Content uploaded by Jeffrey K Snyder
Author content
All content in this area was uploaded by Jeffrey K Snyder on Dec 01, 2020
Content may be subject to copyright.
The dominance dilemma: Do women really
prefer dominant mates?
JEFFREY K. SNYDER,aLEE A. KIRKPATRICK,bAND H. CLARK BARRETTa
aUCLA Center for Behavior, Evolution, and Culture and bThe College of William
and Mary
Abstract
Previous research has led to a widely accepted conclusion that heterosexual women prefer mates who are high in
dominance. Three experiments designed to distinguish dominance from prestige and examine moderating contextual
factors challenge this conclusion. College women at 2 U.S. universities evaluated hypothetical, potential mates described
in written vignettes. Participants in Study 1 preferred a high-prestige to a high-dominance target. With dominance and
prestige manipulated independently in Study 2, participants preferred high to low prestige but also preferred low to high
dominance. Participants in Study 3 preferred high to low dominance, but only (a) when displayed in the context of an
athletic competition and (b) in ratings of attractiveness and desirability as a short-term (vs. long-term) mate.
Sadalla, Kenrick, and Vershure (1987) pub-
lished evidence indicating that women prefer
men who are high in dominance over men who
are low in dominance as potential dates (i.e.,
potential short-term relationship partners) and
rate them as more attractive. Since then,
despite numerous studies pointing to limita-
tions of this result, it seems that a simplistic
version of their conclusion—that ‘‘women
prefer dominant mates’’—has become conven-
tional wisdom in psychology and related
fields.
Here, we wish to reopen the analysis of this
mate preference phenomenon for scholars of
romantic relationships with the following
goals. First, we intend to clarify and reexamine
Sadalla and colleagues’ (1987) initial findings.
In the pursuit of this goal, we argue that there
is a problematic lack of consensus regarding
what dominance as a construct is and suggest
that subsequent attempts to clarify Sadalla and
colleagues’ work (Jensen-Campbell, Graziano,
& West, 1995) were insufficient. Second, we
provide evidence suggesting that women’s
preferences in regard to status are contingent
on several factors, including (a) the distinction
between prestige-based and dominance-based
status, (b) the social context in which the
behavior is observed, and (c) the particular
dimension of desirability being assessed.
Background
Humans are unusual among primates in having
pair bonds characterized by high levels of
coordination and cooperation in raising off-
spring. Evolutionarily minded researchers
have argued that this type of mating system
was favored by increased reliance on complex
extractive foraging and prolonged juvenile
periods during the course of human evolution.
Jeffrey K. Snyder, Department of Anthropology, UCLA
Center for Behavior, Evolution, and Culture and The Col-
lege of William and Mary; Lee A. Kirkpatrick, Depart-
ment of Psychology, The College of William and Mary;
H. Clark Barrett, Department of Anthropology, UCLA
Center for Behavior, Evolution, and Culture.
The authors thank Martie Haselton, Daniel M. T. Fess-
ler, Rob Boyd, Joan Silk, and Francisco Gil-White for their
comments on this work. The authors also express gratitude
to multiple reviewers, editors, and the participants of the
studies herein.
Correspondence should be addressed to Jeffrey K.
Snyder, UCLA Department of Anthropology, 341 Haines
Hall, Box 951553, Los Angeles, CA 90095-1553, e-mail:
jksnyder@ucla.edu.
Personal Relationships,15 (2008), 425–444. Printed in the United States of America.
Copyright Ó2008 IARR. 1350-4126=08
425
Because of the cooperative nature of human
pair bonds, evolutionary psychologists have
proposed that finding suitable mates would
have posed important adaptive problems for
both men andwomen in ancestral environments.
Evolutionary scholars have also proposed
that human females faced several adaptive
problems that were different from those males
faced throughout evolutionary history (Buss,
1994; Symons, 1979). For example, women’s
fitness is likely to have been heavily dependent
on the availability of food for themselves and
their offspring during times when they were
least able to obtain those resources due to the
demands of gestation, nursing, and prolonged
juvenile periods. Women probably faced the
risk of sexual coercion to a greater degree than
men (Smuts, 1992). In addition, because men
face a trade-off between investing in offspring
and pursuing additional mating opportunities
(Gangestad & Simpson, 2000; Trivers, 1972/
2002), finding a suitable mate who will not
divert resources elsewhere poses a particular
problem for women. For these reasons, natural
selection has favored women’s preferences for
men who are able and willing to invest in
women and their offspring (including resource
provisioning and physical protection from
conspecifics; see Buss, 1994; Buss & Schmitt,
1993; Ellis, 1992; Geary, 2002; Symons, 1979,
for reviews of empirical evidence).
Dominance (sometimes social dominance)
is one of the several characteristics that
researchers have identified as influencing
women’s preferences for opposite-sex roman-
tic partners because a man who is high in dom-
inance may provide women with better access
to resources and protection from danger, pro-
vided that dominance includes physical prow-
ess or at least a willingness to aggress against
competitors. Furthermore, dominant men’s
genetic material may offer potential socially
competitive advantages to offspring to the
extent that traits associated with dominance
are heritable (Jensen-Campbell et al., 1995;
Sadalla et al., 1987). For example, Sadalla
and colleagues (1987) conducted four studies
in which they asked participants to rate high-
or low-dominance opposite-sex targets in
terms of their perceived characteristics such
as attractiveness, desirability as a date or
spouse, and likability. Their primary findings
include that women find high-dominance men
more attractive and more desirable as a date than
low-dominance men, whereas manipulations of
dominance in opposite-sex targets did not alter
ratings of attractiveness or desirability by men.
This has been the standard approach to wom-
en’s preferences for dominance in men who are
potential romantic partners for nearly two dec-
ades for nontrivial reasons. Darwin (1871) sug-
gested that both male–male competition
(intrasexual selection) and female choice (inter-
sexual selection) of competitive males can
exaggerate dominance-related traits such as
size, aggressiveness, or weaponry (antlers, can-
ines, etc.). In addition, parental investment the-
ory (Trivers, 1972/2002) suggests that the lower
investing parent will be more competitive for
access to mates than the higher investing parent.
Researchers have often assumed that mem-
bers of the higher investing sex (typically
female) accrue benefits by preferring to mate
with competitive members of the low-invest-
ing sex (typically male) because males who are
successful in same-sex competitions have bet-
ter access to resources and offer potentially
better genetic quality to their offspring. Con-
trary to this assumption, recent nonhuman ani-
mal literature indicates that in some cases,
traits that lead to success in same-sex compet-
itions are neutral or dispreferred by the higher
investing sex (Berglund, Bisazza, & Pilastro,
1996; Bonduriansky & Rowe, 2003; Moore,
Gowaty, & Moore, 2003; Moore & Moore,
1999; Rosenthal, Wagner, & Ryan, 2001; Sih
& Watters, 2005; Wong, 2004). This might
occur because the traits leading to success in
male–male competition (dominance) can have
costs to females that outweigh the benefits of
mating with these males. For example, domi-
nant males might be no better than less domi-
nant males in providing resources, parental
care, or higher quality offspring. In turn,
females might risk injury, death, or lower
fecundity by mating with these high-dominant
mates (Kokko, 2003; Qvarnstro
¨m & Forsgren,
1998) because females and their offspring may
be the victim of the investing male’s aggres-
sive behaviors or because of conflicts of repro-
ductive interest or allocation of investment in
offspring.
426 J. K. Snyder, L. A. Kirkpatrick, and H. C. Barrett
Among humans, we may see parallel phe-
nomena. For example, women do not find
hypermasculine faces attractive (Perrett et al.,
1998), although they generally prefer more
masculine faces (a trait associated with domi-
nance) during the most fertile phase of the
menstrual cycle (Penton-Voak & Perrett,
2000; Penton-Voak et al., 1999). These results
reflect a trade-off that women must negotiate
between genetic quality and parental invest-
ment. While masculine facial features might
indicate genetic quality, for example, immu-
nocompetence in spite of high levels of testos-
terone during development, women in these
studies also associated them with increased
dominance and other negative or undesirable
qualities such as coldness and dishonesty.
How can we reconcile these conflicting
findings regarding dominance and women’s
mate preferences? The conventional view in
the evolutionary psychology literature is that
highly dominant men are desirable as mates as
long as their behaviors reflect high dominance
and agreeableness or generosity simulta-
neously (Jensen-Campbell et al., 1995). The
problem, in our view, is the behaviors that lead
to dominance in same-sex competitions do not
always simultaneously reflect socially desir-
able traits such as agreeableness or charitable
generosity. As the nonhuman animal literature
suggests, the value of dominance might
depend on context because dominance could
entail trade-offs if there are situations in which
the costs of mating with a dominant male out-
weigh the benefits. This is what we call ‘‘the
dominance dilemma.’’ Here, we will attempt
to tease apart the various meanings of the term
dominance, as well as the related but distinct
concepts of prestige and status, in order to
explain why we think this dilemma arises
and to explain the problems women face due
to this dilemma.
What is dominance?
Sadalla and colleagues (1987) wrote that ‘‘it
would be inappropriate to conclude that any
manipulation of dominance will result in anal-
ogous effects on attraction. It is clear that the
term dominance is semantically close to several
other concepts and has multiple behavioral
effects’’ (p. 734). Consistent with this view,
the authors found that women’s preferences
for high-dominance men were attenuated when
the authors characterized the dominance as
aggressive and domineering. This leads to the
question, what do we mean when we speak of
dominance and are there components that we
can tease apart?
