Biological and Classical theories

psychologists who study crime, but also true, despite notable exceptions, for most sociologists. Let's turn now to biological explanations of criminality to see what they might offer to the understanding of crime.

The first positivist research on crime was primarily biological. Today biological ex- planations enjoy a renewed popularity but remain controversial because of their so- cial policy implications and occasional methodological shortcomings. We'll examine several types of biological explanations from over the years and then review the controversy.

Phrenology One of the earliest biological explanations of crime, phrenology, concerned the size and shape of the skull and was popular from the mid-1700s to the mid-1800s (Jones 1986). An Austrian physician, Franz Gall (1758-1 828), was its major proponent. Gall thought that three major regions of the brain (intellectual, moral, and lower) govern three distinct types of behavior and personality characteristics. The lower region was associated with criminal behavior and would be largest in criminals. Since phrenologists could not directly measure the size of the three brain regions, they reasoned that the size and shape of the skull corresponded to the brain's size and shape. They thus thought that skull dimensions provided good evidence of criminal tendencies.

As an explanation of crime, phrenology was popular for a while but never re- ally caught on. We now know, of course, that its assumptions were mistaken: The brain neither works as compartmentally as phrenologists thought nor can be mea- sured by measuring the skull. But perhaps the most important reason phrenology faded was that its biological determinism clashed with the classical emphasis on free will, still popular in the early 1800s. The determinism of positivism didn't be- come widely accepted until much later that century.

Cesare Lombroso: Atavism If Cesare Beccaria was the founder of the classical school of criminology, then Ce- sare Lombroso~(1835-1909), an Italian physician, was the founder of the positivist school (Wolfgang 1972). Influenced by Darwin's work on evolution, Lombroso thought criminals were atavists, or throwbacks to an earlier stage of evolution, and saw criminality as the result of atavism. In essence, criminals were evolutionary ac- cidents who resembled primitive people more than modern (i.e., nineteenth-cen- tury) people. Lombroso felt that atavistic criminals needed to be executed since any other punishment would not work. After doing extensive research, Lombroso published his atavist theory in 1876 in his famous book, LfUorno Delinquent (The Criminal Man) (Lombroso 1876). What was Lombroso's evidence for his theory of atavism? He measured the bodies of men in Italian prisons and decided that they looked more like primitive men than modem men. Among other things, their arms were abnormally long, their skulls and jaws abnormally large, and their bodies very hairy.

. CHAPTER 5 / Explaining Crime: Emphasis on the Individual + 1 1 5

4 Cesare Lombroso believed that criminals were atavists, or evolutionary accidents. His views were influenced by Danuin's theory of evolution and i n turn led other scholars to study whether criminals are biologically difierent from noncriminals.

Given the intense interest in evolution generated by Darwin's work, Lom- broso's ffdiscovery" attracted much attention and his atavist theory of crime became quite popular. Unfortunately, Lombroso's research was methodologically flawed (Vold, Bernard, and Snipes 1998). Because the Italian criminal justice system at the time was hardly a fair one, it's likely that many of his prisoners had not actually committed crimes. His control groups probably included people who had com- mitted crimes without being imprisoned, as is still true today. Many differences he found between his prisoners and control group subjects were too small to be sta- tistically significant. It's also possible that Lombroso unconsciously measured his subjects in ways that fit his theory. Even if we were to assume that his prisoners did look different from how men in a proper control group would have looked, it's possible that their imprisonment resulted more from fear of, and bias against, their unusual appearance than from their criminality. Finally, some of the traits Lom- broso described characterize Sicilians, who have long been at the bottom of Italy's socioeconomic ladder. Lombroso's prisoners might have looked like atavists not be- cause his theory made any sense but because his atavistic traits happened to be ones belonging to Sicilians.

By the end of his career, Lombroso had modified his view of atavism. While he continued totthink that the most serious criminals were atavists, he reasoned that this group comprised only about one-third of all offenders. The remainder were criminals who developed brain problems long after birth, and a large category of "occasional" criminals whose behavior stemmed from problems in their social en- vironment. Two of Lombroso's students, Raffaele Garofalo (1852-1934) and Enri- co Ferri (1856-1929), carried on his views and made their own contributions to criminology's development. Garofalo continued to emphasize biological bases for crime, while Ferri stressed that social conditions also play a role. Both scholars at- tacked the classical view of free will and crime, and argued for a more positivist, determinist view of crime causation.

As the founder of modern positivist criminology, Lombroso left a lasting lega- cy, as his assumption that criminals were biologically different continues to guide today's biological research on crime (Vold, Bernard, 'and Snipes 1998). It should

116 + CHAPTER 5 / Explaining Crime: Emphasis on the Individual

come as no surprise, however, that his atavist theory has long been discredited. In 191 3 English psychiatrist Charles Goring (1 870-19 19) published his book The Eng- lish Convict. Goring measured the body dimensions of 3,000 English prisoners and of the members of a large control group. He didn't discover the differences be- tween the two groups that Lombroso uncovered and thus found no support for atavism. Goring did find, however, that the criminals he studied were shorter and less heavy than control group subjects and thought this indicated a hereditary basis for crime. This possibility notwithstanding, Goring's critique of Lombroso's re- search struck home, and Lombroso's theory soon fell out of favor.

In Chapter 3 I noted that until recently few criminologists studied women crimi- nals. Lombroso was one of these few. That's the good news. The bad news is that his explanation of female criminality, reflecting the sexism of his time, rested on antiquated notions of women's biology and physiology. Lombroso published The Female Offender in 1895. In it he wrote that women were more likely than men to be atavists and that "even the female criminal is monotonous and uniform com- pared with her male companion, just as in general woman is inferior to man" (Lombroso 1920 [1903]:122). He also thought that women "have many traits in common with children," that their "moral sense is deficient," and that "they are revengeful, jealous."

In view of these terrible qualities, how did Lombroso explain why women commit so little crime? He reasoned that women were naturally passive and viewed their "defects [as] neutralized by piety, maternity, want of passion, sexual cold- ness, weakness and an undeveloped intelligence." A woman who managed to com- mit crime despite these crime-reducing traits must be, thought Lombroso, "a born criminal more terrible than any man," as her "wickedness must have been enor- mous before it could triumph over so many obstacles" (Lombroso 1920 [1903]:150-152). Although most modern criminologists consider Lombroso's views hopelessly outdated, his emphasis on women's physiology and supposed biolog- ical nature remains influential in the study of women's crime and other behaviors (Chesney-Lind 1997; Klein 1995).

