the need for new operations or constant
vaginal dilations.56 One study found a failure rate of 64 percent in

Suzanne J. Kessler, Lessons from the Intersexed (New Brunswick, NJ: Rutgers
University Press, 1998), pp. 41“43. Also see Domurat Dreger, Hermaphrodites,
p. 195 and Julie A. Greenberg, ˜˜Defining Male and Female: Intersexuality and the
Collision between Law and Biology,™™ Arizona Law Review, 41, 1999, p. 272. In one
case, a surgeon decided against cutting off a micropenis, despite thinking that it
was too small for reproductive purposes, because it was so ˜˜well formed™™
(Fausto-Sterling, Sexing the Body, p. 59).
Kessler, Lessons from the Intersexed, p. 25.
In Diamond and Sigmunson, ˜˜Sex Reassignment at Birth,™™ Web-based version),
Milton Diamond cites two pieces of clinical advice. ˜˜Because it simpler to construct
a vagina than a satisfactory penis, only the infant with a phallus of adequate size
should be considered for a male gender assignment™™ (from J. W. Duckett and
L. S. Baskin, ˜˜Genitoplasty for Intersex Anomalies,™™ European Journal of Pediatrics,
152, 1993 Suppl. 2, p. 580). ˜˜It is easier to make a good vagina than a good penis and
since . . . the absence of an adequate penis would be psychosexually devastating,
fashion the perineum into normal looking vulva and vagina and raise the individual
as a girl.™™ As at least one surgeon has also remarked, ˜˜You can make a hole but you
can™t build a pole™™ (Fausto-Sterling, Sexing the Body, p. 59).
See Kessler, Lessons from the Intersexed, pp. 53“64, for an out-and-out condemnation
of the operations. Another assessment is more mixed. See Claude J. Migeon et al.,
˜˜Ambiguous Genitalia with Perineoscrotal Hypospadias in 46, XY Individuals:
Long-Term Medical, Surgical, and Psychosexual Outcome,™™ Pediatrics, 110 (3),
2002, www.pediatrics.org/cgi/content/full/110/3/c31.
T H E T R A G E D Y O F D A V I D R E I M E R 37

attempts to repair hypospadias.57 Nonetheless, because identity as a
boy or girl, woman or man depends upon a set of capacities, activities,
and proclivities, the AAP continues to see intersexed conditions as
˜˜social emergencies™™58 and recommends operations to correct them.
Results of clitoral surgery are mixed. For example, preservation of nerve conduction
in the neurovascular bundle of the phallus was reported after excision of the
corporeal bodies in infants with ambiguous genitalia, although long-term sexual
function remains to be investigated in these patients. A second article reported an
excellent cosmetic and functional outcome after clitoral recession; however, an
unwanted outcome of clitoral necrosis can occur. When cosmetic outcomes of
several types of clitoral surgeries were considered together (recession, reduction,
and amputation), the post-surgical appearance of the genitalia were considered to be
poor by Creighton et al.: The young age of some of the participants in these
studies makes it difficult to interpret the functional significance of the findings. In
addition, some of the above-mentioned studies used measures of cosmetic outcome
that were determined by the investigators, not by the patients themselves.
Follow-up studies of vaginoplasty are also limited in number. A large study was
conducted in women with mullerian agenesis or Mayer“Rokitansky“Kuster“Hau-
¨ ¨
ser syndrome (also referred to as Rokitansky syndrome) by Rock et al.: In this
group of patients, the McIndoe vaginoplasty procedure was rated to be successful in
terms of post-surgical vaginal depth for sexual activity by all women. Women who
had CAH as a result of 21-hydroxylase deficiency and underwent a McIndoe pro-
cedure reported a lower success rate (62 percent) in terms of comfortable penovagi-
nal intercourse. Outcome studies of the McIndoe procedure in women with
complete androgen insensitivity syndrome reported satisfactory intercourse post-
operatively in 72 percent of patients and orgasm in 78 percent. In a similar group of
patients with complete androgen insensitivity syndrome, good sexual function in
terms of patients™ satisfaction with their genitalia (78 percent), satisfactory libido (71
percent), and orgasm (77 percent) were reported by study participants. Creighton
et al. reported the results of several vaginoplasty procedures in girls and young
women who were affected by a variety of urogenital abnormalities resulting in
ambiguous genitalia. In these patients, the vaginal introitus was absent or small in
82 percent, vaginal length was inadequate in 27 percent, and additional vaginal
procedures were required in 75 percent.
Investigations of masculinizing surgeries, like feminizing surgeries, in 46,XY
intersex individuals with perineoscrotal hypospadias are limited. The only study to
have evaluated exclusively the surgical outcome of the most severe cases of
hypospadias reported on nineteen men, approximately half of whom experienced
difficulties with micturition, urologic function, and ejaculation. Roughly one-third
of patients were affected by marked impairment in quality of life resulting from
their ambiguous genitalia, ranging from mild depression to severe psychiatric im-
pairment. Another follow-up assessment of hypospadias repair in adults included
eight men with perineoscrotal hypospadias. In all cases, multiple hypospadias rep-
airs were attempted with a post-operative complication rate of 64 percent. Repeated
surgical procedures and complications are of particular concern because of scarring
and loss of tissue associated with each surgery, as well as the presumed negative
impact on sexual function.
See Migeon et al. ˜˜Ambiguous Genitalia,™™ p. e31.
Committee on Genetics, ˜˜Evaluation of the Newborn,™™ p. 138.
38 A F T E R I D E N T I T Y

Ironically, however, if we follow the Kansas and Texas courts
only those surgeries that align the female sex with two X chromo-
somes and those that align the male sex with an X and a Y chromo-
some will pass muster for one™s legal identity. In states and countries
with rules similar to those of Kansas and Texas, when surgically
treated intersexuals become adults, they will be able to select as
marriage partners only those individuals whose chromosomes do not
match their own, regardless of their reconstructed anatomies or the
gender in which they have been raised. In some cases, then, if they are
to marry at all, they will have to marry those whose chromosomes
differ but whose genitalia and gender identity match their own. Given
the Texas and Kansas marriage codes, which explicitly preclude gay
and lesbian marriages, this result seems somewhat bizarre. Then
again, it is surely a welcome turn of events for those couples who
may have been barred from marrying each other because of their
anatomies and should now be able to marry because of their
And what of another ˜˜condition?™™ Androgen Insensitivity
Syndrome (AIS) is a condition in which individuals possess one X
and one Y chromosome but are born with female-like genitalia and
at puberty develop breasts because their bodies cannot ˜˜read™™ or proc-
ess the testosterone in their bodies. According to Natalie Angier,
such individuals are often models since their androgen insensitivity
gives them a tall stature, large breasts, and beautiful hair and skin.
Indeed, she refers to them as ˜˜mama mia™™ women.60 If this is so, a