It is important to distinguish between domi-
nance as a behavioral outcome—a man winning
a competition with another man—and the vari-
ous dispositional traits associated with domi-
nance, such as aggressiveness and other
domineering qualities. Dominance is closely
related to other concepts related to hierarchical
position and differential access to resources such
as leadership, status, and prestige. From one per-
spective, dominance may refer to a position
within a hierarchy or the ability to supplant
a competitor, particularly in ethological studies
of nonhuman animals. From another perspective,
the disciplines of sociology and anthropology
most often use the term dominance following
the ordinary definition of dominance as authority
or control over another or others. In psychology,
the construct of dominance frequently refers to
a cluster of individual characteristics or person-
ality traits that tend to put their possessors into
leadership positions. For example, Sadalla and
colleagues (1987) operationally defined domi-
nance as including traits such as commanding,
authoritative, and masterful, and Jensen-Camp-
bell and colleagues (1995) defined dominance
as including traits such as active, assertive, bold,
talkative, and verbal. These operational defini-
tions have significant overlap with another con-
struct that personality psychologists refer to as
surgency (Buss, 1991; Digman, 1990; Goldberg,
1981). Surgency is a form of extraversion that
includes warmth, gregariousness, assertiveness,
activity, excitement seeking, positive emotion,
a seemingly leader like quality, competitiveness,
approachability, ambitiousness, sociability, and
having interpersonal ascendancy.
Evolutionary psychologists often use etholog-
ical definitions of dominance and nonhuman ani-
mal observations to generate hypotheses and
predictions about human status but acknowledge
additional means to the acquisition of status
besides domineering behaviors, such as social rea-
soning and strategizing in reciprocal relationships
The dominance dilemma 427
(Cummins, 1996, 1999; Hawley, 1999, 2002,
2003a, 2003b; Hawley & Little, 1999). There is
much to recommend from this view and we stress
that it is critical to consider the multiple pathways
to status that are available to humans. In particu-
lar, following Henrich and Gil-White (2001), we
suggest that it is important to distinguish between
dominance and prestige as alternate (though not
entirely mutually exclusive) pathways to social
status.
Dominance versus prestige
Henrich and Gil-White (2001) suggested that
researchers frequently conflate prestige and
dominance as equivalent terms for status. They
stated: ‘‘Status can be viewed as a hierarchy of
rewards or as a hierarchy of displays,’’ that
‘‘high status entails greater access to desirable
things’’ (p. 166) and went on to distinguish two
alternative pathways or strategies for enhancing
status. Henrich and Gil-White conceptualize
dominance to be characterized by forced or
coerced compliance to leadership. Alternately,
prestige, ‘‘the noncoerced, interindividual,
within group, human status asymmetries’’ (p.
166), includes a recognition of certain abilities
by peers that leads to freely conferred status.
Individuals who are knowledgeable or skilled
are conferred prestige-type status by deferent
sycophants (in this context, the term sycophant
is not used with a negative connotation but
rather the term is synonymous with clients, fol-
lowers, or admirers) who, in turn, gain access to
the knowledge individuals hold (infocopying)
and save themselves the costs of trial-and-error
learning. This saving of energy expenditure pro-
vided the selection pressures that favored skill-
seeking, skill-attainment behaviors; the ability
to identify skilled peers; and the propensity to
defer to such individuals as derived traits in the
human lineage (in contrast to dominance as an
ancestral trait).
Furthermore, persuasive appeals to compli-
ance with the leadership of high-status individ-
uals usually characterize prestige, whereas
grandstanding and agonistic methods of lead-
ership mark dominance. Therefore, the presti-
gious individual is influential, honored, and
revered by sycophants rather than feared by
subordinates. Sycophants put themselves in
close proximity to and maintain eye contact
with prestigious individuals in order to gain
information. In contrast, subordinates will
maintain greater distance from and maintain
less eye contact with dominant individuals.
Sycophants offer praise to prestigious individ-
uals who respond with self-deprecation. Pres-
tigious individuals freely offer information
and counsel. Because of this free exchange
of status for information, prestigious individu-
als may appear to be more kind, generous, and
willing to help than dominant individuals.
If this perspective is correct, previous experi-
ments may have conflated dominance and pres-
tige. For example, Sadalla and colleagues
(1987) and Jensen-Campbell and colleagues
(1995) used Mehrabian’s (1969) nonverbal
characterizations of high-dominance targets in
videotape manipulations such as sitting back in
a chair with relaxed, asymmetrical posture,
good eye contact, high rates of gesturing, and
low rates of head nodding. Mehrabian’s charac-
terizations of nonverbal dominant behavior
include nonverbal behaviors of both dominance-
based status (eye contact and gesturing) and
prestige-based status (relaxed, asymmetric
posture) individuals as Henrich and Gil-White
(2001) outline. In contrast, Argyle’s (1969,
1972) nonverbal descriptors of dominant indi-
viduals such as standing erect, chest expanded,
and hands on hips are more congruent with
Henrich and Gil-White’s characterizations of
dominance-based status individuals.
Why and how the dominance dilemma is
manifested in relationships
Ellis (1992) defined dominance as ‘‘a measure
of one individual’s ability to prevail over
another in competitive encounters that involve
a face-to-face physical component, whether
implicitly or explicitly’’ (p. 274). This ‘‘ability
to prevail’’ might in some cases imply charac-
teristics or qualities, such as aggressiveness,
which can lead to behavioral dominance.
While such qualities might have tended to
lead to success for men in male–male compe-
tition (dominance) over evolutionary time,
these same qualities might have been detri-
mental to their mate’s fitness in the household
context. Biparental care was likely to have
428 J. K. Snyder, L. A. Kirkpatrick, and H. C. Barrett
been important to offspring health and survival
and was thus an important contributor to both
maleand femalefitness (Kaplan, Hill, Lancaster,
& Hurtado, 2000). Therefore, the ability of
fathers and mothers to cooperate and coordi-
nate behavior in the context of pair bonding
and child care would have been important.
Behaviors and traits associated with high dom-
inance in men—particularly domineering,
aggressive traits, and behaviors—have the
potential to compromise coordination and
cooperation.
For example, in a sample of Zimbabwean
women, Hindin (2000) documented that
women married to men who exclusively con-
trolled household expenditures had 3%–7%
lower body mass index and greater energy
deficiency (indicative of low investment of
food resources by husbands) compared to
women who contributed to household deci-
sions, which ‘‘can lead to poorer reproductive
outcomes as well as a decreased capacity to
produce food for themselves and their fami-
lies’’ (p. 1525). These results suggest that
domineering characteristics in men can have
real reproductive costs for women. It is not
clear why men in this sample directed resour-
ces away from their household. Perhaps
because men face a trade-off between paternal
investment in offspring and allocating effort to
acquiring additional matings, these domineer-
ing men were directing investments elsewhere
in search of additional matings. Regardless of
the reasons, these men are prevailing over
their wives in a conflict of interest regarding
distribution of resources.
Consistent with this, Sadalla and colleagues
(1987) found that women did not prefer dom-
inant targets who were simultaneously domi-
neering and aggressive. Moreover, even when
women rated targets as more attractive and
desirable as dates, they also rated them as less
likable, tender, and warm. This finding trou-
bled Ellis (1992) and led him to ask whether
‘‘the very qualities that that make up the ability
to invest (e.g., high dominance, achievement
of status) stand in opposition to qualities that
indicate willingness to invest’’ (p. 277).
Closely related to this issue is Jensen-Camp-
bell and colleagues’ (1995) conclusion that
women prefer high-dominance men as poten-
tial romantic partners only if they appeared to
be simultaneously agreeable and helpful. Sim-
ilarly, Geary (2002) has posited that men com-
pete for positions within coalitional dominance
hierarchies but also that women prefer men
who are highly cooperative.
Because of these apparent paradoxes, we
suggest that the qualities that lead to behavioral
dominance might also, in some cases, entail
domineering and aggressive behaviors and
a corresponding deficit in characteristics such
as agreeableness. It might be difficult or impos-
sible for an individual to prevail over another in
a competitive encounter while appearing agree-
able and without displaying explicit or implicit
domineering behaviors. Even the most subtle
verbal and nonverbal displays of dominance
(Burgoon, Buller, Hale, & deTurck, 1984; Puts,
Gaulin, & Verdolini, 2006; Ridgeway, 1987;
Weisfeld & Beresford, 1982) such as interrupt-
ing or talking over someone, posturing, eye-
gazes, and modulation of voice pitch might be
coercive by nature or include veiled threats of
force when displayed with the intent to supplant
or assert dominance rank over a conspecific.
Ultimately, high-dominance men are more
likely to be aggressive and domineering than
low-dominance men.
It is easy to imagine that dominance is an
attractive quality in men as potential mates
because the act of supplanting conspecifics
in a hierarchy may lead to better access to
resources and protection for mates and their
offspring. Nonetheless, not all the traits that
lead to behavioral dominance are necessarily
beneficial to mates, and some might therefore
be unattractive in a man. If the man tends to
utilize dominance behaviors as a global strat-
egy of goal attainment in interpersonal rela-
tions, there might be spillover of such
behaviors directed toward his mate or off-
spring (Date & Ronan, 2000; Mauricio &
Gormley, 2001; Straus, 2004; Sugihara &
Warner, 2002). Furthermore, women risk
being the victims of partner violence to the
extent that aggression is related to dominance
since aggressiveness as a general trait in men
predicts partner-directed aggression in the
home (Lorber & O’Leary, 2004; O’Leary,
Malone,& Tyree, 1994). Dominance is not auto-
matically equivalent to aggression; nonetheless,
The dominance dilemma 429
aggressive coercive tactics in the pursuit of
agentic self-interest frequently characterize
dominance as a personality trait (Gurtman,
1992; Trapnell & Wiggins, 1990). Physical
aggression in the home would obviously be
very costly to a woman and her offspring.