Earnest Hooton: Biological Inferiority After Goring's 1913 refutation of Lombroso's atavism theory, criminologists tem- porarily abandpned the idea that criminals were physiologically different. Then in 1939 Harvard University anthropologist Earnest Hooton (1887-1954) revived in- terest in physiological explanations with the publication of two books, The Ameri- can Criminal: An Anthropological Study (1939a) and Crime and the Man; (1939b). In these books Hooton reported the results of his measurement of 14,000 male pris- oners and 3,200 control group subjects. Compared to the control group, Hooton said, prisoners tended to have, among other things, low foreheads, crooked noses, narrow jaws, small ears, long necks, and stooped shoulders. Not one to mince words, Hooton labeled criminals "organically inferior" and "low-grade human or- ganisms" and concluded that the "primary cause of crime is biological inferiority.. . . The penitentiaries of our society are built upon the shifting sands and quaking bogs of inferior human organisms" (Hooton 1939b:130). He further concluded that criminals' body shapes influenced the types of crime they committed. Murderers

CHAPTER 5 / Explaining Crime: Emphasis on the Individual + 117

tended to be tall and thin, for example, while rapists were short and heavy. Men with average builds did not specialize in any particular crime because they, like their physical shape, had no specific orientation.

Hooton's belief in the biological inferiority of criminals led him to urge the government to reduce crime by undertaking "the extirpation of the physically, mentally, and morally unfit, or . . . their complete segregation in a socially aseptic environment" (Hooton 1939a:309). Put more simply, Hooton was advocating that the government sterilize criminals or exile them to reservations.

Hooton's research suffered from the same methodological flaws as Lombroso's, including the assumptions that all his prisoners had committed crimes and accu- rately represented criminals, and that all his control group subjects were non- criminals. It's also doubtful that Hooton's control group adequately represented the general population, since a majority were either firefighters or members of the Massachusetts militia. Given their occupations, their physical fitness, size, and so forth may well have differed from those of the population at large. Because of these and other weaknesses, Hooton's work did not become popular, especially with the onset of World War I1 and the "extupation" of the millions of people whom the Nazis considered biologically inferior.

William Sheldon: Body Shapes Although assumptions of biological inferiority grew less fashionable, interest in physiology and criminality continued. In 1949 William Sheldon (1898-1977) pub- , lished his book, Varieties of Delinquent Youth, in which he outlined his theory of so- i i matology (Sheldon 1949). This theory assumes that people's body shapes affect 1 their personalities and hence the crimes they commit. Sheldon identified three such body types (see Figure 5.1). Endomorphs are heavy with short arms and legs; they

Endomorph Mesomorph Ectomorph

A FIGURE 5.1 WILLIAM SHELDON'S THREE BODY SHAPES WILLIAM SHELDON A ~ R I B U T E D CRIME TO PEOPLE'S BODY SHAPES, WHICH HE THOUGHT

INFLUENCED THEIR PERSONALITIES AND THUS THE CRIMES THEY COMMITED.

S o u ~ c e : Adapted from Sheldon 1949.

\,~ 118 + CHAPTER 5 / Explaining Crime: Emphasis on the Individual

tend to be relaxed and extroverted, and relatively noncriminal. Mesomorphs are ath- letic and muscular; they tend to be aggressive and particularly apt to commit vio- lent crimes and other crimes requiring strength and speed. Finally, ectomorphs are thin, introverted, and overly sensitive. Sheldon compared 200 male delinquents in an institution to a control group of some 4,000 male college students. Compared to the students, the delinquents tended to be mesomorphic, as Sheldon predicted.

Despite some appeal to his theory, Sheldon's research suffered from the same methodological flaws of the work of Lombroso, Hooton, and other early biologists (Curran and Renzetti 1994; Vold, Bernard, and Snipes 1998). In addition, even if Sheldon's delinquent subjects were more mesomorphic, he couldn't rule out the possibility that their muscular, athletic bodies made it more probable that they worried juvenile justice officials and hence more likely that they would be institu- tionalized. These many flaws, coupled with memories of the Holocaust, minimized the popularity of Sheldon's somatological theory of crime.

Family, Heredity, and Genes Biologists and medical researchers have long noticed that crime tends to "run in families" and naturally assume that criminal tendencies are inherited (Rowe and Farrington 1997). To these researchers, crime is analogous to disease and illness. Just as many cancers, heart disease, and other medical problems are often genetically transmitted, so, they say is crime and, for that matter, other behavioral problems such as alcoholism and schizophrenia. Work on heredity and crime now occupies a central place in biology and crime research. Let's review early and current efforts in this area.

The first notable study of family transmission of crime was The Jukes: A Study in Crime, Pauperism, Disease, and Heredity, published in 1877 by Richard Dugdale (Dug- dale 1877). Noticing that six members of the Jukes family were behind bars in rural New York, Dugdale researched their family tree back 200 years and found that about 140 of 1,000 Jukes during that time had been imprisoned. Unfortunately, Dugdale had no control group and thus could not determine whether the Jukes' level of criminality was higher than other families'. Henry H. Goddard's 1912 study, The Kallikak Family: A Study in the Heredity of Feeblemindedness, was sounder in this regard (Goddard 1912). Goddard examined the descendants of one Martin Kallikak, who fathered children through two different women in the late 1700s. Goddard found a higher'proportion of crime and other problems in one set of Kallikak's de- scendants than in the other. Despite the interesting control group, learning and en- vironmental factors explain Goddard's findings better than heredity. The "deviant" set of Kallikak's descendants, for example, lived in poverty, while the "normal" set lived in wealth.

The ideal way to study heredity and crime would be to take individuals at birth, clone them genetically, and randomly assign them and their clones to families (one individual or clone per family) living in various kinds of circumstances across the country. You would then monitor the individuals' and clones' behavior for the next

CHAPTER 5 / Explaining Crime: Emphasis on the Individual + 1 1 9

forty years or so. At regular intervals throughout this long study, you would de- termine whether individuals and clones tend to act alike. If crime is inherited, then individuals who commit crime should have clones who also commit crime. For each individual-clone pair, you would thus determine whether (1) both members of the pair commit crime, (2) both members do not commit crime, or (3) one mem- ber commits crime and the other doesn't. When both members of a pair act alike, as in (1) or (2), we have concordance; when they don't act alike, as in (3), we have discordance. If crime is inherited, you would find a higher level of concordance than discordance in all your individual-clone pairs; if crime is not inherited, you would find similar levels of concordance and discordance.

For better or worse, in the real world we can't do such an "ideal" study. Juras- sic Park and other science fiction notwithstanding, we can't yet clone whole di- nosaurs or humans, despite recent advances in cloning involvingother animals. Even if we could clone humans, we wouldn't be allowed to assign babies and their clones randomly to families across the land. The same holds true for identical twins, who are the genetic equivalent of clones.