Since the Fourth District™s ruling in the Littleton case, Bexar County has legiti-
mated at least two such marriages, over public protest. (See John Gutierrez-Mier,
˜˜Two More Women Obtain County Marriage License,™™ San Antonio Express News,
September 21, 2000.) Greenberg notes three more such marriages in Oregon, Ohio
and England in which women were allowed to marry male-to-female transsexuals
because the latter are still legally men. (Greenberg, ˜˜Defining Male and Female,™™
p. 268.)
Natalie Angier, Woman: An Intimate Geography (New York: Anchor Books Edn.,
2000), p. 34.
T H E T R A G E D Y O F D A V I D R E I M E R 39

chromosomal criterion for status as a woman will not only undermine
the legal point of some surgeries on so-called intersexuals; it may also
prevent certain Kansas and Texas men from marrying the super-
models of their dreams.
When we follow the medical profession through its examina-
tion of Brenda Reimer and its operations on intersexed infants, we
learn to understand sex and gender assignments in connection with a
more or less tightly woven complex of anatomy, roles, behaviors,
interests, and sexual orientation. Intersexuality and accidental cas-
trations can be corrected by surgeries that align anatomy with rec-
ognizably male and female forms together with psychosexual
counseling that helps develop the roles and gender identity appropri-
ate to one of two strictly delineated kinds of bodies. Something
similar holds for ˜˜gender dysphoria™™; it can be ˜˜cured™™ by creating
an anatomy to fit the individual™s sense of his or her sex and gender.
When we follow Texas and Kansas state law, in contrast, we learn
that intersexuality, accidental castrations, and gender dysphoria
cannot be cured in this way because one remains what one™s chromo-
somes are no matter who one thinks one is, what one is capable
of, or what one likes to do. Suppose, then, that David Reimer had
complied with his surgeons and psychologists in adapting to the
female gender roles and behavior they wanted for him. Brenda
Reimer would have been unable to marry a man in Kansas, Texas,
or a set of other countries and American states. Nor would the
ambiguities of her legal status have been limited to marriage. It is
not clear that Brenda would have been able to claim sex discrimina-
tion in employment in the United States if she were discriminated
against as a woman. Instead, courts might have concluded that
although she was discriminated against as a woman, since chromo-
somally he was a man, Title VII did not apply. And what about
certain insurance benefits, the right to be incarcerated with members
of one™s own sex, and the requirement that men register for the draft?
Should AIS supermodels register for the draft? Should they be incar-
cerated with men?
40 A F T E R I D E N T I T Y

The Kansas and Texas decisions on marriage followed similar
rulings in New York and Ohio, both of which at least sometimes tie
sex to chromosomes at birth. New Jersey, on the other hand, looks to
what it terms psychological sex.61 In the case of M. T. v. J. T, the New
Jersey court ruled that ˜˜For marital purposes, if the anatomical or
genital features of a genuine transsexual are made to conform to the
person™s gender, psyche or psychological sex, then identity by sex
must be governed by the congruence of these standards.™™62 As the
Kansas Supreme Court itself conceded, England and Australia have
also decided transsexual marriage cases in terms of a harmonization
of body sex and psychological gender.
Courts are not only divided on what counts as male and female
for the purposes of marriage. They are also divided on the issue with
regard to other questions. For instance, Ohio and Oregon deny trans-
sexuals a right to change their sex designation on their birth certif-
icates while for the past twenty-five years New York City has
allowed them to do so if they have had sex-reassignment surgery.63
Likewise, the United States allows transsexuals to change their
sex designations on their passports on the condition that they pro-
vide proof that they have undergone or are about to undergo sex-
reassignment surgery.64 In 1977, the United States Tennis
Association (USTA) sought to prevent Renee Richards, a male-to-
female transsexual, from competing in the United States Open
Tennis Tournament as a woman. Using the Barr chromosome test,
the USTA claimed that Richards remained a man and therefore could
not compete in the women™s division. Richards sued and the New

See Greenberg, ˜˜Defining Male and Female,™™ pp. 301“302.
M. T. v. J. T. 140 NJ Super 77 (1976).
In December 2006, the New York City Board of Health unanimously rejected a
recommendation to allow transsexuals to change their sex designation on their
birth certificates with or without sex-reassignment surgery. (See Damien Cave,
˜˜New York Plans to Make Gender Personal Choice,™™ New York Times, November
7, 2006, pp. A1, A21.)
See Greenberg, ˜˜Defining Male and Female,™™ p. 315.
T H E T R A G E D Y O F D A V I D R E I M E R 41

York Supreme Court found in her favor, ruling Richards a woman for
the purposes of tennis competition. ˜˜The requirement,™™ the court
said, ˜˜that this plaintiff pass the Barr body test in order to be eligible
to participate in the women™s singles of the United States Open is
grossly unfair, discriminatory and inequitable, and violative of her
rights under the Human Rights Law of this State.™™65 Nevertheless,
in the 1988 Olympics, officials from the International Olympics
Committee (IOC), conducting similar Barr tests on athletes on the
various women™s teams, disqualified Maria Martinez Patino, a mem-
ber of the Spanish women™s hurdling team. Although Martinez
Patino had always considered herself a female, a cotton swab of
material from the inside of her cheek found that she possessed XY
chromosomes. Further examination found testes hidden by her labia
and neither a uterus nor ovaries. Consequently, Olympic officials
declared her a man, barred her from the women™s competition and
stripped her of all her previous medals.66 Martinez Patino went to
court in Spain and eventually, after two years of inactivity in which
she lost her national scholarship, her athletic residence, her coach
and her boyfriend, the IOC reinstated her status as a woman, appa-
rently on the basis of her shoulder structure.67 Shoulder structure
does not seem to be a widespread measure of female status but, then
again, no measure appears to have general clout. Instead, as this
review of cases and decisions indicates, medical and legal profes-
sions disagree with one another, the courts of different states and
countries disagree, and different parts of the medical profession
make different decisions about sex and gender identity. We can
chart these disagreements:

Richards v. United States Tennis Association 93 misc. 2d 713 (1977).
See Susan K. Cahn, Coming on Strong: Gender and Sexuality in Twentieth Century
Women™s Sport (Cambridge, MA: Harvard University Press, 1994), p. 264 and
Colette Dowling, The Frailty Myth: Women Approaching Physical Equality (New
York: Random House, 2000), pp. 176“178.
Fausto-Sterling, Sexing the Body, pp. 1“2.
42 A F T E R I D E N T I T Y

Female Heterosexual
Candidates XX Anatomy interests, orientation as
for female chromo- without a behaviors, measured by
status somes penis and roles anatomy Final status

Bruce No Yes No No Woman
Brenda No Yes No No Man
Brian No No Yes Yes Man
Second No Yes No No Alloyed
Money woman
Littleton No Yes Yes Yes Man
Ball No Yes Yes Yes Man
Martinez No Yes No Unknown Man (first
Patino IOC)
No Yes No Unknown Woman
court and
Richards No Yes No Yes Woman
AIS model No Yes Yes Yes Woman

