«Why Men Commit Crimes (and Why They Desist)* Satoshi Kanazawa Indiana University of Pennsylvania Mary C. Still Cornell University Hirschi and ...»
Why Men Commit Crimes (and Why They Desist)*
Indiana University of Pennsylvania
Mary C. Still
Hirschi and Gottfredson (1983) claim that the relationship between age and crime is
similar in all social and cultural conditions and that no current sociological or criminological theory can account for this similarity. We introduce the new field of evolutionary psychology and extend Daly and Wilson’s (1988) work on homicide to construct
a general theory of male criminality, which explains why men commit violent and property crimes. The theory can also explain the age-crime curve. It might also account for some empirical anomalies such as why physically smaller boys are more delinquent, and why violent criminals desist more slowly.
In their highly influential 1983 article “Age and Explanation of Crime,” Hirschi and Gottfredson claim that the relationship between age and crime is invariant across all social and cultural conditions at all times. In every society, for all social groups, for all races and both sexes, at all historical times, the tendency to commit crimes and other analogous, risktaking behavior rapidly increases in early adolescence, peaks in late adolescence and early adulthood, rapidly decreases throughout the 20s and 30s, and levels off during middle age.
Figure 1 presents the typical age-crime curve ~Hirschi and Gottfredson 1983: Figures 1, 5, 7–8!. Although some of their claims have been controversial and there have been some minor variations observed around the invariant age curve ~Greenberg 1985; Hirschi and Gottfredson 1985; Steffensmeier et al. 1989!, the essential shape of the curve for serious interpersonal crimes remains uncontested in the criminological literature.
While Hirschi and Gottfredson claim that the age curve is invariant and holds in all societies at all times, they provide no explanations for this universal observation. They instead argue that no theoretical or empirical variable currently available in criminology can explain it. If the age-crime curve is truly constant across all populations, then any factor that varies across such populations cannot explain it. Just as a constant cannot explain a variable, a variable cannot explain a constant. The invariant age curve must be explained by something that is constant across all societies and cultures.
Some sociologists have provided just such explanations for why crime peaks in late adolescence and early adulthood. However, these explanations typically suffer from two *We thank Deborah Blum, Mark E. Bouton, David M. Buss, Rebecca Lynn Churchill, Barbara J. Costello, Martin Daly, Lee Ellis, Paula England, David P. Farrington, Christine Horne, Allan Mazur, Paul J. Quirk, David C. Rowe, Anthony Walsh, and Margo Wilson for their comments on earlier drafts. This paper was completed while the first author was Summer Scholar in the Institute on Violence and the Life Course ~co-directed by Kenneth A. Dodge and Robert J. Sampson! at the Center for Advanced Study in the Behavioral Sciences. He thanks the Andrew W. Mellon Foundation for the financial support. We dedicate this paper to Travis Hirschi.
Travis’ numerous achievements during his illustrious career as a criminologist include the publication of a highly influential book, Causes of Delinquency ~1969! early in his career, the presidency of the American Society of Criminology in 1983, and the very successful collaboration with Michael R. Gottfredson later in his career on many books and articles, on one of which we focus in this paper. One of the few low points in his career was having to serve on the dissertation committee of the first author. Direct all correspondence to: Satoshi Kanazawa, Department of Sociology, Indiana University of Pennsylvania, Indiana, PA 15705-1087. E-mail: Kanazawa@ grove.iup.edu.
problems. The first is that the explanation simply begs the question. For instance, Gove ~1985: 134! notes that “many adolescents find their roles lacking in intrinsic rewards and turn to sensate activities to achieve a sense of self.” Gove argues that this “need for stimulation,” combined with heightened levels of testosterone, leads many adolescents to commit crime. But why is it that adolescents find criminal activities stimulating and fulfilling but adults do not? Why is it that they have a need for stimulation that adults ~presumably!