One way to approximate this ideal study of heredity and crime is to compare identical twins who continue to live with their natural parents with siblings who are not identical twins and thus not genetically the same. We can then determine whether the level of concordance between the identical twins is higher than that be- tween non-twin siblings. Researchers have performed several such studies and usually find higher concordance among the identical twins than among the other siblings. This evidence is widely interpreted as supporting a genetic basis for crime (Jacobson and Rowe 2000).

However, critics argue that other reasons may account for the concordance. Compared to other siblings, identical twins spend more time together, tend to have the same friends, are more attached to each other, and tend to think of themselves as alike. They are also more likely than other siblings to be treated the same by their parents, friends, and teachers. AU these likenesses produce similar attitudes and behaviors among identical twins, including delinquency and crime (Walters 1992; Walters and White 1989).

To rule out these possibilities, some researchers study identical twins separated shortly after birth and raised by different sets of parents. Because the twins do not live together, any concordance must stem from genetic factors. However, identical twins separated at birth are very rare, and too few studies exist to infer a genetic basis for crime. Their results are also mixed: Some find a high level of concordance, while othersldo not. Moreover, most of the identical twins in these studies who were reared "separately" were usually raised by parents who were close family members or neighbors. The twins thus lived in roughly the same environments, with many of them even spending a lot of time with each other. Since the twins were not really raised that separately after all, any concordance found may simply reflect their similar environments and not their genetic sameness (Lewontin, Rose, and Kamin 1984).

Other researchers look at non-twin siblings who through adoption are raised by different sets of parents. In this kind of study researchers determine whether nat- ural parents who are criminals tend to have children, though adopted and raised by other parents, who also become criminals (Carey 1994). These studies usually find that the criminality of natural parents is statistically related to the criminality of their adopted children. For example, a study of almost 1,000 adopted boys born

120 + CHAPTER 5 / Explaining Crime: Emphasis on the Individual

in Denmark between 1927 and 1941 found that 49 percent of boys with criminal records had natural fathers with criminal records, versus only 31 percent of boys without such records (Hutchings and Mednick 1977). A similar study of some 4,000 adopted Danish males found criminal conviction rates of 24.5 percent among those with natural parents who had been convicted, versus only 14.7 percent among those with natural parents who had not been convicted (Mednick et al. 1987).

Although many researchers interpret such evidence as support for a genetic basis for crime, others argue that many siblings in adoption studies are adopted sev- eral months after birth and thus experience similar environmental influences be- fore adoption at a critical stage of their development. These influences might thus account for any similarity found later between their behavior and their natural parents'. Another problem is that adoption agencies usually try to find adoptive parents whose socioeconomic status and other characteristics match those of the natural parents. The resulting lack of random assignment in adoption studies cre- ates a bias that may account for the statistical relationships found (Walters and White 1989).

Despite the methodological problems in heredity and crime research, many biologists and some criminologists continue to feel that the "heritability of ag- gression and nonviolent offending is substantial," to quote a recent review (Jacob- son and Rowe 2000:341). Others are optimistic that the research will one day prove a genetic link but concede that it hasn't yet done so (Fishbein 1996). Still other scholars criticize any rush to judgment on heredity and crime. As another review put it, "Current genetic research on crime has been poorly designed, ambiguous- ly reported, and exceedingly inadequate in addressing the relevant issues" (Wal- t e r ~ and White 1989:478). Even if crime does "run in families," socialization and role modeling, not heredity, may well be the reasons (Widom 1992). Given the method- ological problems in heredity and crime research, a genetic basis for crime cannot yet be assumed. Even if a strong genetic link is one day established beyond a doubt, the social environment also plays an important role in crime, as biological propo- nents readily point out (Jacobson and Rowe 2000).

Before leaving the world of genetics, we should touch briefly on the issue of ab- normal chromosomes. As you might remember from your biology classes in high school and college, each person normally has 23 pairs of chromosomes, or 46 chro- mosomes altogether. The twenty-third pair determines the sex of the child at the moment of conception. Two X chromosomes (XX) mean the fetus will be female, while one X chromosome and one Y chromosome (XY) mean it will be male. Al- though sperm,usually carry either one X or one Y chromosome, occasionally a sperm will carry two X's, two Y's, neither an X nor a Y (designated "O), or both an X and a Y. The chromosome pattern that results in a fertilized egg will thus be either XXX, XYY, XO, or XXY respectively.

The pattern that most interests some criminologists is XYY, which was dis- covered in 1961 and is found in fewer than 1 of every 1,000 men. Compared to nor- mal, XY men, XYY men are more likely to be tall with long arms and severe acne and to have low intelligence. The relatively few studies of XYY men find that they are considerably more likely than normal XY men to be arrested or imprisoned, mainly for petty thefts (Carey 1994). Because the XYY abnormality is so rare, how- ever, sample sizes in these studies are quite small. Some who view the XYY ab- normality as a cause of crime attribute this link to the low intelligence of XYY men. However, others feel that their arrests and imprisonment are more the result of

CHAPTER 5 / Explaining Crime: Emphasis on the Individual + 12 1

bias against their unusual and even menacing appearance. In any event, because the XYY abnormality is so rare, at most it explains only a very minuscule fraction of crime (Ellis 1982).

Neurochemical Mechanisms The human body is filled with many kinds of substances that act as "chemical mes- sengers" to help its various parts perform their functions. Since these functions in- clude behavior, biologists have tried to determine the role chemical substances might play in crime. Two substances that have received considerable attention are hormones and neurotransmitters.

In the human body, endocrine glands secrete hormones into the blood, which then transports them throughout the body. After arriving at the various organs or tissue for which they're intended, hormones enable certain functions to occur, including growth, metabolism, sex and reproduction, and stress reaction. One of the most popular modern biological explanations of crime centers on testosterone, the so- called male hormone. As Chapter 3 indicated, men commit much more crime than women. As you undoubtedly already know, men also have more testosterone than women. Combining these two basic sex differences, many biologists and other scholars interested in crime argue that testosterone, or, to be more precise, varia- tion in the amount of testosterone, is an important cause of male criminality. Testos- terone differences explain not only why men commit more crime than women but also why some men commit more crime than other men.

Ample evidence exists of a correlation between testosterone level and aggres- sion or criminality. In the animal kingdom, testosterone has often been linked to ag- gression; among humans, the sex difference both in testosterone and in crime is obvious. Many studies also find that male adolescents and adults with records of violent and nonviolent offending have higher testosterone than males with no such records (Brain 1994; Dabbs and Morris 1990). A recent study of 4,462 Vietnam-era male veterans found a testosterone-criminality relationship. After measuring men's testosterone levels and interviewing them about their offending at various stages in their lives, the researchers found "a moderately strong and significant relation- ship between testosterone and adult deviance" (Booth and Osgood 1993).