Were alien anthropologists to focus on the operations performed on
Bruce Reimer, they might assume that one™s identity as a human boy
or girl, man or woman depends on one™s anatomy. If one has a penis,
one is a boy or man and if one does not have one, one is a girl or woman.
Since Bruce Reimer did not have a penis he could not be a boy. Yet,
were the anthropologists then to look at psychological assessments of
Brenda Reimer, they might come to think that one™s identity as a girl
or boy also involves one™s interests and activities. Since Brenda did not
T H E T R A G E D Y O F D A V I D R E I M E R 43

have the correct interests and behavior, she could not be a girl. One
must have not only the right anatomy but the right interests, desires,
and behaviors as well. Since Bruce had the wrong anatomy for his
interests and Brenda the wrong interests for her anatomy, alien
anthropologists might assume that David eventually corrected the
problem by manufacturing a penis. If so, however, they would be
confused by the sex and gender identity that courts assigned to
J™Noel Ball and Christine Littleton since both also aligned their anat-
omies with their interests and behavior. While David Reimer was
accepted as a man, neither Ball nor Littleton was legally recognized
as a woman. The anthropologists might then move on to a chromoso-
mal criterion: David Reimer was a man because of his chromosomes;
Ball and Littleton were also men because of their chromosomes.
Nevertheless, this conclusion would raise the question of what doc-
tors thought they were doing when they first operated on Bruce
Reimer and it would also fly in the face of common perceptions of
AIS women.
Alien anthropologists would also remain confused by the differ-
ences in the assumptions behind psychological skepticism about
Brenda Reimer™s femininity, the USTA™s decision on Renee
Richards, and the IOC™s decision on Martinez Patino. To the extent
that the female sex and gender are identified with such activities as
cleaning house, we can assume that Olympic athletes will fail to
conform. Indeed, Colette Dowling argues that the lack of conformity
is the real motivation behind sex testing of female athletes. It is not
what it is advertised to be: namely, to prevent genetic males from
cheating by joining female sports teams and thereby skewing the
competitions in their favor.68 When a genetic male did try to cheat
in this way on behalf of Germany in the 1936 Olympics, three women
beat him.69 When Helene Mayer won the US national fencing title in

Dowling, The Frailty Myth, p. 175.
Fausto-Sterling, Sexing the Body, p. 2. Hermann Ratgent was bitter about the
experience: ˜˜For three years I lived the life of a girl. It was most dull,™™ (Dowling,
The Frailty Myth, p. 178).
44 A F T E R I D E N T I T Y

1938, the governing body of the sport banned competition between
men and women and revoked Mayer™s title.70 When Zhang Shan won a
mixed shooting event in 1992, the Olympic committee decided to
divide the event into separate male and female competitions for the
next Olympics, thereby guaranteeing that no woman could beat a man
again.71 Accordingly, Dowling suggests that the point of the sex test-
ing refers back to conceptions of sex and gender. Because an interest in
exercise and competition is considered a male interest, athletic
women seem so anomalous that they must prove that they really are
women.72 Genetic testing in sports thus reflects the same assump-
tions that lead Colapinto to question the gender of a woman who has a
blue-collar job. Although Martinez Patino and Richards ultimately
won the right to be labeled as women at least for athletic purposes,
Brenda™s similar inability to conform to the interests and behaviors of
girls meant that she could not really be one.
For his part, Brian Reimer is a boy because he possesses male
chromosomes, a male anatomy, and a heterosexual orientation
although, like Brenda, he lacks the interests, behaviors, and roles
suitable for the status attributed to him. Our anthropologists might
think, then, that a heterosexual orientation, as measured by one™s
anatomy and ˜˜confirmed™™ by one™s chromosomes, is the key to sex
and gender. In other words, if one lacks a penis, possesses two X
chromosomes, and is attracted to those with a penis, one counts as a
woman. Yet, this solution excludes Money™s unnamed second patient
and all lesbians who cannot count as women under the definition. If
the anthropologists were to drop the requirement of a heterosexual
orientation, they could include lesbians but would still have to
exclude Christie Littleton and J™Noel Ball as well as every XY super-
model. If they were to drop the chromosomal requirement, they could

70 71
See Cahn, Coming on Strong, p. 210. See Dowling, The Frailty Myth, p. 193.
Ibid., pp. 179“180, Fausto-Sterling, Sexing the Body, p. 3. Dowling quotes Laura
Wakwitz: ˜˜Sex testing is not an issue of how tall a woman is or what percentage of
her mass is composed of muscle; it is an attempt to maintain control over women
who challenge the expectations of femininity by entering a stereotypically defined
˜male™ arena™™ (The Frailty Myth, pp. 179“180).
T H E T R A G E D Y O F D A V I D R E I M E R 45

include Littleton, Ball, and XY supermodels within the female gender,
but might have to exclude Brian Reimer from the male one.
To be sure, it may seem more than a little hyperbolic for alien
anthropologists to highlight such irregularities in our sex and gender
identifications. After all, the irregularities refer to comparatively rare
cases of ablated penises, intersexuality, and voluntary changes in sex
and gender assignment. We might therefore try to allay the confusions
of our anthropologists by assuring them that the cases causing their
perplexity are few in number and constitute gray areas in otherwise
clear conceptions. Indeed, we could assure them that the gray areas in
our sex and gender attributions and identities are really no different
than similar gray areas in our other attributions and identities. It may
not always be clear, for example, who is to count as a genius. While
some investigators may consider a numerical score on an IQ test
sufficient for determining whether or not someone is a genius, other
observers might want to look at other indices, such as performances
and accomplishments, and still others might find IQ tests completely
irrelevant. Moreover, those who agree that a certain score on an IQ test
indicates status as a genius might disagree about whether to attribute
the status of genius to those who score just under it. Yet neither this
imprecision in the conception of a genius nor possible disagreements
at its edges affects our ability to employ the designation in most cases.
Similarly, it is a mistake to think that we need to agree on completely
clear standards or a set of necessary and sufficient conditions for
identity as a man or a woman. The line between identity as a genius
and identity as a non-genius is often vague. Different people will draw
it at different points for different purposes and most will allow for
ambiguous cases. The same arguably holds for those on the edges of
our sex and gender categories. Hence, while our alien anthropologists
might be unclear about the boundaries of the set of men and the set of
women, we could reassure them to be comfortable with their under-
standing of its center.
Yet, while our reassurances are surely relevant to the question of
sex and gender identities, our anthropologists might not be satisfied.
46 A F T E R I D E N T I T Y

In the first place, they might wonder how infrequent cases of penis
ablation, ambiguous genitalia, or transsexuality really are. It is diffi-
cult to calculate cases of intersexuality in part for the reason we have
been looking at: namely, what constitutes unambiguous identity as a
girl or boy is unclear and medical professionals can therefore differ in
what they take intersexuality to be. Indeed, some doctors count only
cases in which infants are born with ambiguous genitalia whereas
others include cases of odd sorts of hair growth and male pattern
baldness in women.73 All researchers seem to agree, however, that
cases of intersexuality are more numerous than those outside the
medical profession assume, perhaps as high as 4 percent of the world™s
population.74 The same ambiguity surrounds identifications of trans-
sexuality: transsexuals, medical professionals, and legal and political
authorities may all define it quite differently, depending on whether
they think surgery is necessary to secure status as a man or woman
and, if it is, what sort is necessary and to what extent.
In the second place, the anthropologists might point out that
identities as geniuses and identities as women or men behave quite
differently. Identities and identifications as geniuses do not come with
expectations about behaviors, roles, preferences, and sexual interest
whereas identities and identifications as men and women do. One
does not walk or urinate as a genius. Nor is there a particular set of
interests that geniuses are meant to have, activities they are meant to
enjoy, or roles they are meant to perform. The failure to have the
appropriate sexual interests does not mean that one is not a genius.
To be sure, we often presume that geniuses will be less than capable in
practical matters and one might argue that their identities therefore do
come with expectations. It is because they are geniuses, because they
are preoccupied with deeper and more abstruse matters, that we can
expect them to forget to put gas in their cars or to leave the coffee pot
on, for example. Yet, if a genius does remember to put gas in his or her