We might therefore expect women to avoid
dominant men, at least under some circum-
stances, rather than seek them out as potential
romantic partners.
To the extent that dominance displays neces-
sarily entail undesirable qualities, then, the
potential selective advantages of mating with
dominant men must be balanced against poten-
tial selective disadvantages. In short, it seems
that dominance in pair-bonded men not only
has the potential to confer benefits to their part-
ners but also the potential to inflict high costs on
their partners because most men will have dif-
ficulty being the optimal competitor in same-sex
competitions while being the optimal cooperator
at home.
We suggest two possible solutions to the
apparent mate selection dilemma women face.
The first arises from the fact that many or most
of the benefits that might accrue to women
from mating with men high in dominance are
equally attainable, with fewer costs, by mating
with men high in prestige. There are a few
possible important exceptions to this. Men
high in prestige but low in dominance-related
characteristics such as aggression might have
fared poorly in contexts that required, for
example, physical prowess, such as warfare
or physical protection from conspecifics. That
said, we suggest that in many but not all social
contexts (e.g., perhaps very violent societies),
the dominance dilemma could be solved by
preferring high prestige over high-dominance
men as mates.
A second potential solution to the domi-
nance dilemma would be to identify men
who are likely to be successful in intrasexual
competitions but not likely to be dominant in
interactions with their mates and offspring. As
Ellis (1992) stated, ‘‘[A woman] may want
a man who is dominant (and therefore less
‘‘warm, likeable, and tender’’) when he is in
competition with other men, but who is warm,
likable, and tender toward her’’ (p. 277). That
is, the desirability of dominance may be
importantly moderated by the context in which
dominance is displayed.
Several of the results Sadalla and colleagues
(1987) report are consistent with the proposal
that women are attempting to solve the domi-
nance dilemma. Participants in their Study 1, as
noted above, watched videotapes of actors dis-
playing (ostensibly) high-dominance or low-
dominance behaviors. All videotapes displayed
interactions between the target and a same-sex
audience. Sadalla and colleagues’ Studies 2 and
3 between-subjects design presented partici-
pants with targets who were equally skilled ten-
nis players but varied in the extent to which
they displayed dominant behaviors while play-
ing tennis. In these contexts, women preferred
‘‘high-dominance’’ men. It remains unknown
whether participants would have viewed the
targets as equally attractive if targets behaved
in a dominant manner toward women as well as
toward other men (in Study 1) or displayed
dominance outside a socially sanctioned ath-
letic competition (in Studies 2 and 3).
We suggest that athletic competitions pro-
vide a social context in which dominant behav-
iors are normative. Competitive displays in leks
(in which males gather in an ‘‘arena’’ during
mating seasons to engage in competitive plum-
age, dance, or vocal displays) and ritualized
fighting among males play a major role in
female mate selection in other species (Zahavi
& Zahavi, 1997). Similar to these nonhuman
animal examples, dances and sports among
men may constitute sanctioned competitions
between males as competitive displays (Geary,
Byrd-Craven, Hoard, Vigil, & Numtee, 2003)
of ambition, physical prowess, and athletic abil-
ity. For example, the Maasai (von Mitzlaff,
1988/1994) use such public displays for the pur-
pose of facilitating mate selection for women.
The present studies
The primary objective of the current studies is
to reevaluate women’s preferences for domi-
nance using Henrich and Gil-White’s (2001)
theoretical conceptualization of status, domi-
nance, and prestige. In this work, we attempted
to evaluate whether women attempt to solve
the dominance dilemma by preferring men
who pursue a nondomineering-based strategy
430 J. K. Snyder, L. A. Kirkpatrick, and H. C. Barrett
of attaining status. This question led us to
examine how women’s preference for domi-
nance in potential mates varied according to
the context of the dominance strategies. We
predicted that measurable differences between
dominance and prestige are observable in
human opposite-sex mating preferences. Spe-
cifically, we predicted that participants would
judge high-prestige men as preferable to
high-dominance men as potential romantic
partners when provided a direct comparison
of two men (Study 1). Second, we predicted
that in contrast to prestige-based status, pref-
erences for men with dominance-based status
would be attenuated in the context of long-
term mate preferences when compared to
short-term mate preferences (Studies 2 and 3).
Finally, we predicted that the social context
would moderate women’s preferences for
men who display dominance-based behaviors
(Study 3).
All the studies that follow utilize target
descriptions (vignettes) as the sole methodol-
ogy for an important reason. Human status
hierarchies are highly complex. Some status
positions are nested within larger hierarchies
while some hierarchies are orthogonal to other
hierarchies. It is very likely that any given
individual will occupy positions in numerous
hierarchies—some obtained primarily with
dominant-based strategies and others with
prestige-based strategies. Furthermore, a com-
bination of dominance- and prestige-based
strategies may be necessary for obtaining cer-
tain hierarchical positions. Any given individ-
ual could display multiple behaviors or cues of
dominance or prestige at any given time. We
therefore utilized a vignette methodology that
enabled us to manipulate and control carefully
the particular behaviors and cues to which par-
ticipants were exposed, as well as to eliminate
potential overlap between the theoretical con-
structs of dominance and prestige.
Study 1
We designed Study 1 to examine the relative
effects of dominance and prestige on women’s
mating preferences. Participants read vignette
descriptions of both a high-dominance target
and a high-prestige target who obtain social
status according to Henrich and Gil-White’s
(2001) characterizations of dominance and
prestige. We predicted that provided the alter-
native of a high-prestige target, women would
rate the high-dominance target as a less attrac-
tive and less desirable mate.
Method
Seventy-one female students from the College
of William and Mary, a highly selective small
state university in Virginia, participated in
Study 1. We intentionally utilized samples of
university students in this reexamination of
dominance preferences because previous
researchers (Jensen-Campbell et al., 1995;
Sadalla et al., 1987) utilized similar sampling
procedures in their work on this topic. (This
sampling procedure also presents some limita-
tions as discussed below.) Participants signed
up for participation in groups of 5 to 30. Par-
ticipants received the questionnaire for Study
1 including fictional vignettes and rating
scales in one packet. Participants read fictional
vignettes describing two male students at dif-
ferent colleges (see Appendix A). Each target
became president of the debating club at his
school. One paragraph described a high-dom-
inance target who took the position by domi-
nating the peer group and grandstanding
during meetings. Another paragraph described
a high-prestige target who was given the posi-
tion by the peer group and was casual during
meetings. We counterbalanced the order of
presentation of the high-dominance and high-
prestige targets.
Instructions after the vignette encouraged
participants to imagine what the fictional char-
acter was like before responding to the rating
scales and then honestly and accurately report
their first impressions. (We utilized these
instructions in all studies reported in this
work.) Participants rated the targets relative
to each other on a scale from 4 to 4 with 0 as
the midpoint (4–3–2–1–0–1–2–3–4). The
names of the targets, ‘‘Bill’’ and ‘‘Dave,’’
anchored the 9-point scales. We utilized
dependent measures adapted from Sadalla
and colleagues (1987) to obtain ratings of tar-
get’s relative attractiveness and desirability as
a romantic partner.
The dominance dilemma 431
Results and discussion
We arranged the raw scores so that positive val-
ues reflected preferences for the high-prestige
target and negative values reflected preferences
for the high-dominance target, with a score of
0 indicating no preference. We computed one-
sample ttests (against a null value of l¼0) for
this analysis. With regard to desirability as
a romantic partner, we combined ‘‘desirable as
a date,’’ ‘‘desirable as a boyfriend,’’ and ‘‘desir-
able as a spouse’’ into a single aggregate (a¼
.91). Cronbach’s alpha is a measure or reliabil-
ity. Women rated the high-prestige target as
more desirable as a mate, t(70) ¼11.34, p,
.001, with a mean difference 2.15 and standard
error 0.19.
The aggregate (the mean of combined indi-
vidual dependent measures) of judged attrac-
tiveness ratings combined the dependent
measures ‘‘not ugly,’’ ‘‘physically attractive,’’
and ‘‘sexually attractive’’ (a¼.82). (We uti-
lized the same procedure for aggregated depen-
dent measures throughout this series of studies.)
Again, women judged the high-prestige target
as more attractive than the high-dominance tar-
get, t(70) ¼3.73, p,.001, with a mean dif-
ference 0.54 and standard error 0.14.
As predicted, these results demonstrated
a clear preference for high-prestige targets
over high-dominance targets in direct compar-
ison to each other. Conceptualized from the
perspective of Henrich and Gil-White (2001),
both targets were high in status and thus
equally high in resource-acquisition potential.
As we have argued, dominance in a mate
comes with risks that are absent from presti-
gious mates. Therefore, one potential strategy
that has the potential to resolve the dominance
dilemma is to choose prestige over dominance
given the opportunity to do so.
Study 2
We designed Study 2 with several goals in
mind. Participants in Study 1 strongly pre-
ferred high prestige to high dominance, but
the question remains whether this is because
the high-prestige target was particularly
appealing or whether the high-dominance tar-
get was particularly unappealing. Therefore,
we designed Study 2 to be a between-groups
factorial experiment in which we manipulated
dominance (high vs. low) and prestige (high vs.
low) independently. We predicted that partici-
pants would consistently favor high-prestige
targets over low-prestige targets, whereas pref-
erences for high-dominance would depend on
other factors described below.