What accounts for the relationship between testosterone and offending? Biol- ogists and other scholars cite several mediating factors that increase the chances of juvenile and adult crime (Booth and Osgood 1993). In particular, higher testos- terone is 'thought to increase aggression and perhaps risktaking and impulsive- ness, and thus also low self-control, all important components of delinquency and crime. These effects may also reduce the interpersonal bonds that inhibit offend- ing (see Chapter 7), again leading to higher rates of deviance. As you can see, testos- terone is thus thought to interact with many social factors in producing criminality. In conjunction with the evidence of a testosterone-criminality correlation, this in- teraction of the biological and the social makes a hormonal explanation of crimi- nal behavior very appealing.

However, several methodological problems indicate the testosterone expla- nation may well be a Trojan horse that is appealing on the outside but flawed upon closer inspection. Consider, for example, the common assumption that testos- terone produces aggression throughout the animal kingdom. Although this link

122 + CHAPTER 5 / Explaining Crime: Emphasis on the Individual

is commonly found, it's also true that in many animal species, among them guinea pigs and lions, females are more aggressive than males, although they have lower testosterone. Moreover, neuroendocrinologists who study hormones and behav- ior caution against extrapolating from animal studies to human behavior. Although it's true that hormones strongly affect many behaviors of lower animals, including primates, the human central nervous system is so complex that simple endocrine influences cannot be assumed.

The evidence among humans of testosterone-induced offending is also open to question. Although many studies have found a link between testosterone and ag- gression or offending, other studies have found no such link. As you know from Chapter 1, moreover, correlation does not mean causation. A correlation among human males between high testosterone and high offending does not necessarily mean that testosterone affects offending. Metl~odologically, it is just as plausible that offending affects testosterone, or that some third factor leads to both high testosterone and high offending. In the animal kingdom, for example, aggression and dominance lead to high testosterone in certain species. Although this has not been widely investigated among humans, it's possible that delinquency and adult criminality lead to feelings of dominance and thus to higher testosterone (Miczek et al. 1994).

The sex difference in testosterone and criminality is also obviously subject to other interpretations. As Chapter 3 discussed, sex-,role socialization produces dif- ferent behaviors in girls and boys and different opportunities for offending. To most sociologists, a testosterone-based explanation of the gender difference in crime seems much less plausible than one based on social and structural factors.

In view of these problems, a significant effect of testosterone on human ag- gression cannot be assumed. A recent review commissioned for the National Acad- emy of Sciences concluded that the testosterone-aggression correlations often found among human males "are not high, they are sometimes difficult to replicate, and importantly, they do not demonstrate causation. In fact there is better evidence for the reverse relationship (behavior altering hormonal levels). . . . [Wlinning--even in innocuous laboratory competitions-can increase testosterone" (Miczek et al. 1 994: 6-7).

HORMONES: PMS AND CRIME BY WOMEN One final hormonal explanation of criminality focuses on women. In some women, hormonal changes in the days before menstruation appear to be linked to increased stress, tension, lethargy, and other problems. These women are said to suffer from premenstrual syndrome, or PMS. Thinking this emotional condition might lead to aggression and*other offending, some researchers study whether crime by women tends to occur in their premenstrual phase. If PMS were not related to women's crime, their offending would occur randomly throughout their menstrual cycles. If PMS did lead women to offend, their deviance would tend to occur in their pre- menstrual phase (see Figure 5.2 on page 124).

To study this possibility, researchers have asked women in prison to think back to when they committed the offense for which they were arrested and to remem- ber the dates of their menstruation. From this information researchers can deter- mine whether offenses occurred randomly throughout the women's cycles or instead were concentrated in their premenstrual phase. A leading researcher in this field, Katharina Dalton, has found such a concentration, with about half of the pris- oners she studied reporting they committed their offenses in the eight-day period immediately preceding and during menstruation (Dalton 1961).

CHAPTER 5 / Explaining Crime: Emphasis on the Individual + 123

Egg Release and

Endometrial Changes

Estrogen Progesterone

Possible ~;$~,~;;;p:;;;g~~~,~. ,,,?!

Possible 3.(:> !&$p$@)p:<::*i:':;; :zci:P g3!gg4f;q xeF LI, -,.v. 5; L?,> " Possible Fertility @t~%$~~<:~~<:,:$;;> ,-+,," , - Fertility

<>+* ,,!$~i$~$;>y~~;>p2$:,,, , ;?,$,

A FIGURE 5.2 THE MENSTRUAL CYCLE SOME STUDIES ATTRIBUTE CRIMES BY WOMEN TO PREMENSTRUAL SYNDROME, BUT METHODOLOGICAL PROBLEMS CAST DOUBT ON THIS CONCLUSION.

Dalton attributed their criminality to their emotional condition and increased lethargy and clumsiness during this time: Their emotional condition prompted them to commit the crimes, and their lethargy and clumsiness made it more diffi- cult for them to avoid detection and arrest. To support her view of women's phys- ical ineptitude, Dalton noted that half of women drivers involved in serious auto accidents are also in the eight-day premenstrual-menstrual phase. Dalton's and other researchers' findings have led some criminal defense attorneys to claim PMS as a defense when women have been tried for various crimes. In England in 1980, for example, one woman murdered her boyfriend by driving her car into him, and another killed a co-worker in a London pub. Both claimed that PMS led to their vi- olence, and both received probation instead of imprisonment.

As you might expect, the PMS explanation for women's crime is quite contro- versial. Many scholars feel that it takes us back to the myths of "raging hormones," when women were considered unfit to be airplane pilots, president of the United States, and other positions because they could not be trusted to act rationally dur- ing "that time of the month."

Beyond these ideological concerns, the PMS research is also methodologically flawed, perhaps fatally (Horney 1978; Katz and Chambliss 1995). It assumes that women can accurately remember when menstruation occurred, since even a few days' error can place their crime outside of their premenstrual phase. Some women are very regular and can remember the dates of their menstruation but others can- not. More important, it is well known that stress and other problems can disrupt women's cycles, with menstruation occurring either sooner or later than expected. If the stress of committing crime, or the stress leading up to the crime, hastens men- struation, it may appear artificially that the crime occurred in a woman's premen- strual phase only because menstruation occurred sooner than normal. In recalling Dalton's finding that half of all women drivers in serious accidents were in the eight-day premenstrual-menstrual phase of their cycles, consider that half of all women passengers involved in accidents are in the same phase (Horney 1978). As

124 + CHAPTER 5 / Explaining Crime: Emphasis on the Individual

Janet Katz and William J. Chambliss (1995:290) aptly put it, "Unless we wish to argue that the passenger's lethargy somehow caused the accident, it would ap- pear that the trauma of the accident triggered menstruation, not vice versa."