See Fausto-Sterling, Sexing the Body, p. 52“3.
See Greenberg, ˜˜Defining Male and Female,™™ p. 267.
T H E T R A G E D Y O F D A V I D R E I M E R 47

car or to turn off the coffee, we do not therefore doubt the validity of
his or her attributed identity as a genius. In contrast, because Brenda
liked guns and because Money™s unnamed second patient had a blue-
collar job some observers did doubt the validity of Brenda™s attributed
identity as a girl and the second patient™s attributed identity as a
woman, respectively.
Taking their cue from the different ways that identities as men
and women and identities as geniuses behave, then, our anthropolo-
gists could point out to us that the former are much more rigid than
the latter. Indeed, they could point out that we simply do not allow
gray areas to remain gray areas in the case of sex and gender. Courts
and medical authorities do not take it upon themselves to issue deter-
minative rulings on who is and is not a genius. Instead, our culture
allows different institutions and associations to use different criteria
for genius status as befits their different functions. Moreover, a person
can be ˜˜sort of a genius,™™ or a genius in a certain way. One cannot be
sort of a woman or a man in a certain way. Conversely, we sometimes
say that a certain woman is ˜˜like a man™™ but not that a certain woman
is ˜˜like a genius.™™
It might seem remarkable to our alien anthropologists that dif-
ferent courts and different medical establishments differ in which sex
and gender identities they attribute to which individuals. Yet, perhaps
it would be no more remarkable to them than the circumstance that
these institutions make legally or surgically binding sex and gender
decisions at all. If our answer to the aliens™ perplexity is to be the
reference to gray areas, then it would seem to follow that we should
give those gray areas the same status they have for other identities. We
would think it medically unethical and even insane to break and re-set
a child™s legs if they had almost but not quite enough turn out to allow
him or her to be a ballet dancer. Why is it not equally unethical and
insane to castrate a child to if his anatomy almost but not quite allows
him or her to be a girl?
In chapters 2 and 3, I want to look at sorts of identities that are
perhaps closer in their history and characteristics to men and women
48 A F T E R I D E N T I T Y

than geniuses and ballet dancers are. In the recent past of the United
States, racial and ethnic identities had the same relation to law that
gender identities possess now. Contemporary debates consider their
relation to medicine. We might then look at our racial identities to see
if they can offer our anthropologists any clues to our sex and gender
identities. In chapter 2, I consider the quandaries that racial identities
have caused in the history of the United States. In chapter 3, I propose
a way of thinking about racial identities that I hope will transfer to the
case of sex and gender identities.
Racial identification and

Attributions of identities as a man sometimes depend on the presence
of a penis; sometimes they depend on the possession of XY chromo-
somes; in one instance, they required not having a baby. Similarly,
attributions of identities as a woman are sometimes contingent upon
capacities and proclivities, sometimes they look to sexual orientation,
and at least once they were linked to shoulder structure. What remains
constant in these various standards for sex and gender identity is their
association with some part of some set of behaviors, roles, and prefer-
ences, including sexual ones. What is inconstant is that these parts
and sets vary. Racial and ethnic status in the United States famously
possesses the same sort of variation. I shall therefore begin this chapter
with what W. E. B. Du Bois called the ˜˜exasperations of race,™™1 to see
what help they may be in considering exasperations of sex and gender.

Americans have been puzzling over their racial attributions for a very
long time. In suits for freedom by slaves before the Civil War, in
prosecutions for miscegenation between whites and non-whites after
the Civil War, and in racial prerequisite cases from 1789 until 1952,
states and federal courts had to determine whether particular individ-
uals were black, white, American Indian, or whatever. Until the ˜˜one-
drop rule™™ became widespread after the Civil War,2 different states
employed different standards to decide the issue. Some insisted that

See W. E. B. Du Bois (1897), ˜˜The Conservation of Races™™, in Robert Bernasconi and
Tommy L. Lott, eds., The Idea of Race (Indianapolis, IN: Hackett, 2000, pp. 108“117),
p. 109.
See Randall Kennedy, Interracial Intimacies: Sex, Marriage, Identity and Adoption
(New York: Pantheon Books, 2003), p. 223.
50 A F T E R I D E N T I T Y

one was black if one-fourth of one™s total ˜˜blood™™3 was of African
descent while others were satisfied with one-sixteenth of one™s
blood. In Virginia, one could be white with 24 percent black ancestry
until 1910.4
Courts also disagreed in how to apply state standards to individ-
ual cases. Ian F. Haney Lopez relates the divisions in the Virginia
Supreme Court in 1806 when it had to decide whether the Wright
family should be freed from slavery on the basis of a misidentification
of its race.5 In 1806, race in Virginia was a matter of maternal descent.
The Wrights claimed they were descendants of an American Indian
woman, Butterwood Nan, and, as such, after at the very latest 1705
they could not be legally enslaved.6 Challenging this claim, their
owner argued that Butterwood Nan had been a ˜˜negro™™ and that what-
ever American Indian heritage the Wrights possessed ran through the
male line. Hence the family members were legally his slaves. While all
the courts agreed with the Wrights, different judges at different judi-
cial levels had different reasons for their conclusions. The Chancellor
of the High Court of Chancery used color as the index of race, noting
˜˜that the youngest of the appellees was perfectly white and that there
were gradual shades of difference in colour between the grand-mother,
mother and grand-daughter.™™7 Judge Roane of the Virginia Supreme
Court cited ˜˜the general reputation and opinion of the neighbour-
hood™™8 with regard to Butterwood Nan™s daughter, Hannah Wright.

See, for example, Jones v. The Commonwealth, Gray v. Commonwealth Supreme
Court of Virginia 80 Va. 538 (1885). ˜˜If his [the accused™s] mother was a yellow
woman with more than half of her blood derived from the white race, and his father
a white man, he is not a negro. If he is a man of mixed blood he is not a negro, unless
he has one-fourth at least of negro blood in his veins,™™ p. 544.
Kennedy, Interracial Intimacies, p. 223.
Hudgins v. Wrights Supreme Court of Virginia 11 Va. 134 (1806). See Ian F. Haney
Lopez, ˜˜The Social Construction of Race: Some Observations on Illusion, Fabrication,
and Choice,™™ Harvard Civil Rights“Civil Liberties Law Review, 29, 1994.
There was some confusion on this point because Virginia had allowed for the enslave-
ment of American Indians brought into the colony between 1679 and 1691 or 1705.
The judges seemed to conclude that the age of the youngest Wright ruled out this
possibility. See Counsel in Hudgins v. Wrights.
7 8
Hudgins v. Wrights, cited in ˜˜Prior History™™. Hudgins v. Wrights, p. 142.
R A C I A L I D E N T I F I C A T I O N A N D I D E N T I T Y 51

Hannah Wright was now also dead but many of her neighbors had
thought that she was an Indian and had repeatedly urged her to sue for
her freedom. Hence, according to Judge Roane, she probably was an
Indian and would have been granted her freedom had the times in
which she lived ˜˜been as just and liberal on the subject of slavery as
the present.™™9 Judge Tucker, however, rejected both skin color and
reputation and relied instead on hair texture. Indeed, he wrote, ˜˜So
pointed is this distinction between the natives of Africa and the
aborigines of America that a man might as easily mistake the glossy,
jetty clothing of an American bear for the wool of a black sheep, as the
hair of an American Indian for that of African, or the descendent of an
Fortunately for the Wrights the testimony indicated that Hannah
had possessed long black hair. Yet, straight hair was not enough for
Thomas Gary in Arkansas in 1858. Under Arkansas law of the time one
was a ˜˜negro™™ if one had more than one-sixteenth ˜˜African blood.™™
When the sixteen-year-old Gary sued for freedom, three doctors testi-
fied that he did not. Dr. Brown found no trace of ˜˜negro blood™™ in his
eyes, nose, mouth, or jaw, noting in addition that his hair was ˜˜smooth
and of sandy complexion, perfectly straight and flat with no indication

