Male Aggression and Sexual Coercion of Females in Nonhuman Primates and Other Mammals: Evidence and Theoretical Implications

Barbara B. Smuts    Departments Of Psychology and Anthropology and Center For Human Growth and Development University Of Michigan Ann Arbor, Michigan 48109

Robert w. Smuts    Ann Arbor, Michigan 48104

I INTRODUCTION

Although Trivers’ general rule has many exceptions, it accurately identifies the primary source of conflict between the sexes: in most sexual organisms most of the energy and time invested in offspring comes from females. From this basic fact it follows that, for males more than females, reproductive success is limited by the number of matings with fertile partners. For females more than males, on the other hand, reproductive success is limited by the time and effort required to garner and transfer energy to offspring and to protect and care for them (Bateman, 1948; Trivers, 1972). Males therefore are usually more eager than females to mate at any time with any partner who may be fertile, while females are usually more careful than males to choose mates who seem likely to provide good genes, protection, parental care, or resources in addition to gametes (Trivers, 1972; Alexander and Borgia, 1979).

Combined with female interest in mate quality, male interest in mate quantity creates a widespread conflict of interest between the sexes (Borgia, 1979; Parker, 1979; Hammerstein and Parker, 1987). The conflict is mitigated when males court females by offering them the benefits females want from males, such as food, protection, or help in rearing young. These benefits are typically costly in terms of male time and energy, however, and males may often be able to overcome female reluctance at lower cost to themselves by using force or the threat of force, behavior that we call “sexual coercion.”

Theoretical treatments (e.g., Hammerstein and Parker, 1987) indicate that sexual coercion can function as an important selective force influencing the evolution of both male and female behaviors. However, male aggression toward females, including sexual coercion, has rarely been a focus of study, and for the vast majority of animals, including most mammals, quantitative information is unavailable. These limitations severely constrain our ability to determine the evolutionary significance of sexual coercion.

This article aims to stimulate research and theorizing about sexual coercion by reviewing the relevant evidence for nonhuman primates and some other mammals in which sexual coercion is especially well documented. Two contrasting goals guide this review. On the one hand, we hope to persuade the reader that sexual coercion is an important phenomenon worthy of further study. On the other, we wish to highlight important gaps in our knowledge of sexual coercion. We have tried to balance these two goals by using limited evidence from a small number of species to generate hypotheses, while emphasizing that, to test these hypotheses, we need much better information from a larger number of species.

We begin by describing aggressive male behaviors that appear to function as sexual coercion, the costs that this male aggression imposes on females and young, and the counterstrategies that females employ to reduce these costs. The data that we review for primates and other mammals reveal extensive variation in the form and frequency of male aggression against females, and we propose several hypotheses to help account for this variation. We also consider the kinds of evidence needed to determine whether particular cases of male aggression against females function as sexual coercion. In the final section, we argue that sexual coercion has been underestimated as a significant force in social evolution, and indicate how more attention to intersexual coercion as a form of sexual selection can enhance our understanding of animal societies.

THE CONCEPT OF SEXUAL COERCION

We define sexual coercion as use by a male of force, or threat of force, that functions to increase the chances that a female will mate with him at a time when she is likely to be fertile, and to decrease the chances that she will mate with other males, at some cost to the female. The functional consequences of male sexual coercion distinguish it from other instances of male aggression against females (e.g., in the context of feeding competition) that do not appear to involve manipulation of sexual opportunities.

Our definition of sexual coercion as a subset of aggressive male behaviors toward females that is delineated by their function means that sexual coercion is not a purely behavioral concept, but involves a combination of behavioral description and functional explanation. Sexual coercion cannot be identified by observing only the immediate behavior of the aggressor; it is also necessary to observe the subsequent behavior of the aggressor, the target, and even of other individuals. It is not an easy concept to work with, but we believe it is nevertheless useful because it accurately reflects the complexity of agonistic sexual behavior in animals. Toward the end of this article, we consider in some detail how one can test the hypothesis that particular acts of male aggression against females fit the functional definition of sexual coercion given here. We delay this discussion until later because it requires a basic understanding of the wide variety of male aggression toward females that is observed in nature. Thus, we will proceed for the moment on the assumption that sexual coercion does indeed exist, while keeping in mind the need to examine functional consequences before accepting the hypothesis that a particular aggressive act (or set of acts) actually functions as sexual coercion.

Our definition also limits sexual coercion to behavior that involves the use or the threat of force. Although males can (and do) manipulate female mating behavior to their own advantage by inflicting other kinds of costs or by withholding benefits, such a broad definition of sexual coercion would encompass so large a part of all interactions between males and females that it would prove useless.

II MALE AGGRESSION AND SEXUAL COERCION IN NONHUMAN PRIMATES

In what follows, we concentrate on polygynous primates living in groups in which a single male monopolizes matings with two or more females, or multiple males compete for mating opportunities with multiple females. Because polygyny is typically associated with much more intense male–male competition for mates (Clutton-Brock and Harvey, 1976, 1978), these species are expected to show more male sexual coercion than species living in monogamous or polyandrous groups. Reduced sexual coercion is especially likely in monogamous and polyandrous primates, because these species invariably establish long-lasting pair bonds and defend territories against other groups (Goldizen, 1987; Robinson et al., 1987; Robbins Leighton, 1987), minimizing opportunities for contact between opposite-sex individuals other than mates. This is in contrast to the situation in many monogamous birds and in most human groups, in which, because of high mobility and/or communal living, mated individuals may frequently encounter opposite-sex individuals other than their mates (Westneat et al., 1990; Rodseth et al., 1991). Although, on theoretical grounds, sexual coercion is expected to be considerably less common in monogamous and polyandrous nonhuman primates, we do not imply that it is entirely absent in these species, and, indeed, the possible significance of sexual coercion of females by mates (e.g., Goldizen, 1989) or male neighbors during encounters between family groups, or by strange males when they encounter lone females, or by mates when females approach territory boundaries, deserves further attention.

We focus on information from wild primates, when it is available, because wild groups are more likely to reflect socioecological conditions that obtained during the species’ evolutionary history, but we also include relevant evidence from provisioned and captive animals. Caution is necessary when such information is used to support an argument related to selection pressures in the wild. However, evidence from captive or provisioned animals can provide useful indications of behavioral potentials not typically shown in the wild, which may nevertheless reflect a species’ evolved capacity to respond adaptively to novel circumstances (R. W. Smuts, 1993).

Finally, we conclude our discussion of primates with a special section on chimpanzees. We discuss chimpanzees as a separate “case study” because much more information is available on male aggression against females in this species than in any other primate, and we wish to present this information as a coherent whole.

A FREQUENCY OF MALE AGGRESSION AGAINST FEMALES AND CONTEXTS OF OCCURRENCE

Male aggression against females is frequently mentioned in passing or briefly described in the literature on wild nonhuman primates, which suggests its widespread occurrence through the Primate order (Tracy and Crawford, 1992). However, few quantitative data are available on male aggression against female nonhuman primates. Smuts (1985) determined...