Another problem with these explanations is that their applicability is limited to contemporary industrial societies. Gove ~1985: 138! and Walsh ~1995: 184–85! explain crime among adolescents as a function of the combination of high autonomy and low responsibility during the teenage years. Such a combination accurately describes teenagers in the United States and other western nations today, and thus Gove and Walsh might be able to explain the relationship between age and crime in the western society in the latter half of the twentieth century. However, if Hirschi and Gottfredson are right and the age-crime curve is universal in all societies at all times, then it should hold both in the developing nations today and the western nations in the past. Any explanation applicable only to the contemporary western societies is inadequate.
Other explanations of the age-crime curve tend to be incomplete. Grogger ~1998! presents an explicitly decision-theoretic model in which actors choose between crime and labor market as alternative means of earning money. Among other things, he uses his model to explain the age-crime curve. While his model successfully explains why property crime precipitously drops in early adulthood ~while their market wages and opportunity costs of crime rapidly increase!, it cannot explain why crime participation increases during early to middle adolescence. Nor can his model explain violent crimes. There has therefore been no satisfactory theory of crime that can explain all aspects of the curve ~both the incline and decline over the life course! and is applicable to all crimes in all societies at all times.
In this paper, we will first introduce the emerging field of evolutionary psychology ~Barkow, Cosmides and Tooby 1992!, and then propose a theory of male criminality derived 436 SOCIOLOGICAL THEORY from this perspective. The theory explains, among other things, the Hirschi-Gottfredson age-crime curve. By the age-crime curve we mean the relationship between age and criminality where: 1! criminality rapidly rises during adolescence; 2! it peaks in late adolescence or early adulthood; 3! it rapidly decreases during adulthood; and 4! it remains very low for the rest of the life. While the age-crime curve is not exactly identical for all crimes in all societies ~Greenberg 1985; Steffensmeier et al. 1989!, these four features are present for all serious interpersonal crimes ~except such minor crimes as gambling, fraud, and drug use! in all societies for which data are available. We seek to explain these four features of the age-crime curve.
Before we present the theory, however, it is important to state its limitations explicitly.
First, ours is a theory of criminality, not crime or criminal behavior. Crime is a function of criminality ~propensity to commit crime! and external factors ~opportunities and constraints! ~Hirschi and Gottfredson 1986: 58!. As the routine activity approach ~Cohen and Felson 1979! demonstrates, whether someone with criminal propensities actually commits a crime depends largely on the circumstances, and one of us has elsewhere explained differential crime rates among nations in terms of such situational and institutional factors ~Hechter and Kanazawa 1993!. Our current theory only explains why people want to commit crimes, not whether they actually do so.
Second, ours is a theory of male criminality, and it does not explain female criminality at all. To the extent that women commit crimes ~although women commit very few serious crimes worldwide!, our theory is not able to explain it. However, given how inherently different men and women are, we would not expect men and women to commit crimes for the same reason.1 Parsimony is not as important a criterion for theory as logic and evidence ~Kanazawa 1998: 197n!. Our theory only explains some of the reasons why men might want to commit crimes.
Third, our theory explains intraindividual variations in criminality, not interindividual variations. We explain why men at certain ages or in certain life-stages are more criminal than the same men would otherwise be. Our evolutionary psychological theory cannot explain why some men are inherently and consistently more criminal throughout their lives than others. In Moffitt’s ~1993! terminology, our theory explains the age-crime curve of those who are criminal only during adolescence, not why some people are criminal only during adolescence and others are throughout their lives. Behavior-genetic ~Ellis and Walsh 1997; Rowe, Vazsonyi and Figueredo 1997! and developmental ~Moffitt 1993; Patterson and Yoerger 1993! theories of crime, among others, account for such interindividual differences in criminality.
PRINCIPLES OF EVOLUTIONARY PSYCHOLOGY
Because evolutionary psychology is a very new field and because sociologists in particular are unlikely to be familiar with it, we will first explain its foundational principles before we proceed to our theory derived from this perspective.2 Evolutionary psychology seeks to discover universal human nature, which is a collection of domain-specific psychological mechanisms. A psychological mechanism is an information-processing procedure or decision rule that natural selection has equipped humans to possess in order to solve a particular adaptive problem ~a problem of survival or reproduction!. Unlike decision rules in microeconomic subjective expected utility maximizaWe thank Joanne Savage for making this point.