We also designed Study 2 to examine
whether it was possible to replicate the posi-
tive effect of dominance demonstrated in
Sadalla and colleagues’ (1987) Studies 2 and
3 in a different context, specifically one not
involving a socially sanctioned athletic con-
text. Therefore, we utilized a context that
involved a form of group leadership that would
be relevant to the sample of participants—in
this case, a collegiate fraternity. We predicted
that dominance behaviors in this context
would have the effect of making a potential
mate less rather than more attractive and desir-
able as a mate—especially as a long-term
partner.
Finally, we included additional scales to cre-
ate separate dependent measures of desirability
of the target as a short-term and as a long-term
mate (in addition to attractiveness). We pre-
dicted, consistent with sexual strategies theory
(SST), that women would be more discerning in
their judgments of potential long-term mates as
compared to potential short-term mates (Buss,
1994; Gangestad, Simpson, Cousins, Garver-
Apgar, & Christensen, 2004; Pillsworth, Hasel-
ton, & Buss, 2004; Simpson, Gangestad, Chris-
tensen, & Leck, 1999). Presumably, one of the
functions of short-term mating is to obtain off-
spring of good genetic quality while the focus
of long-term mating is the coordinated effort
toward raising offspring. Assuming that the
traits that lead to dominance are heritable, off-
spring of high-dominance men may have a com-
petitive edge. Therefore, women may value
dominant men as short-term mates that carry
‘‘good genes.’’ Because female strategies of
mate selection vary between seeking a short-
term partner and a long-term partner and
because men frequently use competitive dero-
gation tactics to attract women as short-term
mates, we predicted that women may be more
tolerant of or even prefer dominant character-
istics in potential short-term mates than in
potential long-term mates. In contrast, we
432 J. K. Snyder, L. A. Kirkpatrick, and H. C. Barrett
expected high-prestige targets to be more
highly valued in the context of long-term rela-
tionships than in short-term relationships or on
the attractiveness measure.
Method
One hundred and two female students from the
College of William and Mary, the same state
university where we conducted Study 1, partic-
ipated in Study 2. Participants signed up for
participation in groups of 10–35 and received
the vignettes and rating scales in one packet. All
the vignettes described the target becoming
president of his fraternity. Therefore, Study 2
manipulations varied dominance-based and
prestige-based strategies for obtaining status
within the context of a same-sex competition
for coalition leadership. We described the tar-
get’s demeanor in the first paragraph with
adjectives synonymous with dominance or pres-
tige. The second paragraph of the vignette
described the target in social interaction and
his leadership style with an account of how he
became president. The third and final paragraph
of the vignette stated that the target became
president and described the target as either dom-
inant, prestigious, both, or neither. All vignettes
described John as ‘‘59100tall, 165 lbs.’’ (See
Appendix B for vignettes.)
The rating scale included 40 bipolar adjec-
tives. These bipolar adjectives included the
intended dependent measures as well as neutral
‘‘distracter’’ items intended to make the hypoth-
esis of the study less transparent to participants
and thereby avoid potential demand character-
istics. These adjectives consisted of most of the
same dependent measures utilized in Study 1
with alterations attempting to capture differen-
ces between short-term and long-term mating
strategies. Participants rated each pair of bipolar
adjectives on a 1–9 scale. (Relevant scale
anchors follow in the Results and Discussion
section immediately below. All scale anchors
are available from the first author upon request.)
Results and discussion
We reverse scored the dependent measures
when appropriate and created an aggregated
dependent measure to decrease the probability
of Type I error in a 2 (dominance) 2 (prestige)
analysis of variance (ANOVA). We utilized an
aggregate of ‘‘ugly–handsome,’’ ‘‘physically
unattractive–physically attractive,’’ and ‘‘sexu-
ally unattractive–sexually attractive’’ (a¼.75)
to measure participants perceptions of target
physical attractiveness. ANOVA yielded a main
effect for prestige, F(1, 98) ¼10.22, p¼.002, g
¼.09 (gis the effect size), reflecting women’s
ratings of the high-prestige target (M¼6.37, SE
¼0.06) as significantly more attractive than the
low-prestige target (M¼5.99, SE ¼0.06). (See
Table 1 for mean ratings per condition.) We
found no other significant main effects or inter-
actions for attractiveness judgments.
We identified the same trend for ratings of
target short-term desirability. We measured
short-term desirability of the opposite-sex targets
using an aggregate of ‘‘not desirable as a sex part-
ner–desirable as a sex partner,’’ ‘‘not desirable to
hook up with–desirable to hook up with,’’ ‘‘not
desirable to self as a short-term romantic partner–
desirable to self as a short-term romantic partner,’’
and ‘‘not overall desirable as a short-term roman-
tic partner–overall desirable as a short-term
romantic partner’’ (a¼.74). ANOVA yielded
a significant main effect for prestige, F(1, 98) ¼
13.70, p,.001, g¼.12, such that the high-
prestige target (M¼6.65, SE ¼0.12) was more
desirable than the low-prestige target (M¼5.62,
SE ¼0.12). We found no other significant main
effects or interactions for this measure.
We measured long-term desirability of the
target with an aggregate of three dependent
measures: ‘‘not desirable as a spouse–desirable
as a spouse,’’ ‘‘not desirable to self as a long-term
romantic partner–desirable to self as a long-term
romantic partner,’’ and ‘‘not overall desirable as
a long-term romantic partner–overall desirable as
a long-term romantic partner’’ (a¼.89).
ANOVA yielded a main effect for dominance,
F(1, 98) ¼24.11, p,.001, g¼.20, reflecting
that women rated the low-dominance target (M
¼6.14, SE ¼0.14) as significantly more desir-
able than the high-dominance target (M¼4.61,
SE ¼0.14). A main effect for prestige, F(1, 98)
¼30.11, p,.001, g¼.24, indicated women
rated the high-prestige target (M¼6.08, SE ¼
0.14) as significantly more desirable than the
low-prestige target (M¼4.67, SE ¼0.14). The
interaction was not significant, p..05.
The dominance dilemma 433
In addition, we conducted a three-way
mixed ANOVA in which we treated differences
among the three dependent measures—attrac-
tiveness, short-term, and long-term desirabili-
ty—as a within-subjects factor. (We report
Greenhouse–Geisser tests of within-groups
effects throughout the article.) As predicted
by SST, this analysis yielded a significant main
effect for the within-subjects variable, F(1.73,
169.65) ¼16.53, p,.001, g¼.14, such that
mean ratings were lower (i.e., participants were
more discriminating) with respect to long-term
desirability than short-term desirability or
attractiveness. More important, the within-sub-
jects variable interacted significantly with dom-
inance, F(1.73, 169.65) ¼15.49, p,.001, g¼
.14, and with prestige, F(1.73, 169.65) ¼9.61,
p,.001, g¼.09, such that the preference for
the low-dominance (vs. high-dominance) target
was much stronger (Ms¼6.17 vs. 4.81) with
respect to long-term desirability than with
respect to either attractiveness (Ms¼6.17 vs.
6.21) or short-term desirability (Ms¼6.24 vs.
5.90). Similarly, the preference for the high-
(vs. low-) prestige target was stronger with
respect to long-term desirability (Ms¼6.25
vs. 4.73) than short-term desirability (Ms¼
6.49 vs. 5.63) or attractiveness (Ms¼6.38 vs.
6.01). In sum, women appeared to avoid poten-
tial romantic partners who achieved coalition
leadership through dominant strategies and to
prefer men who attained such leadership due to
prestige. These patterns were particularly true
when evaluating potential long-term mates.
Study 3
These results for dominance stand in marked
contrast to those of Sadalla and colleagues’
(1987) Studies 2 and 3 in which women gener-
ally preferred high-dominance to low-domi-
nance partners. Although Sadalla and
colleagues’ studies are not directly comparable
to our Study 2 above, our hypothesis is that the
crucial difference concerns context: Male dom-
inance may be attractive to women when dis-
played in a socially sanctioned athletic context
but not (especially in a long-term partner) when
displayed in other contexts. Therefore, we con-
ducted a third study manipulating the context of
dominance within a single experiment.
Method
Sixty female students at the University of Cal-
ifornia, Los Angeles, the largest University of
California campus in terms of enrollment and
one of the few public research universities
Table 1. Study 2: Mean ratings of targets
Dominance of target Prestige of target MSDN
Attractiveness judgments
High High 6.33 0.50 25
Low 6.08 0.79 25
Low High 6.43 0.53 27
Low 5.91 0.60 25
Desirability as short-term partner
High High 6.23 1.27 25
Low 5.53 1.48 25
Low High 6.75 0.80 27
Low 5.72 1.08 25
Desirability as long-term partner
High High 5.36 1.60 25
Low 4.25 1.34 25
Low High 7.14 1.12 27
Low 5.20 1.52 25
Note. Higher values represent higher respective ratings of attractiveness and desirability.
434 J. K. Snyder, L. A. Kirkpatrick, and H. C. Barrett
located in a major city, volunteered to partic-
ipate in Study 3. Volunteers participated indi-
vidually and received US$5.00 cash payment
for their participation. Participants received
a questionnaire for Study 3 including fictional
vignettes and rating scales in one packet. We
utilized high- and low-dominance paragraphs
of the athletic context condition verbatim from
Sadalla and colleagues’ (1987) Study 2. We
constructed the high- and low-dominance
paragraphs of the nonathletic condition to be
parallel to the athletic context paragraphs. The
nonathletic context depicted a target who was
competing for leadership in informal decision
making among peers (see Appendix C). The
rating scale included 42 bipolar adjectives to
be used as dependent measures. These adjec-
tives consisted of the same dependent meas-
ures utilized in Study 2 with minor revisions
intended to refine the measure. Volunteers
returned the research materials, received pay-
ment, and were thanked by the experimenter
upon completion of this task.