Hudgins v. Wrights, p. 142.
Hudgins v. Wrights, p. 140; Haney Lopez, ˜˜The Social Construction of Race,™™ p. 2.
Regardless of which physical features it took to be dispositive in racial rulings,
Hudgins v. Wrights established the evidence of ˜˜inspection™™ as the foundation of
racial ascriptions in Virginia. See Hook v. Nanny Pagee and Her Children Supreme
Court of Virginia 16 Va. 379 (1811) and Gregory v. Baugh Supreme Court of Virginia
29 Va. 665 (1831). In North Carolina, the Supreme Court considered the question of
who was to be qualified to perform such inspections and resolved the issue in State
v. Asa Jacobs Supreme Court of North Carolina, Raleigh 51 NC 284 (1859). It did not
require ˜˜a distinguished comparative anatomist to detect the admixture of the
African or Indian with the pure blood of the white race,™™ the court declared.
Instead, just as ˜˜persons accustomed to observe the habits of a certain kind of fish
have been permitted to give in evidence their opinions as to the ability of the fish to
overcome certain obstructions in the rivers . . . any person of ordinary intelligence,
who, for a sufficient length of time, will devote his attention to the subject, will be
able to discover, with almost unerring certainty, the adulteration of the Caucasian
with the Negro or Indian blood™™ (State v. Asa Jacobs, p. 287).
52 A F T E R I D E N T I T Y

of the crisp or negro curl.™™11 Dr. Wilcox conceded that he could not say
that there was ˜˜no negro blood™™ in Gary. Yet, since his eyes were blue,
˜˜his hair straight and light, his complexion sandy™™ no such ˜˜blood™™ was
to be discerned ˜˜from external appearance.™™12 For his part, Dr. Dibbrell
thought that Gary might have a ˜˜small amount of negro blood, not more
than a sixteenth, perhaps not so much™™ and ˜˜would not positively swear
that he had any at all, so vague are the signs of the admixture of the
negro race, in one so remotely removed from the African blood by
crossing with the white.™™ Indeed, he admitted that he had ˜˜no definite
rule™™ and knew of no ˜˜reliable one™™ by which to judge cases such as
Despite such ˜˜expert™™ opinion, the Arkansas Supreme Court
ruled against Gary™s suit for freedom. Its justification lay in the race
of the woman, Susan, who it determined was his mother. Although
she had ˜˜a very light complexion™™ and straight hair, she had never
objected to her enslavement; moreover, she was swarthy with ˜˜rather
thick lips and coarse features.™™13 These facts were sufficient ˜˜to repel
any presumption in freedom in favor of the complainant, even upon
the supposition that the evidence, otherwise, left it as a matter of grave
doubt, whether he belonged to the white or negro race.™™ Indeed, doubt
was ˜˜the utmost that could be claimed for him™™ and more than what
the court thought he was ˜˜entitled to,™™ for, as it continued, no one
could read the evidence and come to the conclusion that it made it
appear that ˜˜he belongs to the white race, or descended from that race
on his mother™s side.™™14
For its part, although the Virginia Supreme Court relied on hair
texture in the Wrights case, it did not always do so. In 1877, a lower
court found Rowena McPherson and George Stewart guilty of ˜˜illicit
intercourse™™ even though they were husband and wife. According to

Gary v. Stevenson 19 Ark. 580 (1858) p. 583. Also see Jason A. Gilman, ˜˜Suing for
Freedom: Interracial Sex, Slave Law and Racial Identity in the Post-Revolutionary
and Antebellum South,™™ North Carolina Law Review, January 2004, p. 538.
Gary v. Stevenson, p. 583.
Gary v. Stevenson, p. 585; Gilman, ˜˜Suing for Freedom,™™ p. 608.
Gary v. Stevenson, pp. 586“587.
R A C I A L I D E N T I F I C A T I O N A N D I D E N T I T Y 53

the court, McPherson was a ˜˜negro™™ and hence her putative marriage
to Stewart, a white man, was illegal. By 1877, Virginia™s criterion for
belonging to the ˜˜negro™™ race no longer lay in the question of maternal
or paternal descent. Instead, it lay in the amount of ˜˜negro™™ blood one
had. Whether one inherited this blood from one™s male or female
ancestors, more than one-fourth of it made one a ˜˜negro.™™ Using this
criterion, the Virginia Supreme Court interpreted the ˜˜certificate of
facts™™ in the McPherson case to say that McPherson™s father was
white, that her great-grandfather was white and that her great-grand-
mother was ˜˜brown.™™ ˜˜It was said in the family,™™ the court went on,
˜˜that the . . . brown skin woman was a half-Indian, a fact which is
confirmed by the color of her skin.™™ Indeed, had McPherson™s great-
grandmother been a full-blooded ˜˜negro,™™ her skin would have been
black because the skin of full-blooded ˜˜Negroes™™ ˜˜is black and never
brown.™™ The court concluded that McPherson had ˜˜certainly derived
at least three-fourths of her blood from the white race.™™ Moreover, of
that fourth that was not derived from the white race some ˜˜residue™™
must have been American Indian and ˜˜if any part of the said residue of
her blood, however, small, was derived from any other source than the
African or negro race, then Rowena McPherson cannot be a negro.™™15
In Arkansas, then, blackness lay in the possession of more than
one-sixteenth ˜˜negro blood™™ while in Virginia it lay in a ˜˜negro™™ female
ancestor in a case in the beginning of the nineteenth century and in
more than one-fourth negro blood in a case at the end. The evidence of
˜˜blood,™™ for its part, lay in coarse features and failure to pursue one™s

McPherson v. The Commonwealth 69 Va. 939 (1877), p. 940: ˜˜It appears from [the
certificate of facts] ˜that her father was a white man; that her mother was also by a
white man, out of a brown skin woman; that Washington Goode, the half-uncle of
the said Rowena McPherson, testified that the said brown skin woman, who was his
grandmother and the great grandmother of said Rowena McPherson, told him that
she was a half-Indian; and that his mother, her daughter also told him the same.™ It
thus appears that less than one-fourth of her blood is negro blood . . . Besides having
certainly derived at least three-fourths of her blood from the white race, she derived a
portion of the residue from her great-grandmother, who was a brown skin woman,
and of course, not a full-blooded African or negro, whose skin is black, and never
brown.™™ Also see Kennedy, Interracial Intimacies, p. 224.
54 A F T E R I D E N T I T Y

freedom in Arkansas while in Virginia it lay in hair texture in the
beginning of the nineteenth century and in skin color towards the
end. An 1835 ruling in South Carolina took up Judge Roane™s idea in
the Wrights case and insisted that the evidence of one™s race lay in ˜˜the
general reputation and opinion of the neighbourhood.™™ Here, the South
Carolina Supreme Court declared that, ˜˜The condition of the individual
is not to be determined solely by a distinct and visible mixture of Negro
blood, but by reputation, by his reception into society and [by] his
having commonly exercised the privileges of a white man.™™16
Over a hundred years later, the 1940s case of Bennett v. Bennett
took the same position. The facts of this case mirror those of In re Estate
of Marshall G. Gardiner. In the Gardiner case, a nephew sued to
prevent his uncle™s estate from going to his wife, J™Noel Ball, because
she was really a man. In Bennett v. Bennett, Franklin Bennett™s daugh-
ter sued to prevent her father™s estate from going to his second wife,
Louetta Chassereau Bennett, because she was really a ˜˜Negro.™™ Louetta
Bennett fared better than J™Noel Ball, however, for while the court in the
Gardiner case looked to J™Noel™s chromosomes rather than what it
might have seen as her feminine and wifely behavior, the Bennett
court looked at Louetta™s behavior and the behavior of others with
regard to her. Hence, although the daughter alleged that ˜˜she had
more than an eighth of negro blood in her veins,™™17 the court noted that:

Upon the death of the defendant™s father and mother she was first
taken into the home of white people, then she was placed in a
church orphanage for white children, she was confirmed . . . as
a communicant of the Holy Communion Church of Charleston,
a white church; she was taken from the orphanage and placed in a
white home as a member of the family, and from there into another
white home as a member of that family; she married a white
man . . . she votes in the democratic primaries, both City and State,

Kennedy, Interracial Intimacies, pp. 227“228.
Bennett v. Bennett 10 S.E. 2d (SC 1940), p. 2. See also Kennedy, Interracial
Intimacies, pp. 226“227.
R A C I A L I D E N T I F I C A T I O N A N D I D E N T I T Y 55

whose rules bar negroes from voting; her children attend the white
public schools of Walterboro of which one of the plaintiff™s attor-
neys is a trustee.18

Randall Kennedy remarks on what might have been a sufficient
rationale for appealing to reputation in the Bennett case: ˜˜In the after-
math of a contrary holding, any white South Carolinian might have
felt compelled to peer into the mirror with a new intensity and ask
nervously, ˜Where will it end?™™™19 Judge Roane and the 1935 South
Carolina court may have had the same thought. Anyone™s family
history could be scoured for evidence of ˜˜Negro™™ blood and, moreover,
given the conditions of slavery, it might be quite easy to find it. Hence,
it would be far safer to rely on reputation.
Nevertheless, it is not clear what to make of the differences and
even contradictions in different courts™ assessments of the criteria and
evidence of race. One might think that the Southern states resolved
their differences after the Civil War with the one-drop rule that
counted as a black person anyone with one African ancestor. Yet, the
McPherson and Bennett cases both used different criteria of identity.
Moreover, since the solution requires deciding who has African
ancestry, it raises complexities of its own, as the Wrights and Gary
cases might have predicted,20 and it requires stipulating how far back
in a particular family tree to go in looking for evidence of African
ancestry.21 In addition, the solution simply shifts the problem of racial

18 19
Bennett v. Bennett, p. 5. Kennedy, Interracial Intimacies, p. 228.
See Hudgins v. Wrights and Gary v. Stevenson.
See State v. William Chavers 50 N.C. 11 (1857). Chavers was charged with carrying a
shot-gun in violation of the prohibition against free blacks doing so. His lawyer tried
to argue that Chavers was a white man, and hence entitled to his gun, because he
was five generations removed from a pure African ancestor. Instead, the court agreed
with the Brunswick Superior Court:

Take . . . two families, the father of one family a white person and the mother a
negro, and the father of the other family a negro and the mother a white woman;
the members of these families are of the half blood, and in the first generation
from a negro, let them intermarry, and their descendants intermarry, until by
generation, they are removed beyond the fourth generation from the pure negro
ancestors, the father of the one, and the mother of the other, from whom they are
56 A F T E R I D E N T I T Y

determinations from the question of who is black to the question of
who is white.
From 1790 until 1952, the United States restricted naturalized
citizenship to ˜˜whites,™™ amending the law in 1870 to include ˜˜persons
of African nativity and African descent.™™ Consequently in petitions for
citizenship brought by natives of different countries, federal and state
courts had to decide which foreign-born applicants could count as
white or black and which, in contrast, were excluded from possible
citizenship on the basis of being neither. In his documentation of
these cases, Haney Lopez shows that courts failed to agree with one
another on what they meant by a white person and that they some-
times failed even to agree with their own recent rulings.22 In 1922, for
instance, in the case of Takao Osawa v. United States, a native of
Japan, Takao Osawa, argued for his eligibility for citizenship on the
basis of the white color of his skin. As he put the point: ˜˜The Japanese
are of lighter color than other Eastern Asiatics, not rarely showing the
transparent pink tint which whites assume as their own privilege.™™23
Nevertheless, the US Supreme Court rejected his petition and denied
that the words ˜˜white person™™ referred to color:

descended, are they any the less free negroes in the fifth than they were in the
first generation from their negro ancestors? They still have half negro blood in
their veins, and that is all they had in the first generation. In the fourth gener-
ation they were unquestionably free negroes, but they certainly had no more
negro blood than their children . . . Can it be that a remove by one generation has
the effect, in law, of turning a half negro into free white man in spite of the color
of his skin or the kinking of his hair? It seems to me both unreasonable and
absurd . . . No person in the fifth generation from a negro ancestor becomes a
free white person, unless one ancestor in each generation was a white person . . .
and unless there is such purification it makes no difference how many gener-
ations you should have to go back to find a pure negro ancestor; even though it
should be a hundred, still the person is a free negro. (State v. William Chavers,
pp. 12“13)

In Germany, Himmler decided on 1650 as the stopping point for prospective SS
members. If research showed their family trees to be free of Jews back to that time,
they were declared Aryan. See Berl Lang, ˜˜Metaphysical Racism (Or: Biological
Racism by Other Means),™™ in Race/Sex: Their Sameness, Difference, and
Interplay, Naomi Zack, ed. (New York: Routledge, 1997), p. 19.
Ian F. Haney Lopez, White by Law: The Legal Construction of Race (New York: New
York University Press, 1996).
Ibid., p. 81.
R A C I A L I D E N T I F I C A T I O N A N D I D E N T I T Y 57

Manifestly the test afforded by the mere color of the skin of each
individual is impracticable as that differs greatly among persons of
the same race, even among Anglo-Saxons, ranging by imperceptible
gradations from the fair blond to the swarthy brunette, the latter
being darker than many of the lighter hued persons of the brown or
yellow races. Hence to adopt the color test alone would result in a
confused overlapping of races and a gradual merging of one into the
other, without any practical line of separation.24

Scientifically more legitimate, the court decided, was an equa-
tion of the meaning of white with Caucasian. While Osawa might be
light in complexion, he remained Japanese and since the Japanese were
not Caucasian, Ozawa was not white. The same court decided United
States v. Bhagat Singh Thind three months later. Relying on Takao
Osawa, Bhagat Singh Thind argued that he was a high-class Hindu of
the Aryan race and noted that experts from Johann Friedrich
Blumenbach on had identified Aryans with Caucasians.25 Since the
Takao Osawa decision defined whites as Caucasians, Thind argued
that he was indisputably white. Nevertheless, shortly after rejecting
Osawa for naturalization as a non-Caucasian, the Supreme Court said
that ˜˜The Aryan theory as a racial basis seems to be discredited by
most™™ and that ˜˜the word Caucasian [the word the court had itself
used three months earlier] is in scarcely better repute.™™26 ˜˜Mere ability
on the part of an applicant . . . to establish a line of descent from a
Caucasian ancestor will not ipso facto and necessarily conclude the
inquiry,™™ the court ruled, for ˜˜˜Caucasian™ is a conventional word of
much flexibility.™™ Indeed, rejecting all ˜˜scientific classification™™ the
court based its decision on the immigrants from the ˜˜British Isles and
Northwestern Europe™™ whom the framers ˜˜must have had affirma-
tively in mind™™ along with the ˜˜immigrants from Eastern, Southern