The human nervous system consists of billions of cells called neurons and bundles of neurons called nerves. Nerves carry messages from the brain throughout the body and messages from the body back to the brain and spinal cord. Certain neu- rons called receptors are found in the sense organs of the body such as the eye. Receptors send messages, or impulses, through nerves back to the brain. After ob- taining these messages, the brain sends instructions for particular actions back to various parts of the body. Neurons transmit impulses to each other across synaps- es with the aid of chemical substances called neurotransmitters. In studying ag- gression, scientists have been particularly interested in one particular neurotransmitter, serotonin (Jacobson and Rowe 2000).

In animal studies, low levels of serotonin are linked with higher levels of ag- gression. Many studies of humans have found low levels of serotonin in violent of- fenders (Moffitt et al. 1998). Although some researchers find the serotonin research particularly interesting, it suffers from several of the methodological problems al- ready discussed, including inconsistent measurement of offending and the possi- bility that serotonin levels result from aggression rather than the reverse, which make it premature to assume a strong role for serotonin in human aggression. Some studies even find higher levels of serotonin in aggressive individuals. A recent re- view concluded that "serotonin is not a very discriminating marker for violence" (Wallman 1999:24).

Diet and Nutrition In late 1978 a San Francisco city supervisor named Dan White allegedly murdered George Moscone, the city's mayor, and Harvey Milk, a city supervisor and gay ac- tivist. I lived near San Francisco at the time and will never forget how much the murders shocked the Bay Area. People even stopped shopping for Christmas pre- sents for several days, as the whole community shared in collective grief. When White was tried for the two murders, his attorney claimed he had been eating too much junk food. The sugar and various additives in the food supposedly deep- ened his depression and reduced his ability to tell right from wrong. White's "Twinkie defense" worked, as he was convicted only of manslaughter, not first- degree murder. His conviction on the lower charge outraged Bay Area residents (Weiss 1984).

As this example indicates, diet and nutrition are popularly thought to play a role in aggression and crime (Kanarek 1994). Researchers investigate this role with two kinds of studies. In the first kind, they control the levels of various nutrients given to animal or occasionally human subjects in the laboratory and then compare the behavior of these subjects to control groups. In the second kind, they compare the diet and nutrition of offenders, usually juveniles, to those of nonoffenders. From this body of evidence several diet and nutritional factors have been identi- fied as producing aggression and other forms of offending. High amounts of sugar and refined carbohydrates, excessive levels of chemical additives, and deficiencies in vitamin B and other vitamins have received the most attention.

CHAPTER 5 / Explaining Crime: Emphasis on the Individual + 125

Despite the common-sense appeal of this research, it, too, suffers from sever- al methodological problems that cast doubt on its findings, including small Sam- ples, possibly spurious findings, and ambiguity in defining offending. Moreover, several studies of diet and nutrition do not find them linked to antisocial behav- ior. A review of the research concluded that diet and nutrition have at most a "rel- atively minor" effect on criminality (Kanarek 1994:535).

Pregnancy and Birth Complications Some of the most interesting biological research concerns the effects of pregnancy and birth complications. These complications are often referred to as perinatal prob- lems. Poor nutrition or the use of alcohol and drugs during pregnancy are thought to harm fetal development, with potentially long-lasting effects on central nervous system (CNS) functioning that in turn can lead to antisocial behavior. For example, a recent study found that boys born to women who smoked during pregnancy were twice as likely as those whose mothers did not smoke to be arrested for a vi- olent crime by age 34 (Conlon 1999). CNS functioning can also be impaired by com- plications during difficult births. Although the evidence is mixed, some studies , find such neurodevelopmental problems to be linked to later onset of childhood and i adolescent behavioral problems, especially violent aggression. A recent review I noted that "delivery complications appear to be more predictive of later criminal behavior than pregnancy problems or low birthweight" (Brennan, Mednick, and Volavka 1995:76).

Although this body of research suggests a biological role in offending, other in- terpretations arr possible. In particular, pregnancy and birth complications may often be the fault of mothers who don't observe standard advice for promoting fetal health and development. As biological researchers Elizabeth Kandel and Sarnoff A. Med- nick (1991:526) observe, "Such poor prenatal mothering might be related to poor mothering after birth, which would render the birth complications-violent behavior association spurious." To rule out this possibility, postnatal parenting would have to be observed, or adopted babies with perinatal complications studied.

Smoking during pregnancy may impair fetal development, - causing potentially long-lasting effects on central nervous system functioning, Although the evi- dence is mixed, some studies find that such neurodevelopmental problems are linked to the later onset of childhood and adolescent behavioral problems, especial- ly violent aggression.

126 + CHAPTER 5 / Explaining Crime: Emphasis on the Individual 1 I

CHAPTER 5 / Explaining Crime: Emphasis on the Individual + 127

Critique of Biological Explanations Many biologists and some criminologists are enthusiastic about the potential of biological theories to explain crime (Fishbein 1998; Jeffery 1994; Moffitt et al. 1998). Most sociologists and other criminologists are more wary (Akers 1997; Curran and Renzetti 1994; Vold, Bernard, and Snipes 1998). Let's look at some of their concerns.

One problem is that crime is simply too diverse. Even if biological factors ac- count for some violent aggression, they wouldn't explain the vast majority of criminality. Among other problems, they can't easily account for the "relativity" of deviance; that is, they can't explain why someone with a biological predispo- sition to violence turns to street crime instead of, say football or any other activity involving physical force. As McCaghy and Capron (1994:43) point out, violence "in a bar brawl" makes you a criminal; violence "on a battlefield" makes you a hero. A biological explanation of violence is thus not the same thing as a biolog- ical explanation of criminal violence.

Another problem is methodological. As our review of biological explanations indicated, several methodological problems-including small, unrepresentative samples of offenders, inadequate control groups, and correlations between bio- logcal factors and offending that are subject to many interpretations-make it dif- ficult to infer any firm conclusions from biological research.

A third problem concerns what I call "group rate differences." As we saw earlier in this chapter, sociologists are interested in different crime rates from one group or setting to another. Thus we are less interested, for example, in why a par- ticular individual in a big city commits a crime than in why urban areas have higher crime rates than rural areas. Biological explanations cannot easily account for group rate differences. Take the fact that the United States has a much high- er homicide rate than Western European nations. How would you explain this bi- ologically? Is it really conceivable that U.S. residents are different biologically from residents of England, Germany, or Denmark in a way that leads to more crime in the United States? Can the high crime rates of big cities as compared to rural areas really be attributed to biological problems in big-city residents? Can biological explanations account for why street crime rose in the United States during the 1960s and fell in the 1990s? Even if they might explain why some in- dividuals commit crime, they cannot easily account for different crime rates among groups or locations or for changes in crime rates. As one observer notes, the explanation of these kinds of trends "will come not from research in neuro- biology but from understanding economic forces, variation in cultural mores re- garding the acceptability of violence, diversity in the kinds and efficacy of means of informa1,and formal social control, and availability of firearms" (Wallman 1999:24).