Results and discussion
We reverse scored the dependent measures
when appropriate and created aggregated
dependent measures to decrease the probabil-
ity of Type I error in 2 (dominance) 2 (con-
text) between-groups ANOVA. We formed an
aggregate dependent measure of participants’
judgments of the targets attractiveness utiliz-
ing the same combination of dependent meas-
ures as the previous study (a¼.81). ANOVA
yielded no significant main effects, although
the interaction approached significance, F(1,
56) ¼3.41, p¼.070, g¼.06 (see Figure 1).
Analysis of simple effects revealed that partic-
ipants judged high-dominance targets (M¼
5.42, SD ¼0.96) to be more attractive than
low-dominance targets (M¼4.36, SD ¼
1.15) in the context of an athletic competition,
F(1, 56) ¼6.54, p¼.013, g¼.10, but not in
the nonathletic context.
We measured short-term desirability of the
target using an aggregate of ‘‘not desirable as
a sex partner–desirable as a sex partner,’’ ‘‘not
desirable go out with–desirable to go out with,’’
‘‘not desirable as a date–desirable as a date,’’
‘‘not desirable to self as a short-term romantic
partner–desirable to self as a short-term roman-
tic partner,’’ and ‘‘not overall desirable as
a short-term romantic partner–overall desirable
as a short-term romantic partner’’ (a¼.84).
ANOVA yielded no significant main effects.
Analysis did reveal a significant interaction,
F(1, 56) ¼6.34, p¼.015, g¼.10, such that
participants preferred the high-dominance tar-
get (M¼5.23, SD ¼0.93) to the low-domi-
nance target (M¼4.13, SD ¼1.13) in the
context of an athletic competition but preferred
the low-dominance target (M¼4.63, SD ¼
1.07) to the high-dominance target (M¼4.19,
SD ¼1.51) in the context of nonathletic com-
petition (see Figure 2). Analysis of the simple
effects revealed a significant difference
between ratings of athletic targets, F(1, 56) ¼
6.44, p¼.014, g¼.10, but not between non-
athletic targets.
Analysis revealed the same trend for ratings
of target long-term desirability. The aggregate
measure of the long-term desirability of the
target consisted of four dependent measures:
‘‘not desirable as a boyfriend–desirable as
a boyfriend,’’ ‘‘not desirable as a spouse–
desirable as a spouse,’’ ‘‘not desirable to self
as a long-term romantic partner–desirable to
self as a long-term romantic partner,’’ and
3.00
3.50
4.00
4.50
6.00
5.50
5.00
Low
Dominance
High
Dominance
Attractiveness
Athletic
Context
Interpersonal
Context
Figure 1. Study 3: Dominance by context
interaction for perceived attractiveness of
target.
Note. Higher scores reflect higher ratings of
attractiveness.
The dominance dilemma 435
‘‘not overall desirable as a long-term romantic
partner–overall desirable as a long-term
romantic partner’’ (a¼.88). ANOVA yielded
no significant main effect for context,
although the main effect for dominance
approached significance, F(1, 56) ¼3.40,
p¼.071, g¼.06, reflecting higher overall
ratings of the low-dominance target compared
to the high-dominance target. There was a sig-
nificant interaction, F(1, 56) ¼9.89, p¼.003,
g¼.15, such that participants preferred the
high-dominance target (M¼4.18, SD ¼1.10)
to the low-dominance target (M¼3.78, SD ¼
1.19) in the context of an athletic competition
but preferred the low-dominance target (M¼
4.55, SD ¼1.27) to the high-dominance target
(M¼3.02.19, SD ¼1.20) in the context of
nonathletic competition (see Figure 3). Anal-
ysis of the simple effects within this interac-
tion indicated significant differences, F(1, 56)
¼12.44, p¼.001, g¼.18, in target ratings in
the context of nonathletic competition but not
the athletic competition (p..05).
A mixed 2 (dominance) 2 (context) 3
(attractiveness, short-term, and long-term
desirability) ANOVA indicated, as in Study
2, that women were more discerning (i.e., pro-
vided lower mean ratings) in their judgments
of potential long-term romantic partners com-
pared to judgments of attractiveness and short-
term partners, F(1.82, 101.78) ¼18.07, p,
.001, g¼.24. In addition, the within-subjects
variable interacted significantly with domi-
nance, F(1.82, 101.78) ¼8.66, p¼.001, g
¼.13, confirming that the size of the domi-
nance effect differed across measures. Aver-
aging across contexts, the effect of dominance
is slightly positive for attractiveness, relatively
neutral for short-term, and strongly negative
for long-term desirability.
In sum, similar to Study 2, the results of
Study 3 indicated that the effects of dominance
on mate preferences can be highly variable in
predictable ways. Consistent with the results of
our Study 2, participants generally preferred
high-dominance targets to low-dominance tar-
gets in the athletic contexts, but the reverse was
true in nonathletic contexts. Also consistent
with Study 2, participants rated high dominance
more negatively with respect to long-term
desirability than with respect to attractiveness
or short-term desirability.
General Discussion
Collectively, the results of these studies dem-
onstrate clearly that dominance and prestige
are distinct constructs, with very different
3.00
3.50
4.00
4.50
6.00
5.50
5.00
Low
Dominance
High
Dominance
Sort-term Desirability
Athletic
Context
Interpersonal
Context
Figure 2. Study 3: Dominance by context
interaction for target desirability as a short-
term romantic partner.
Note. Higher scores reflect higher ratings of
desirability.
3.00
3.50
4.00
4.50
6.00
5.50
5.00
Low
Dominance
High
Dominance
Long-term Desirability
Athletic
Context
Interpersonal
Context
Figure 3. Study 3: Dominance by context
interaction for target desirability as a long-
term romantic partner.
Note. Higher scores reflect higher ratings of
desirability.
436 J. K. Snyder, L. A. Kirkpatrick, and H. C. Barrett
effects, in the context of women’s mate pref-
erences. Study 1 showed that women preferred
targets on whom peers conferred prestige-
based status because of specific knowledge
or skills to targets who achieved dominance-
based status through strategies of force or the
threat of force, given equivalent status out-
comes. Study 2 further demonstrated that,
when dominance and prestige are manipulated
independently, women consistently prefer
high prestige to low prestige.
The effects of dominance were variable and
more complex. Women displayed preferences
for high dominance only within the context of
male–male competitions. This is consistent
with the idea that women’s preferences for
dominance reflect a desire for romantic part-
ners who are able to protect them from male
conspecifics or that are able to dominate other
men in physically competitive contexts more
generally. Women did not prefer targets who
used dominance-based strategies to achieve
status in a different context outside of a socially
sanctioned athletic contest; indeed, they gen-
erally preferred low dominance to high domi-
nance in the nonathletic contexts in Studies 2
and 3. We believe that women were not only
evaluating cues of success in male–male com-
petition but were also evaluating cues of how
likely men were to direct dominance toward
potential romantic partners. Perhaps directing
domineering behaviors toward same-sex in-
group members, as opposed to one or several
out-group individuals, is a good indication that
a given man is domineering across multiple
contexts—even across contexts in which coop-
eration, coordination, and compromise may be
more valued such as household decision mak-
ing. If this interpretation is correct, willingness
to direct dominance strategies against peers
may be serving as a cue of willingness to direct
dominance strategies against romantic partners.
This interpretation also explains why
women in this study preferred low-dominance
targets as potential long-term romantic part-
ners. Not only are men more likely than
women to employ intrasexual strategies of
competitor derogation when competing for
a particular mate (Buss, 1994), but Simpson
and colleagues (1999) and Gangestad and
colleagues (2004) found that women prefer
direct strategies of competitor derogation in
potential short-term romantic partners. None-
theless, these contrasting patterns do not point
to an intractable contradiction of findings.
First, as described above, women may prefer
direct and forceful behavioral displays in men
toward competitors but avoid men who display
the same behaviors toward coalition members.
Second, Simpson and colleagues and Gangestad
and colleagues (2004) used characteristics
such as ‘‘asserting superiority’’ and ‘‘using
a direct approach’’ as a measure of social dom-
inance, whereas the current studies used char-
acteristics such as the use of force or the threat
of force to describe dominance.
One potential alternative interpretation of
our results that is important to consider is that
our results reflect differences not in domi-
nance or prestige, but rather in the social desir-
ability or general likability of our targets. In
our vignettes, the high-dominance target’s use
of coercive, forceful means to supplant com-
petitors may, for example, have made our
high-dominance targets appear simply too dis-
agreeable. This interpretation cannot, how-
ever, reasonably explain the results of Study
3—in which participants preferred the high-
dominance to the low-dominance target in
one context, but displayed the reverse prefer-
ence in the other—nor the results from Studies
2 and 3 showing that the type of relationship
(i.e., long term vs. short term) moderated the
dominance effect.