Takao Osawa v. United States 260 US 178 (1922). See Haney Lopez, White by Law,
Appendix B, p. 220.
Counsel in United States v. Bhagat Singh Thind 261 US 204 (1923).
United States v. Bhagat Singh Thind. See Haney Lopez, White By Law, Appendix B,
p. 223.
58 A F T E R I D E N T I T Y

and Middle Europe™™ who ˜˜were received as unquestionably akin to
those already here.™™27
In these racial prerequisite cases, then, white sometimes refers
to skin color; sometimes it means Caucasian, and sometimes it means
European. In trying to determine whether a person was black, courts
sometimes relied on inspection,28 sometimes on ancestry, and some-
times on reputation. Similarly, in trying to determine whether a
person was white, courts sometimes appealed to science, sometimes
to common sense, and sometimes to the intentions of the framers. Nor
did the courts become clearer about their criteria for racial ascription
after 1927. In twelve cases between 1923 and 1942, numerous ethnic-
ities were dubbed ˜˜not white,™™ including Japanese, ˜˜Asian Indians,™™
Armenians, Punjabis, Filipinos, Afghanis, and ˜˜Arabians.™™ In 1944,
however, Arabians became ˜˜white.™™ The rationale for the federal
court in Michigan calling them not white in 1942 referred to both
common knowledge and legal precedent. So too did the rationale for
the federal court in Massachusetts that in 1944 called them white.29

A quick survey of court cases in the nineteenth and first half of the
twentieth century indicates just how much time federal and particu-
larly state courts had to spend defining and policing racial lines. For
this reason, it is even less plausible in the case of inconsistencies in
racial attributions than in the case of inconsistencies in sex and gender
attributions to suppose that the irregularities represent gray areas in
mostly stable conceptions. Given the extent of the variations in the
racial attributions different authorities made at different times, and
even in the racial attributions that the same authorities made at
almost the same time, our alien anthropologists might well be amazed
at our stubbornness in continuing to insist on racial identities. Indeed,

United States v. Bhagat Singh Thind. See Haney Lopez, White By Law, Appendix B,
p. 224.
And also dealt with the question of who had the authority to do inspections.
See Haney Lopez, White by Law, Appendix A, p. 208.
R A C I A L I D E N T I F I C A T I O N A N D I D E N T I T Y 59

one can imagine that the anthropologists would be at least as exas-
perated about identity in American society as a black or white as
Evans-Prichard was about identity in Zande society as a witch.30 The
Azande, he noted, thought witchcraft was inherited through the male
line and they also believed that it could be detected through a post
mortem examination of a person™s intestines. Given the Zande clan
system, however, just a few post mortem examinations performed on
just a few individuals of different clans would show either that every
Azande was a witch or that none was. Yet, to Evans-Prichard™s cha-
grin, the Azande did not push their beliefs to this logical conclusion
and therefore did not consider their claims about witchcraft to be
problematic. We might make the same point about our notions of
race. On the one hand, the one-drop rule says that a person who has
one African ancestor is a black. On the other hand, a few checks of a
few family trees would reveal just how many of those who function as
whites, conceive of themselves as whites, and are considered by rep-
utation to be white are, on this criterion, black. Indeed, a 1958 study
already found that ˜˜approximately 21 percent of those classified as
white have an African element in their inherited biological back-
ground™™ and that most of those with an African element were living
as whites.31 Either, then, many more people than we tend to think are
˜˜black,™™ or far fewer are. In fact, the number of people who were
˜˜really™™ white or ˜˜really™™ black might be, depending on the criteria
used, either everyone or no one, just as in the case of Zande witches.32
Of course, from the point of view of a more culturally sensitive
anthropology, Evans-Prichard was simply a British imperialist,
unwilling to immerse himself in the Zande conceptual universe.
Peter Winch famously thought it important that the Azande do not
push their beliefs to their logical conclusion because it shows that

See E. E. Evans-Prichard, Witchcraft, Oracles and Magic among the Azande (Oxford:
Oxford University Press, 1937).
See Robert P. Stuckert, ˜˜African Ancestry of the White American Population,™™ Ohio
Journal of Science, 58(3), p. 158. K. Anthony Appiah cites this article in his The
Ethics of Identity (Princeton, NJ: Princeton University Press, 2005), p. 324, n. 48.
Appiah, The Ethics of Identity, esp. pp. 184“186.
60 A F T E R I D E N T I T Y

their witchcraft conceptions have a different function. Whereas the
European anthropologist conceives of witchcraft as a theoretical sys-
tem offering a pseudo-scientific understanding of the world, the Azande
do not.33 We could likewise say that the question of whether our racial
ideas are coherent does not arise for us “ or, at least, does not require us
to abandon them “ because they are part of our practices, not of our
science. We can debate the scientific question of how many elements
in our inherited biological background and which elements make one
a member of a certain population group, but the question of racial
identity is different: it is a question of identities that are constituted
for us by a universe of action and practice.34 Outside of our practices
and activities, blacks and whites as well as Latinos, Latinas, and
Asians have no more reality than witches. Within it, they have the
same amount.
Another way of putting this point is to say that the identification
of racial identities is a matter of knowing how rather than knowing
that. It may be that we cannot articulate a set of rules and criteria by
which to place particular people in particular racial categories. Nor
can we teach someone how to play basketball by giving them a book of
rules. Rather, someone learns basketball by coming to understand
how to play it and in the United States we know, at least roughly,
how to play our game of race. One tells a beginning basketball player to
pass only to a player who is not being effectively guarded but if the
beginner is constantly stripped of the ball by holding on to it for too
long, it becomes clear to him or her that learning the game is less a
question of reviewing a set of rules in his or her head than under-
standing how to play it.35 Similarly, if a group of alien anthropologists

Peter Winch, ˜˜Understanding a Primitive Society,™™ in Fred R. Dallmayr and Thomas
A. McCarthy, Understanding and Social Inquiry (Notre Dame: University of Notre
Dame Press, 1977), p. 172.
This point seems to be the gist of Lionel McPherson™s and Tommie Shelby™s
criticism of Appiah, in ˜˜Blackness and Blood: Interpreting African American
Identity,™™ Philosophy and Public Affairs 32 (2), 2004.
For this example, see Stanley Fish, ˜˜Fish v. Fiss,™™ Stanford Law Review, 36, 1981,
pp. 1331“1334.
R A C I A L I D E N T I F I C A T I O N A N D I D E N T I T Y 61

are trying to learn how to identify blacks and whites in the United
States, it would presumably become clear to them that racial identi-
fication and self-identification is less a question of memorizing spe-
cific criteria than a question of knowing how to engage in a certain
racial practice.
Yet, basketball is coherent as a game. Are racial identification
and identity? While the Azande may be unconcerned about the contra-
dictions in their practice of witchcraft, they cannot expect outsiders to
be similarly unconcerned. Winch thinks that anthropologists must
find that standpoint from within their own culture from which they
can begin to make sense of the practices and beliefs of another.36 In the
case of Zande witchcraft, he thinks the problem with Evans-Prichard™s
account lies in its Western assumptions linking witchcraft to a prim-
itive form of science or even pseudo-science. Witchcraft, he argues,
becomes much more intelligible once it is linked to Western practices
of prayer and religion. But can the same be said about our racial
practices? From what point of view should we be unconcerned about
their contradictions?
Contemporary social theorists offer us one answer: the pers-
pective of power. Racial identities may be practical identities.
Nevertheless, more salient to dissecting them is an appreciation of
the benefits those with social, political, and legal authority accrue
from making use of the contradictions they involve. On an Evans-
Prichard-type reading of the judicial rulings we have surveyed, they
are exasperating because they fail to sort out consistent or plausible
criteria for determining a person™s race and because they fail to recog-
nize that they have failed. On what we can call a social constructionist
reading, however, unacknowledged contradictions are a boon for those
in power precisely because they open up multiple possibilities for
imposing their own agendas.37