A final concern about biological explanations addresses their social policy implications. One implication is that to reduce crime we have to do something about the biological deficiency that causes it. However, short of some science- fiction world that, thankfully, does not yet exist, we can't easily change biology. And if we can't change biology, we can't reduce crime. Say, for example, that bio- chemical deficiencies explain why people commit crime. If so, what can we do to reduce crime? Perform some genetic engineering? Give them drugs to correct the deficiencies? Given the rapid rise in scientific advances, some of these measures are quickly becoming possible but remain rather frightening. Perinatal research is a notable exception. If it turns out to be true that perinatal problems predict later offending, then it may be possible to reduce crime with social policies aimed at

128 + CHAPTER 5. / Explaining Crime: Emphasis on the Individual

better prenatal health care and nutrition, especially among the poor, where preg- nancy and birth complications are more common.

Responding to the concern that biological explanations imply little chance for reducing crime, some researchers stress that biological traits interact with en- vironmental influences to produce crime: Biological factors may predispose in- dividuals to crime, but the extent and timing of their influence depends on environmental factors (Fishbein 1996; Tibbetts and Piquero 1999). Efforts to change the social environment thus hold much promise for reducing crime. Al- though this "softer" position is more compatible with a sociological framework, it still suggests the need to do something about the biological traits. Thus some scholars recommend, for example, that children be screened for biological traits that may lead to later criminality (Jeffery 1994; Nelkin 1993). Proposals like this raise several ethical and other concerns, including the possibility that children tar- geted in this fashion may be "labeled" as potential criminals (see Chapter 8) and treated that way.

A related problem with biological explanations centers on their potential jus- tification for appalling acts committed against people regarded as biologically different. It's a short step from considering people biologically different to view- ing them as biologically inferior (Rafter 1997; Shipman 1994). History is replete with acts of genocide, lynchings, hate crime, and other actions taken against peo- ple deemed biologically inferior to some ruling group, with their supposed in- feriority justifying the inhumane treatment. Hundreds of thousands of Native Americans were murdered in what came to be called the United States by white Europeans who considered them subhuman. Millions of Africans were brought to the New World in chains and kept for two centuries in slavery. In the early decades of this century, the eugenics movement in the United States led to the in- voluntary sterilization of some 70,000 people, almost all of them poor and many of them African-American (Rafter 1997). Not too long after, Nazi Germany slaugh- tered millions of Jews and others who were thought inferior to the Aryan race. The "evidence" gathered by American eugenicists of biological inferiority rein- forced Nazi ideology (Kuhl1994).

If, then, we find that a biological trait makes certain people more likely to be criminals, history tells us it's very easy for these groups to be considered bio- logically inferior and in need of special, even inhumane, treatment, even if no biologists today advocate such treatment. These groups are usually the poor and people of color, as biological research on crime has centered on street crimes com- mitted by the poor, ignoring the white-collar crimes by wealthier people. It might sound silly to you even to suggest that a defective gene or hormonal imbalance leads corporate executives to engage in price fixing or to market unsafe prod- ucts. Yet, given bur society's prejudices, it might not sound as silly to suggest that a biological problem leads poor people to commit acts of common violence. Many of the early biological researchers were prejudiced against the poor, im- migrants, and people of color and interpreted their findings as "proof" of these groups' biological inferiority (Gould 1981 ). Given continuing racial and ethnic prejudice, we must be very careful in interpreting the findings of contemporary research on biology and crime.

One sociologist wrote a decade ago, "In my opinion, criminologists ultimately must come to grips with biological hypotheses and findings" (Gibbons 1992:7). As we've just seen, however, many sociologists continue to be critical of these same hypotheses and findings. The value of biological explanations will un- doubtedly remain a major source of controversy in criminology for some years to come.

CHAPTER 5 / Explaining Crime: Emphasis on the Individual + 129

As discussed earlier, psychology offers valuable understanding of individual be- havior but says little about the larger social and structural forces also at work. In the area of crime, sociology and psychology together provide a more comprehen- sive explanation than either discipline can provide separately. Sociology tries to explain why certain groups and locations have more crime than others, while psy- chology may be able to tell why a few people with these backgrounds commit se- rious crime while most do not (Andrews and Bonta 1999; Bart01 1999). Let's look at the major psychological explanations for criminal behavior, leaving learning ap- proaches, which are more compatible with a sociological framework, for a later chapter (Chapter 7).

Psychoanalytic Explanations Modern psychoanalytic explanations see delinquency and crime rising from in- ternal disturbances developing in early childhood because of interaction problems between parents and children. These explanations derive from the work of Sig- mund Freud (1856-1939), the founder of psychoanalysis (Freud 1935 [1920]; 1961 [1930]). Although Freud focused more on mental disorders than on criminality per se, his work provided a logical foundation for extensions into delinquency and crime by later theorists. Freud and his followers see mental disorders arising from a conflict between society and the instinctive needs of the individual. The indi- vidual personality consists of three parts: the id, the ego, and the superego. The id, present at birth, consists of instinctual desires that demand immediate gratification: Infants get hungry and don't take no for an answer if they're not fed soon enough. Eventually the ego develops and represents the more rational part of personality. A child learns that he or she can't always expect immediate gratification of his or her needs. The superego comes later and represents the internalization of society's moral code. This is the individual's conscience and leads the individual to feel guilty or ashamed for violating social norms. The development of these three parts of the personality is generally complete by about age 5.

Freud thought that people are inherently pleasure-seeking because of the id, but that too much pleasure-seeking can translate into antisocial behavior. The ego and superego thus need to restrain the id. This happens in mentally healthy individu- als as the three parts of the personality coexist harmoniously. A lack of balance can result when a child's needs for food, emotional comfort, and the like are not met because of parental deprivation, neglect, or overly harsh discipline. According to psychoanalytic theory, if the superego becomes too weak to control the id's in- stinctive impulses, delinquency and crime result. They can also result if the super- ego is too strong, when individual~ feel overly guilty and ashamed. The rational part of the personality, the ego, realizes that if individuals commit a crime they will be punished and thus reduce their guilt. Given this realization, the ego leads the person to break the law.

Psychoanalytic explanations have been valuable in emphasizing the impor- tance of early childhood experiences for later behavior, but their value for under- standing crime is limited for several reasons (Vold, Bernard, and Snipes 1998). First, they suggest that antisocial behavior is mentally disordered behavior, which isn't true for most individuals. Second, they neglect social factors and overemphasize

130 + CHAPTER 5 / Explaining Crime: Emphasis on the Individual

childhood experiences; while these are undoubtedly important, later life-cycle in- fluences are also important (Laub and Sampson 1993). Finally, psychoanalytic re- search relies on case histories of individuals under treatment or on samples of offenders in juvenile institutions, adult prisons, or mental institutions. Such methodology ignores the possibilities that the subjects might not represent the vast majority of offenders not under treatment or institutionalized and that any men- tal or emotional problems in the institutionalized subjects may be the result, and not the cause, of their institutionalization.