We did in fact go to considerable lengths to
control the emotional valence of target traits
with both careful selection and parallel con-
struction of target descriptors across studies
and conditions. At some level, however, ‘‘dis-
agreeableness’’ is to some degree inherent in
dominance as we have defined it here. Indeed,
this is essentially the crux of our argument
about the dominance dilemma: The upside to
choosing a dominant mate—especially a long-
term mate—is the associated high status (and
thus access to resources, etc.), but this comes
at the risk of a significant downside. Our con-
tention is that it is unlikely for men to supplant
same-sex competitors without appearing
explicitly or implicitly domineering or aggres-
sive when achieving dominance-based status.
We therefore disagree with the conclusion of
The dominance dilemma 437
Jensen-Campbell and colleagues (1995) that
women prefer dominance in men as long as
men simultaneously display agreeableness
and altruism. We argue that the researchers
actually described their ‘‘high-dominance’’
targets as high in status—not as high in dom-
inance—and that it therefore was unclear to
participants how the targets achieved status.
Limitations and future directions
We do not intend for this work to be the final
word on women’s preferences for dominance
in heterosexual romantic partners—nor should
it be. Our sample is likely to represent young,
relatively affluent women in a relatively
gender-egalitarian (Schlegel, 1972), Western,
industrialized, state society, who may also be
relatively inexperienced in the domain of
romantic relationships. Having demonstrated
the importance of context, we must also
acknowledge that there may be contexts in
which traits such as aggression and being dom-
ineering are in fact desirable. Cultural groups
that have long histories of raiding and revenge
types of warfare such as the Shuar, Achuar,
Huaorani, and Yanomamo of South America
and the Mae Enga tribes of New Guinea are
likely to value aggressive, warrior-like behav-
iors in potential mates (Chagnon, 1983, 1988;
Descola, 1993/1996; Meggitt, 1977; Redmond,
1994). In fact, preliminary evidence suggests
that the more dangerous the local environ-
ment, the more women value the willingness
of a mate or potential mate to aggress against
opponents (Lee, Navarrete, & Snyder, 2006).
Women are likely to face a trade-off in dan-
gerous environments such that they may have
to sacrifice personal autonomy in favor of pro-
tection from conspecifics although there is no
a priori reason why women should have to
sacrifice personal autonomy except that men
who are dominant in same-sex competitions
also tend to be dominant in the household.
We do not posit a world in which local
beliefs are the sole source of standards of
desirability for mates. Nonetheless, local
norms may construct social environments with
varying cost and benefit structures such that
the expression of women’s evolved preferen-
ces may vary in relatively stable, therefore,
predictable ways. In particular, a husband’s
choice of persuasion or coercion in household
decision making may be an evoked response
based on how local norms constitute the
household hierarchy. Future studies may focus
on accounting for varying contexts and local
norms in residence and marriage patterns and
household decision making by including
cross-cultural comparisons.
For example, some small-scale societies are
patrilocal (the married couple resides in close
proximity to the husband’s family), while
others are matrilocal (the couple resides in
proximity to the wife’s family). It may be the
case that wives are more vulnerable to aggres-
sive coercion from their husbands in patrilocal
societies than in matrilocal societies—assum-
ing that the wife’s kin group would be less
tolerant of aggressive coercion than the hus-
band’s kin group. If the risks of aggressive
coercion in the household are systematically
higher in patrilocal groups, women may be
more cautious and avoid marriage partners that
are high in dominance.
Although we believe that Henrich and Gil-
White’s (2001) distinction between prestige
and dominance represents an important contri-
bution to our understanding of hierarchical
position and asymmetries in access to resour-
ces among humans, there are many complex
issues surrounding the application of these
constructs to be resolved. Neither they nor
we construe these strategies merely as general
personality types. The characteristic behaviors
of the respective strategies are likely to be
highly context specific. Any given individual
in a large-scale society requiring high degrees
of specialization and large social networks is
likely to have numerous positions in many
hierarchies—both nested and orthogonal. Fur-
thermore, attainment of hierarchical position is
often a hybrid combination of dominance and
prestige. An individual may be granted a posi-
tion of institutional authority based on skills
(prestige-based status attainment) but once
occupying that position must enforce institu-
tional norms with coercion or punishment. For
example, a professor may offer a disruptive
student a persuasive appeal to compliance
but if the student continues to be disruptive it
may become the professor’s institutional duty
438 J. K. Snyder, L. A. Kirkpatrick, and H. C. Barrett
(in fairness to other students) to make an
appeal to authority or threaten punishment
according to institutional policy. Moreover,
attributions about an individual that are
formed based on observation of behavior are
likely to vary across different observers. For
example, the tennis opponent may have a dif-
ferent view of a dominant player than an
observer of the tennis game does—perhaps
especially so in a mixed-gender tennis match.
Nevertheless, it stands to reason that stable
individual differences are likely to exist in the
general tendency to employ prestige versus
dominance strategies across time and contexts.
Buttermore and Kirkpatrick (2008) have
shown in a series of recent studies that prestige
and dominance can be reliably measured sep-
arately as distinct personality traits and as such
correlate differentially with a variety of theo-
retically relevant variables. At the same time,
their two scales correlate equally strongly with
one of the most widely used personality meas-
ures of dominance (the California Personality
Inventory), suggesting that much previous
research on dominance as a personality trait
has routinely conflated the two constructs. A
potentially rich area for future research might
involve revisiting these earlier studies in light
of the dominance–prestige distinction.
For example, the dominance–prestige dis-
tinction might be useful for reexamining the
circumplex model of interpersonal behavior
(see Trapnell & Wiggins, 1990). Specifically,
individuals who are oriented toward domi-
nance-based strategies of attaining status may
tend to be low in ‘‘nurturance’’ —expressing
high agentic self-interest—while individuals
who are oriented toward prestige-based strate-
gies of attaining status may be higher in nurtur-
ance as the circumplex model posits.
Conclusions
The main contribution of this work to the
existing literature on women’s preferences
for dominance is the reevaluation of Sadalla
and colleagues’ (1987) original findings in
a way that is distinct from Jensen-Campbell
and colleagues’ (1995) findings. Jensen-
Campbell and colleagues’ primary concern
was that Sadalla and colleagues’ high-domi-
nance targets appeared attractive to women
but were also rated as less likable than low-
dominance targets. As noted above, Jensen-
Campbell and colleagues suggested that
dominance in men is only attractive to
women when men appear to be simulta-
neously agreeable. In contrast, our strategy
has been (a) to suggest that high-dominance
men are not only less likable but are capable
of imposing high costs on their romantic part-
ners through the use of coercive strategies in
the household and (b) to disentangle status
and social rank from the means by which rank
is obtained.
The question at hand here is how women
deal with the dilemma of finding mates that are
able to succeed in dominance competitions,
but who will not direct dominance behaviors
toward their own mates. Our findings suggest
that women prefer potential mates who obtain
status through prestige-based strategies over
potential mates who obtain status through
dominance-based strategies. In addition, the
above evidence suggests that women utilize
specific cues that help them identify potential
mates that hold social rank and are direct in
social interactions but are unlikely to direct
force or the threat of force toward them.
In the course of this endeavor, we have
suggested that the term dominance is used
differently in various contexts and subdisci-
plines of the social sciences, and we suggest
that more precise distinctions between the
various folk and the technical notions of this
term will help to advance research on human
social relationships. In particular, we suggest
that Henrich and Gil-White’s (2001) distinc-
tion between dominance and prestige allows
a more sophisticated perspective on human
status asymmetries, particularly with respect
to distinguishing humans from nonhuman
animals.
References
Argyle, M. (1969). Social interaction.NewYork:Atherton.
Argyle, M. (1972). The psychology of interpersonal
behavior (2nd ed.). London: Penguin Books.
Berglund, A., Bisazza, A., & Pilastro, A. (1996). Arma-
ments and ornaments: An evolutionary explanation of
traits of dual utility. Biological Journal of the Linnean
Society,58 , 385–399.
The dominance dilemma 439
Bonduriansky, R., & Rowe, L. (2003). Interactions among
mechanisms of sexual selection on male body size and
head shape in a sexually dimorphic fly. Evolution,57,
2046–2053.
Burgoon, J., Buller, D., Hale, J., & deTurck, M. (1984).
Relational messages associated with nonverbal
behaviors. Human Communication Research,10 ,
351–378.
Buss, D. (1991). Evolutionary personality psychology.
Annual Review of Psychology,42 , 459–491.
Buss, D., & Schmitt, D. (1993). Sexual strategies theory:
An evolutionary perspective on human mating. Psy-
chological Review,100 , 204–232.
Buss, D. M. (1994). The evolution of desire: Strategies of
human mating. New York: Basic Books.
Buttermore, N. R., & Kirkpatrick, L. A. (2008). Prestige
and dominance: Distinguishing two pathways to social
status. Manuscript submitted for publication.
Chagnon, N. (1983). Yanamamo: The fierce people (3rd
ed.). New York: Holt, Rinehart & Winston.
Chagnon, N. A. (1988). Life histories, blood revenge, and
warfare in a tribal population. Science ,239, 985.
Cummins, D. (1996). Dominance hierarchies and the evo-
lution of human reasoning. Minds and Machines ,6,
463–480.
Cummins, D. (1999). Cheater detection is modified by
social rank: The impact of dominance on the evolution
of cognitive functions. Evolution and HumanBehavior,
20, 229–248.
Darwin, C. R. (1871). The descent of man and selection in
relation to sex. London: John Murray.
Date, A. L., & Ronan, G. F. (2000). An examination of
attitudes and behaviors presumed to mediate partner
abuse: A rural incarcerated sample. Journal of Inter-
personal Violence,15 , 1140–1155.