Winch, ˜˜Understanding a Primitive Society,™™ p. 171.
Daniel Sharfstein argues that Southern courts were well aware that the lines they
were creating between races were arbitrary. See Daniel Sharfstein, ˜˜The Secret
History of Race in the United States,™™ Yale Law Journal, April 2003.
62 A F T E R I D E N T I T Y

Michael Omi™s and Howard Winant™s account of racial forma-
tion is the classic social constructionist account of race.38 What Omi
and Winant term the macro-level of formation involves the social,
historical, and political practices, events, and actions that established
the United States™ particular racial typology. Although early English
encounters with Africans say little about race and more about labor,
gold, and the Portuguese,39 racial forms of identification emerged with
the African slave trade. Slave traders collected people of diverse cul-
tures and dubbed them indiscriminately as black Africans. Slave own-
ers reinforced this form of identification by consciously mixing slaves
of different ethnicities in their individual forces and refusing to rec-
ognize bonds of kinship between spouses, parents, or children.
Individual colonies further entrenched racial identifications by estab-
lishing legal distinctions between African and non-African servants,
establishing African servitude for life and consciously setting black
and white laborers against one another. Virginia™s reaction to Bacon™s
Rebellion is a particularly good example of the last tactic. Recognizing
the threat to the gentry posed by a coalition of African and European
servants and non-landowners, it set out to re-center the conflict from
non-landowner against landowner to white against black. To this end,
it repealed all penalties imposed on Europeans for their participation
in the rebellion but enforced them on African participants; it
employed the European rebels to help quash slave revolts; and it sold
the livestock formerly owned by the African participants to poor
Europeans. The colony also prohibited African slaves from assembling
and moving freely and, in 1691, it passed a series of even more restric-
tive laws prohibiting the manumission of slaves unless their master
paid to remove them from the colony, levying fines on free white

Michael Omi and Howard Winant, Racial Formation in the United States from the
1960s to the 1980s (New York: Routledge, 1986), p. 66.
See Kathleen M. Brown, Good Wives, Nasty Wenches and Anxious Patriarchs:
Gender, Race and Power in Colonial Virginia (Chapel Hill, NC: University of
North Carolina Press, 1996), pp. 37“40.
R A C I A L I D E N T I F I C A T I O N A N D I D E N T I T Y 63

women who gave birth to racially mixed children, and denying free
blacks the right to vote, hold office, or testify in court.40
While actions such as these helped to construct racial identifi-
cations on the macro-level, the internalization of the identifications
constructs what Omi and Winant call the micro-level. Individuals
introject or appropriate their macro-level racial identification and
make it a part of their self-identity. The separation of slaves from
their original ethnic and linguistic consociates as well as from their
families is part of this process since it forced displaced Africans to
forge new systems of solidarity based on their new circumstances as
well as new customs and hybrid forms of religion to overcome their
cultural distance from one another.41 Frederick Douglass writes about
his ties on the plantation where he was enslaved that the slaves ˜˜were
as true as steel, and no band of brothers could have been more loving . . .
We never undertook to do anything of importance, which was likely to
affect each other, without mutual consultation. We were generally a
unit, and we moved together.™™42 Of course, white consciousness of a
separate identity followed a similar trajectory insofar as whites con-
sciously distinguished themselves from blacks and other ˜˜non-
whites.™™ After the Civil War, this racial consciousness flourished in
racial ideologies and Jim Crow laws while black racial consciousness
grew in the efforts of blacks to lead successful lives in a society legally
and politically armed against them.
Ian Hacking emphasizes another element of social construction
in what he calls ˜˜looping effects,™™43 which, in the case of racial con-
structions, we can see as the reciprocal influences of macro- and
micro-levels on one another. Individuals take up their external racial

See Ronald Takaki, A Different Mirror: A History of Multicultural America (Boston:
Little, Brown & Co., 1993), pp. 63“67.
Stephen Cornell and Douglas Hartmann, Ethnicity and Race: Making Identities in a
Changing World (Thousand Oaks, CA: Pine Forge Press, 1998) pp. 104“106.
Frederick Douglass, My Bondage and My Freedom, William Andrews, ed. (Urbana:
University of Illinois Press, 1987), pp. 164“165.
Ian Hacking, The Social Construction of What? (Cambridge, MA: Harvard
University Press, 1999), p. 34.
64 A F T E R I D E N T I T Y

identification as part of who they are; indeed, the racial identities they
adopt become fundamental to them insofar as they shape prospects
and life-plans, determine who is part of which ˜˜unit,™™ and offer reasons
for action. At the same time, lives lived in terms of particular racial
designations ˜˜loop™™ back to develop and change the meaning of the
designations themselves. ˜˜Looping effects are everywhere,™™ Hacking

Think what the category of genius did to those Romantics who
saw themselves as geniuses, and what their behavior did in turn
to the category of genius itself. Think about the transformations
effected by the notions of fat, overweight, anorexia. If someone
talks about the social construction of genius or anorexia, they are
likely talking about the idea, the individuals falling under the idea,
the interaction between the idea and the people, and the manifold
of social practices and institutions that these interactions

Being identified by others as a genius or an anorexic has con-
sequences for the way one thinks about oneself, the goals one sets for
oneself, and the expectations one has. One™s external identification
thus affects and helps to construct an internal identity. In turn, the
goals and expectations that individuals have as geniuses and anorexics
feed back into the designations. Institutions, practices, and medical,
educational, and perhaps even legal discourses develop to deal with
the identities; these discourses then feed back into the way those
designated as geniuses and anorexics think about their prospects and
identify themselves and these feed back into the social and institu-
tional level, and so on. The same holds for identifications as black,
white, Asian, and Hispanic. The ascription establishes the circum-
stances of one™s life, one™s sense of how one fits into one™s society, and
the life trajectory one foresees and establishes for oneself. One™s racial
identification thus arranges the list of possibilities one draws from in

Ibid., p. 34.
R A C I A L I D E N T I F I C A T I O N A N D I D E N T I T Y 65

planning one™s life and it shapes the way one reacts both to others and
to events. In turn, the sense that individuals have of their prospects
and their expectations loops back to develop the meaning of racial
classifications. Institutions, practices, and a series of medical and legal
discourses develop around the identities. Indeed, societies change to
fit and regulate the racialized individuals they have created. Think,
then, of ˜˜the manifold of social practices and institutions™™ created by
the interactions between notions of race, particular racial attributions,
and social and political life in the United States.
In place of a Whiggish exasperation with racial identifications
and identities as well as with interpretive efforts to make sense out of
them, then, social constructionists offer a causal account, the virtue of


. 2
( 8)