Before we leave psychoanalytic explanations, a comment on their view of women's criminality is in order. Although Freud is widely regarded as one of the three or four greatest thinkers of the last two centuries (along with Marx, Darwin, and Einstein), his views on women reflected the sexism of his day (Klein 1973). Freud viewed child rearing as women's natural role in life and thought that fe- males who could not ad just to this role suffered from "penis envyff and hence men- tal disorder. To compensate for the lack of a penis, some women, he thought, tried to act like men in desiring careers. Extending Freud's views to delinquency and crime, Freudian scholars later attributed most girls' delinquency to their sexual needs. In a traditional Freudian framework, then, women and girls with mental disorders or histories of crime and delinquency need to be helped to adjust to their natural child-rearing roles. Thanks to critiques by feminist scholars, this view of fe- male criminality lost popularity in the 1970s but hasn't disappeared.

Moral Development and Crime Since the time of Jean Piaget (1896-1980), psychologists have been interested in children's mental and moral development. Piaget thought that children experience four stages of mental development. The sensorimotor period lasts until the age of two and involves learning about their immediate environment and developing their reflexes. The preoperational period lasts from ages two to seven and consists of learn- ing language, drawing, and other skills. A stage of concrete operations lasts from ages seven to eleven and involves learning logical thinking and problem solving. The final formal operations stage occurs during ages eleven to fifteen and concerns dealing with abstract ideas (Pulaski 1980).

Following in Piaget's footsteps, psychologist Lawrence Kohlberg (1969) de- veloped his theory of moral development. This is the development of the ability to dis- tinguish right from wrong and to determine the ethically correct course of action in complex circumstances. Kohlberg theorized that individuals pass through sev- eral stages in which they develop their ability to reason morally. In the early stages, children's moral reasoning is related solely to punishment: Their view of what is correct behavior is equivalent to behavior that keeps them from getting punished. In later stages they begin to realize as adolescents that society and their parents have rules that deserve to be obeyed in and of themselves, not just to avoid pun- ishment. They also realize that exhibiting the behaviors expected of them will lead others to view them positively. In the final stages of moral development during late adolescence and early adulthood, people recognize that universal moral prin- ciples supersede the laws of any one society. Individuals reaching this stage may decide to disobey the law in the name of a "higherf' law.

Kohlberg further theorized that not everyone makes it through all the stages of moral development. In particular, some people's moral development stops after only the early stages. Because their view of right and wrong is limited to what

CHAPTER 5 / Explaining Crime: Emphasis on the Individual + 13 1

avoids punishment, they haven't developed what many of us would call a con- science and may well engage in harmful behavior as long as they think they won't get punished for it. Kohlberg thus thought that incomplete moral development was a major reason for criminal and other antisocial behavior. Studies by Kohlberg and others of the level of moral reasoning in samples of offenders and nonoffend- ers support his theory (Henggeler 1989; Kohlberg 1969).

One problem with tests of Kohlberg's theory is the familiar chicken-and-egg question of causal order. Even if offenders do have a lower level of moral reason- ing than nonoffenders, it's possible their offending affected their moral reason- ing rather than the reverse. They might have begun to violate the law for other reasons, such as peer pressure or hostility toward their parents, and then adjust- ed their moral reasoning to accommodate their illegal behavior to minimize any guilt or shame.

Intelligence and Crime Researchers have long blamed crime on low intelligence. Studies early in this cen- tury by Goddard and others found low IQs among prisoners and juveniles in re- form schools. Scholars later criticized this research for using small, unrepresentative samples and unreliable tests, and it lost popularity by the 1930s (Gould 1981). In the late 1970s, however, an article by Travis Hirschi and Michael Hindelang in a pre- eminent sociology journal renewed interest in the IQ-crime issue (Hirschi and Hin- delang 1977). They reviewed many studies using both self-report and official data and found that delinquents' IQ scores were about eight points lower than non- delinquents' scores. The authors concluded that low IQ is an important cause of delinquency.

Although findings are not always consistent, later studies also link delin- quency to low IQ. Some scholars go as far as to say that IQ is a stronger predictor of delinquency than either race or social class and that blacks' low native intelli- gence is the major reason they commit more street crimes than whites (Herrnstein and Murray 1994).

Most research on IQ and offending focuses on adolescents, for whom several reasons are thought to explain the apparent causal link between IQ and delin- quency. First, youths with low intelligence do poorly in school. Poor school per- formance in turn leads to less attachment to school and more alienation from it, and thus to higher rates of delinquency. Second, low intelligence leads to a lower abil- ity to engage in moral reasoning and to delay gratification, increasing the likelihood of offending. Third, adolescents with low intelligence are thought to be less able to appreciate t h ~ consequences of their actions and to be more susceptible to the in- fluence of delinquent friends (Hirschi and Hindelang 1977; Lynam, Moffitt, and Stouthamer-Loeber 1993).

If some studies link low intelligence to crime, other research proposes that low intelligence is inherited. Dugdale's early study of the Jukes and Goddard's study of the Kallikaks both assumed that low intelligence is passed biologically from one generation to the next. Research on heredity and intelligence became more so- phisticated early in this century with the development of the IQ test and its ad- ministration to different groups of people. Finding different average IQ scores among these groups and assuming that IQ tests measure natural intelligence, many researchers concluded that IQ is an inherited trait and that different kinds of peo- ple differ in natural intelligence. For example, the low scores of Polish and Russ- ian immigrants to the United States who took the test in the early 1900s indicated

132 * CHAPTER 5 / Explaining Crime: Emphasis on the Individual

to researchers that they were innately less intelligent than whites from Anglo-Saxon backgrounds (Gould 1981).

Over the years, black-white differences in IQ scores have also been found, with the average scores of blacks about 10 to 15 points below those for whites (Wilson and Herrnstein 1985). Some researchers interpret blacks' lower scores as evidence that they are naturally less intelhgent than whites, while others think the scores stem from poor prenatal care among blacks, which leads permanently to lower in- telligence (Herrnstein and Murray 1994; Jensen 1969; Wilson and Herrnstein 1985).

Suppose we accept the assumptions guiding the research linking IQ, race, and crime: (1) IQ tests are valid measures of natural intelligence; (2) blacks are intel- lectually inferior to whites; (3) low natural intelligence produces higher rates of delinquency and crime; (4) low natural intelligence is an important and perhaps the major reason for high rates of street crime by blacks. What do these beliefs imply about efforts to reduce such crime?