Descola, P. (1996). The spears of twilight: Life and death
in the Amazon jungle (J. Loyd, Trans.). New York:
New Press. (Original work published in 1993)
Digman, J. (1990). Personality structure: Emergence of the
Five-Factor Model. Annual Review of Psychology ,41,
417–440.
Ellis, B. J. (1992). The evolution of sexual attraction: Eval-
uative mechanisms in women. In J. H. Barkow,
L. Cosmides, & J. Tooby (Eds.), The adapted mind:
Evolutionary psychology and the generation of culture
(pp. 267–288). New York: Oxford University Press.
Gangestad, S. W., & Simpson, J. A. (2000). The evolution
of human mating: Trade-offs and strategic pluralism.
Behavioral and Brain Sciences,23 , 573–644.
Gangestad, S. W., Simpson, J. A., Cousins, A. J., Garver-
Apgar, C. E., & Christensen, P. N. (2004). Women’s
preferences for male behavioral displays change across
the menstrual cycle. Psychological Science ,15,
203–207.
Geary, D. C. (2002). Sexual selection and human life his-
tory. In R. Kail (Ed.), Advances in child development
and behavior (Vol. 30, pp. 41–101). San Diego, CA:
Academic Press.
Geary, D. C., Byrd-Craven, J., Hoard, M. K., Vigil, J., &
Numtee, C. (2003). Evolution and development of
boys’ social behavior. Developmental Review,23,
444–470.
Goldberg, L. (1981). Language and individual differences:
The search for universals in personality lexicons. In
L. Wheeler (Ed.), Review of personality and social
psychology (Vol. 2, pp. 141–165). Beverly Hills,
CA: Sage.
Gurtman, M. B. (1992). Trust, distrust, and interpersonal
problems: A circumplex analysis. Journal of Person-
ality and Social Psychology,62 , 989–1002.
Hawley, P. (1999). The ontogenesis of social dominance:
A strategy-based evolutionary perspective. Develop-
mental Review,19 , 97–132.
Hawley, P. (2002). Winning friends and influencing
peers: Strategies of peer influence in late childhood .
International Journal of Behavioral Development ,26,
466–474.
Hawley, P. (2003a). Prosocial and coercive configurations
of resource control in early adolescence: A case for the
well-adapted Machiavellian. Merrill-Palmer Quarterly,
49, 279–310.
Hawley, P. (2003b). Strategies of control, aggression, and
morality in preschoolers: An evolutionary perspective.
Journal of Experimental Child Psychology,85 , 213–
235.
Hawley, P., & Little, T. (1999). On winning some and
losing some: A social relations approach to social dom-
inance in toddlers. Merrill-Palmer Quarterly ,45,
215–225.
Henrich, J., & Gil-White, F. L. (2001). The evolution of
prestige: Freely conferred deference as a mechanism
for enhancing the benefits of cultural transmission.
Evolution and Human Behavior,22 , 165–196.
Hindin, M. J. (2000). Women’s power and anthropometric
status in Zimbabwe. Social Science and Medicine ,51,
1517–1528.
Jensen-Campbell, L. A., Graziano, W. G., & West, S. G.
(1995). Dominance, prosocial orientation, and female
preferences: Do nice guys really finish last? Journal of
Personality and Social Psychology,68 , 427–440.
Kaplan, H., Hill, K., Lancaster, J., & Hurtado, A. M.
(2000). A theory of human life history evolution: Diet,
intelligence, and longevity. Evolutionary Anthropol-
ogy,9, 156–185.
Kokko, H. (2003). The evolution of mate choice and mat-
ing biases. Proceedings: Biological Sciences,270,
653–664.
Lee, S. S., Navarrete, C. D., & Snyder, J. K. (2006, June).
Women’s preferences for aggressive mates: Navigat-
ing a dangerous world? Poster presented at the annual
meeting of the Human Behavior & Evolution Society,
Philadelphia, PA.
Lorber, M. F., & O’Leary, K. D. (2004). Predictors of the
persistence of male aggression in early marriage. Jour-
nal of Family Violence,19 , 329–338.
Mauricio, A. M., & Gormley, B. (2001). Male perpetration
of physical violence against female partners: The inter-
action of dominance needs and attachment insecurity.
Journal of Interpersonal Violence,16, 1066–1081.
Meggitt, M. (1977). Blood is their argument: Warfare
among the Mae Enga tribesmen of the New Guinea
Highlands. Mountain View, CA: Mayfield.
Mehrabian, A. (1969). Methods and designs: Some refer-
ents and measures of nonverbal behavior. Behavioral
Research Methods and Instruments,1, 203–207.
Moore, A., Gowaty, P., & Moore, P. (2003). Females avoid
manipulative males and live longer. Journal of Evolu-
tionary Biology,16 , 523–530.
Moore, A., & Moore, P. (1999). Balancing sexual selec-
tion through opposing mate choice and male compe-
tition. Proceedings: Biological Sciences ,266,
711–716.
O’Leary, K. D., Malone, J., & Tyree, A. (1994). Physical
aggression in early marriage: Prerelationship and
440 J. K. Snyder, L. A. Kirkpatrick, and H. C. Barrett
relationship effects. Journal of Consulting and Clini-
cal Psychology,62, 594–602.
Penton-Voak, I., & Perrett, D. (2000). Female preference
for male faces changes cyclically-further evidence.
Evolution and Human Behavior,21 , 39–48.
Penton-Voak, I. S., Perrett, D. I., Castles, D. L., Kobayashi,
T., Burt, D. M., Murray, L. K., et al. (1999). Female
preferences for male faces changes cyclically. Nature,
399, 741–742.
Perrett, D. I., Lee, K., Penton-Voak, I. S., Rowland, D.,
Yoshikawa, S., Burt, M., et al. (1998). Effects of sexual
dimorphism on facial attractiveness. Nature,394,
884–887.
Pillsworth, E. G., Haselton, M. G., & Buss, D. M. (2004).
Adulatory shifts in female sexual desire. Journal of
Sex Research,41 , 55–65.
Puts, D. A., Gaulin, S. J. C., & Verdolini, K. (2006). Dom-
inance and the evolution of sexual dimorphism in
human voice pitch. Evolution and Human Behavior ,
27, 283–296.
Qvarnstro
¨m, A., & Forsgren, E. (1998). Should females pre-
fer dominant males? Trends in Ecology and Evolution,
13, 498–501.
Redmond, E. (1994). Tribal and chiefly warfare in South
America. Memoir of the Museum of Anthropology,
University of Michigan, Ann Arbor.
Ridgeway, C. (1987). Nonverbal behavior, dominance,
and the basis of status in task groups. American Socio-
logical Review,52 , 683–694.
Rosenthal, G., Wagner, W., Jr., & Ryan, M. (2001). Sec-
ondary reduction of preference for the sword ornament
in the pygmy swordtail Xiphophorus nigrensis (Pisces:
Poeciliidae). Animal Behaviour,62, 37–45.
Sadalla, E. K., Kenrick, D. T., & Vershure, B. (1987).
Dominance and heterosexual attraction. Journal of
Personality and Social Psychology,52, 730–738.
Schlegel, A. (1972). Male dominance and female auton-
omy: Domestic authority in matrilineal societies. New
Haven, CT: HRAF Press.
Sih, A., & Watters, J. (2005). The mix matters: Behavioural
types and group dynamics in water striders. Behaviour,
142, 1417–1431.
Simpson, J. A., Gangestad, S. W., Christensen, P. N., &
Leck, K. (1999). Fluctuating asymmetry, sociosexual-
ity, and intrasexual competitive tactics. Journal of
Personality and Social Psychology,76, 159–172.
Smuts, B. (1992). Male aggression against women: An
evolutionary perspective. Human Nature,3, 1–44.
Straus, M. A. (2004). Cross-cultural reliability and validity
of the Revised Conflict Tactics Scales: A study of
university dating couples in 17 nations. Cross-Cultural
Research,38, 407–432.
Sugihara, Y., & Warner, J. A. (2002). Dominance and
domestic abuse among Mexican Americans: Gender
differences in the etiology of violence in intimate
relationships. Journal of Family Violence ,17, 315–
340.
Symons, D. (1979). The evolution of human sexuality.
New York: Oxford University Press.
Trapnell, P. D., & Wiggins, J. S. (1990). Extension of the
interpersonal adjective scales to include the big five
dimensions of personality. Journal of Personality and
Social Psychology,59, 781–790.
Trivers, R. (2002). Parental investment and reproductive
success. In R. Trivers (Ed.), Natural selection and
social theory: Selected papers of Robert Trivers (pp.
60–110). New York: Oxford University Press (Reprin-
ted from B. Campbell, Ed., 1972 , Sexual selection and
the descent of man, pp. 136–179, Chicago: Aldine).
von Mitzlaff, U. (1994). Maasai women: Life in a patriar-
chal society; Field research among the Parakuyo, Tan-
zania (C. G. A. T. Dibdin, Trans.). Mu¨nchen,
Germany: Trichster Verlag. (Original work published
in 1988)
Weisfeld, G., & Beresford, J. (1982). Erectness of posture
as an indicator of dominance or success in humans.
Motivation and Emotion,6, 113–131.
Wong, B. (2004). Superior fighters make mediocre fathers
in the Pacific blue-eye fish. Animal Behaviour ,67,
583–590.
Zahavi, A., & Zahavi, A. (1997). The handicap principle:
A missing piece of Darwin’s puzzle. New York: Oxford
University Press.