Excellent introductions to evolutionary psychology include Barkow et al. ~1992!, Buss ~1994, 1995, 1999!,
Ridley ~1993!, and Wright ~1994!.WHY MEN COMMIT CRIMES ~AND WHY THEY DESIST! 437
tion theory or game theory, however, psychological mechanisms mostly operate behind our conscious thinking.
Male sexual jealousy is an example of an evolved psychological mechanism ~Daly, Wilson and Weghorst 1982!. Because gestation in human and most other mammalian species occurs inside the female body, males of these species can never be certain of the paternity of their mates’ offspring while females are always certain of their maternity. In other words, the possibility of cuckoldry exists only for males. Men who are cuckolded and invest their resources in the offspring of other men end up wasting these resources, and their genes will not be represented in the next generation. Men therefore have a strong reproductive interest in making sure that they will not be cuckolded while women do not share this interest. Accordingly, men have been selected to possess a psychological mechanism that makes them extremely jealous at even the remotest possibility of their mates’ sexual infidelity. The psychological mechanism of sexual jealousy solves men’s adaptive problem of paternal uncertainty. The same psychological mechanism often leads to men’s attempt at mate guarding in order to minimize the possibility of their mates’ sexual contact with other men, sometimes with tragic consequences ~Buss 1988; Buss and Shackelford 1997!.
While men and women are the same in the frequency and intensity of their jealousy in their romantic relationships ~White 1981; Buunk and Hupka 1987!, there are clear sex differences in what triggers jealousy. There is both survey and physiological evidence from different cultures to show that men get jealous of their mates’ sexual infidelity with other men, underlying their reproductive concern for cuckoldry. In contrast, women get jealous of their mates’ emotional involvement with other women because emotional involvement often leads to diversion of their mates’ resources from them and their children to their romantic rivals ~Buss et al. 1992; Buss et al. 1999!.
Note that we do not consciously choose or decide to get jealous. We just get jealous under some circumstances, in response to certain predictable triggers, but otherwise don’t know why. However, what triggers jealousy is always understandable to others, and these triggers are cross-culturally constant ~Thiessen and Umezawa 1998!; otherwise, no romance novels or romantic comedies would ever become international hits. Evolutionary psychology explains human behavior in terms of these evolved psychological mechanisms, and the preferences, desires, and emotions that they produce in us.
Evolutionary psychology is premised on two broad generalizations. The first generalization, to put it bluntly, is that there is nothing special about Homo sapiens. To put it more precisely, “certainly we are unique, but we are not unique in being unique. Every species is unique and evolved its uniqueness in adaptation to its environment. Culture is the uniquely
human way of adapting, but culture, too, evolved biologically” ~van den Berghe 1990:
428!. Human beings are just like other animal species ~Betzig 1997; de Waal 1996; Maryanski and Turner 1992!, and all the laws of nature, in particular, the laws of evolution by natural and sexual selection, apply as much to humans as they do to other species. The second broad generalization is that there is nothing special about the brain as a body part; it is just like the hand or the pancreas or any other body part. Just as a long history of evolution has shaped the hand or the pancreas to perform a specific function, so has evolution shaped the brain to perform certain tasks ~solving adaptive problems!.
The second generalization leads to a very important implication of evolutionary psychology. Just as the basic shape and functions of the hand and the pancreas have not changed since the end of the Pleistocene epoch about 10,000 years ago, the basic functioning of the brain has not changed very much in the last 10,000 years. The human body ~including the brain! evolved over millions of years during the Pleistocene epoch in the African savanna where humans lived during most of this time ~Maryanski and Turner 438 SOCIOLOGICAL THEORY
Figure 2. The Basic Theoretical Structure of Evolutionary Psychology