Appendix A: Stimulus Materials Used in
Study 1
High-dominance male target description
This year Dave felt strongly that he should be
president of the club and leaned heavily on
everyone to convince them to vote for him,
which they did. As president, he regularly calls
club meetings, which everyone makes sure to
attend. At these meetings, Dave is a strong and
almost intimidating leader. He calls meetings
to order when he is ready to start, and then
takes the floor right away, often continuing
to speak for most of the time. He sometimes
speaks very emotionally and powerfully, pac-
ing back and forth across the room and occa-
sionally pounding his fist on the table or
raising his voice to make a point. His fellow
club members always listen to him intently,
though usually without making eye contact
with him; they never interrupt him, and usually
wait to speak until Dave asks them to. Outside
of meetings, he tends to hang out with friends
other than Debating Club members, and
although the club members like him, they are
a little reluctant to approach him outside of
meetings. When club members compliment
him on the good job he is doing as president,
which they do from time to time, he enjoys the
feeling of satisfaction and always thanks them
politely.
High-prestige male target description
This year Bill was pleased when his team-
mates unanimously voted to make him presi-
dent of the club. His teammates regularly ask
The dominance dilemma 441
him to hold club meetings, which he does. At
these meetings, it is clear that everyone looks
to Bill as the leader of the group. They are
always eager to hear what he has to say, and
consequently winds up speaking most of the
time. He always speaks in a relaxed and con-
fident manner, rarely becoming emotional or
raising his voice and usually not even standing
up to speak. The other club members always
listen to him intently with their eyes fixed on
him, never interrupting him, but also not hes-
itating to jump in when Bill is clearly finished
speaking. Outside of meetings, he tends to
hang out with friends other than Debating Club
members, but he is always happy to speak with
club members when they approach him, which
they often do. Club members speak highly of
Bill to their friends whenever they have the
chance. When they compliment him to his
face, however, Bill usually laughs it off and
makes a joke about it at his own expense.
Appendix B: Stimulus Materials Used in
Study 2
High-dominance/high-prestige male target
vignette
John is 59100tall, 165 lbs. He is a second
semester junior and has been a member of
his fraternity since he began college. John
tends to control and take charge of every situ-
ation with his distinct and powerful presence.
He is direct and respectable in both formal and
informal social circumstances. All his move-
ments tend to communicate dignity, domi-
nance, reputability, and authority. He is very
determined, refusing to yield against any
opponents or challengers to his esteemed rep-
utation. When challenged, John will tower
over any competitor with a show of oppressive
will which usually results in the competitor
giving up.
John was determined to become the Presi-
dent of his fraternity this year and was willing
to face anyone who got in his way. When it
came time for nominations, John stood and
paced while he spoke and was emotional at
times when expressing why he should be nom-
inated for president. He reminded the members
of the fraternity that they had frequently asked
him to hold meetings, direct important deci-
sion making, and head important activities.
John went on to remind them of his many pos-
itive attributes, successes, and accomplish-
ments across many areas of his life. When
John spoke forcefully, his peers would avoid
eye contact and slouch in their seats. However,
John began to suspect that he was not going to
encounter any resistance, so he sat and began
to speak in a relaxed and confident manner. At
this time, John’s peers sat up, made good eye
contact and listened intently until he was fin-
ished speaking.
John was pleased when his fellow fraternity
members expressed to him that they had
already unanimously decided among them-
selves to elect him president. Furthermore,
they acknowledged that this decision was
based on John’s consistent successes and fre-
quent accomplishments. Consistent with many
other of John’s experiences, it was clear that
his peers admired him, held him in high
regard, and were unwilling to challenge his
desire to lead them.
High-dominance/low-prestige male target
vignette
John is 59100tall, 165 lbs. He is a second
semester junior and has been a member of
his fraternity since he began college. John
tends to control and take charge of every situ-
ation with his commanding and powerful pres-
ence. He is direct and overbearing in both
formal and informal social circumstances.
All his movements tend to communicate dom-
inance and authority. He is very competitive,
refusing to yield against any opponents or
challengers to his lead. When challenged, John
will tower over any competitor with a show of
oppressive will which usually results in the
competitor giving up.
John is determined to become the President
of his fraternity this year and is willing to chal-
lenge anyone who gets in his way. When it
came time for nominations, John stood and
paced while he spoke and was emotional at
times when expressing why he should be nom-
inated for president. He suspected that he had
competitors for the position and expressed
a subtle threat to any competitors. He reminded
442 J. K. Snyder, L. A. Kirkpatrick, and H. C. Barrett
the members of the fraternity that he fre-
quently led meetings, directed important deci-
sion making, and headed important activities.
John went on to remind them of his many pos-
itive attributes, successes, and accomplish-
ments across many areas of his life. When
John spoke forcefully, his peers would avoid
eye contact and slouch in their seats.
It was obvious that John’s peers held him in
low esteem and were not ready to concede the
presidency to him when they called for further
nominations. John resolved to defeat his com-
petition and pressure his peers until he was
sure to be elected. In spite of the overall low
regard his peers held for him, John was elected
president of his fraternity because of his char-
acteristic dominance and command over his
peers. Consistent with many other of John’s
experiences, it was clear that he could succeed
by oppressing and dominating challengers to
his desire to lead.
Low-dominance/high-prestige male target
vignette
John is 59100tall, 165 lbs. He is a second
semester junior and has been a member of
his fraternity since he began college. John
tends to be illustrious in every situation with
his notable and distinctive presence. He is
prominent and respectable in both formal and
informal social circumstances. All his move-
ments and speech tend to communicate dig-
nity, credibility, and honor. John prefers
working with or influencing others over direct
competition. When challenged in some way,
he will usually blow it off and allow others
to have their way—especially when someone
is resistant to his influence.
John was pleased and willing to accept
a nomination to become the President of his
fraternity this year. This came as no surprise
because the members of the fraternity had fre-
quently asked him to hold meetings, direct
important decision making, and head impor-
tant activities. When asked to speak with
regard to the nomination, he sat and began to
speak in a relaxed and confident manner. John
suspected that he had a competitor but was
unwilling to attempt to control or dominate
the situation. Furthermore, he was unwilling
to act oppressive or overbearing toward a com-
petitor or his peers. John briefly spoke of his
many positive attributes, successes, and
accomplishments across many areas of his life.
While he spoke, his peers sat up, made good
eye contact and listened intently until he was
finished speaking.
John was pleased when his fellow fraternity
members expressed to him that they had
already unanimously voted among themselves
to make him president. Furthermore, they
acknowledged that this decision was based
on John’s consistent successes and frequent
accomplishments. Consistent with many other
of John’s experiences, it was clear that his
peers admired him and held him in high
esteem and regard.
Low-dominance/low-prestige male target
vignette
John is 59100tall, 165 lbs. He is a second
semester junior and has been a member of
his fraternity since he began college. John
tends to occupy positions of high status with-
out being held in high regard by his peers or
taking the positions by force. He is not com-
manding or particularly notable in either for-
mal or informal social circumstances. All his
movements and speech tend to lack authority
and prominence. John prefers working with
others over direct competition. When chal-
lenged in some way, he will usually blow it
off and allow others to have their way.
John requested, and was granted, a nomina-
tion to become the President of his fraternity
this year. When asked to speak with regard to
the nomination, he sat for periods of time and
sometimes stood to ensure that everyone could
hear him and see him. John spoke without
much emotion or confidence. He simply
reminded all the members that he would be
a senior next year, consistently attended meet-
ings, participated in important decision mak-
ing, and important activities. John suspected
that he had a competitor but was unwilling to
attempt to control or dominate the situation.
Furthermore, he was unwilling to be oppres-
sive or overbearing toward a competitor or his
peers. John briefly spoke of his many positive
attributes, successes and accomplishments
The dominance dilemma 443
across many areas of his life. While he spoke,
some of his peers remained attentive and
others appeared uninterested.
John was pleased when his fellow fraternity
members elected him president and promised
to continue to be a willing participant in the
fraternity. Consistent with many of his other
experiences, John found that he could succeed
without being pushy or even admired.
Appendix C: Stimulus Materials Used in
Study 3 (Manipulation of Dominance
Among Peers)
High-dominance, nonathletic, peer context
John is 59100tall, 165 lbs. He has been in col-
lege for 1 year and is currently enrolled in
undergraduate classes. Despite being relatively
new at his school he is very confident and takes
the lead in the decisions he and his peers
make 60% of the time. He is coercive in
his strong arguments by powerfully dis-
counting others’ opinions. These behaviors
lead to success in disputes with his peers, as
he refuses to yield to them—even if they
have been in college much longer. All his
movements tend to communicate dominance
and authority. He tends to psychologically
dominate his peers, forcing them out of their
leadership roles and into following his lead.
Low-dominance, nonathletic, peer context
John is 59100tall, 165 lbs. He has been in
college for 1 year and is currently enrolled in
undergraduate classes. Despite being rela-
tively new at his school he is very confident
and takes the lead in the decisions he and his
peers make 60% of the time. He is not coercive
during arguments but his verbal skills are good
and he makes good points. Although he is
comfortable leading the group, he is just as
comfortable letting others take the lead. He
is not particularly competitive and tends to
yield to his peers when challenged—especially
if they have been in college much longer.
Although he makes good points when arguing,
he easily gives in when his peers speak with
great authority. He enjoys spending time with
his peers but avoids getting into highly com-
petitive situations with them.
(See Sadalla, Kenrick, and Vershure, 1987,
Experiment 2 for high-dominance, athletic
context and low-dominance, athletic context
conditions).
444 J. K. Snyder, L. A. Kirkpatrick, and H. C. Barrett