Since they discount social factors, they suggest, first of all, that efforts to reduce social inequality and other structural problems would do relatively little to reduce black crime rates. Since they emphasize blacks' low natural intelligence as a primary cause of their criminality, they also imply that to reduce black crime rates we have to improve their innate intelligence. But to say that intelligence is innate or natur- al suggests that efforts to raise it will probably be useless; given this futility, we can do little to reduce African-American crime. Since we cannot reduce it, perhaps all we can do is to deter blacks from committing crime by putting even more of them in prison. Some might even say we should sterilize blacks because their low intelligence poses a menace to the rest of society.

This is certainly a pessimistic appraisal, and perhaps a bit simplistic, but is it warranted? Not if the assumptions turn out to be questionable or even false, which is precisely what many critics charge (Gould 1981; Lewontin, Rose and Kamin 1984; Menard and Morse 1984). Without question, the early IQ research was rife with methodological problems; much of it was carried out by researchers with views we would now call racist. Recent IQ studies are more carefully designed, with many using twins and adopted children, but still suffer from the same problems af- fecting heredity and crime research. As a result, their findings on IQ, race, and crime are suspect. In one particular problem, IQ studies often use samples of of- fenders in adult prisons or juvenile institutions. Because incarcerated offenders represent only a very small proportion of all offenders, we cannot safely general- ize these studies' findings to the entire offender population.

Perhaps more important, many critics charge, IQ tests are culturally biased and less a measure of intelligence than of white, middle-class background or of school achievement. Instead of reflecting low natural intelligence, then, blacks' low IQ scores may simply reflect their poorer schooling and the fact that they are not white and usually not middle class.

In sum, race-IQ-crime assumptions are highly questionable at best and patent- ly false at worst, with dangerous racial and class overtones. Although it's too early to rule out the assumed intelligence-criminality link in the race-IQ-crime chain, history tells us we must tread very cautiously in this area.

Personality and Crime Some of the most important work today in psychology and crime focuses on tem- perament, or personality, and aggressive and other antisocial behavior. In an early study, Eleanor and Sheldon Glueck administered Rorschach (ink blot) tests to 500

CHAPTER 5 / Explaining Crime: Emphasis on the Individual + 133

delinquents and the same number of nondelinquents matched on several charac- teristics, and found greater personality problems in the delinquents (Glueck and Glueck 1950). Other research began to use personality inventories, most often the Minnesota Multiphasic Personality Inventory (MMPI) and the California Psycho- logical Inventory (CPI), administering them to samples of incarcerated juvenile and adult offenders. These inventories list several hundred true/false and other items (e.g., "I would do almost anything on a dare") to which subjects respond; the CPI includes about one third of the MMPI items. These studies found person- ality differences between offenders and nonoffenders that are assumed to be re- sponsible for the offending.

Much of the current research focuses on childhood temperament, as many studies, some of them longitudinal, link temperament problems during infancy and childhood with behavioral problems during this time and also with later delin- quency during adolescence. The long list of temperament problems includes such things as attention deficits, impulsiveness, hyperactivity, irritability, coldness, and suspiciousness. Children with temperament problems are more likely to become delinquent in unstable families marked by inadequate parenting than in stable ones where parents are loving and supportive. Although most children with tem- perament problems do not commit serious delinquency during adolescence, the ones with the worst problems are more likely to become delinquent. Further, most serious delinquents are thought to have had childhood temperament problems (Caspi et al. 1994; Farrington 1998).

This body of work has important implications for reducing crime. If early child- hood temperament problems do matter, then attempts to reduce them may reduce delinquency and crime (Zigler, Taussig, and Black 1992). Such efforts include preschool and early family intervention programs. If we wait until adolescence to work with juvenile offenders, it's often too late. As Cathy Spatz Widom and Hans Toch (Widom and Toch 1993:263) observe, "By the time [juveniles] are treated, often after referral by court personnel, typically they have been involved in a long his- tory of antisocial behavior and conduct problems that are not reversed easily."

Although the personality research is appealing, several problems limit its ap- plicability (Einstadter and Henry 1995; Tennenbaum 1977). Perhaps most important,

Although most children with tempera- ment problems do not commit serious delinquency during adolescence, the ones with the worst problems are more likely to become delinquent. Furthermore, chil- dren with temperament problems are more likely to become delinquent in un- stable families marked by inadequate par- enting than in stable ones where parents are loving and supportive.

134 + CHAPTER 5 / Explaining Crime: Emphasis on the Individual

several studies of personality and offending find no differences between offenders and nonoffenders. Also, because most of the research examines juvenile offenders in institutions, the offenders' personality problems may be the result of their in- stitutionalization and not the cause. The personality-offending link may even re- sult from an effect of offending on personality traits rather than the reverse. Because many studies don't control for socioeconomic status, education, and other charac- teristics, their correlation between personality and criminality may also be spuri- ous. In one further methodological problem, the validity of the MMPI and other inventories for assessing the effects of personality on criminality is also open to question, since some of their items ask subjects about their offending. Thus a re- spondent who admits to offending automatically provides an answer indicating a personality problem.

A recent study of youths in Pittsburgh and New Zealand sought to overcome these problems by measuring personality without regard to criminality and by using random samples and self-reports of offending (Caspi et al. 1994). The re- searchers found personality differences linked to delinquency. Youths expressing negative emotionality-a tendency to react to stress with fear and anger-and/or high impulsiveness were more likely to be delinquent than youths with neither trait. Despite its methodological sophistication, however, this study didn't deal ad- equately with the question of causal order and did not eliminate the possibility of spuriousness.

A final problem with personality research is that personality explanations of crime, like their biological counterparts, can't adequately account for the "relativ- ity" of deviance: They cannot explain why individuals psychologically predisposed to thrill-seeking or violence undertake criminal actions instead of legal ones (McCaghy and Capron 2000). Some people with the impulsiveness trait mentioned previously, for example, may pursue car racing or parachute jumping as careers or hobbies; others may choose crime. Personality explanations do not help us under- stand why one behavior instead of the other is chosen.

Taken together, these problems mean we must be cautious in drawing con- clusions from the personality-crime research. As a recent review noted, "While it might be comforting to think that the criminal is a special type of person that can be distinpshed from noncriminals through [psychological] testing, in reality it appears that . . . there are few if any differences between these groups" (Curran and Renzetti 1994:109). Many psychologists, of course, disagree with this assessment, and personality research will no doubt continue to challenge conventional crimi- nological thinking. In this regard, the research project in New Zealand mentioned previously has been extremely valuable, as it has followed subjects from childhood through age 21. Its longitudinal design allows it to assess whether various problems during childhood predict delinquency and other problems several years later. As the International Focus box on page 136 discusses, this project's findings provide a powerful case for the argument that personality and other problems during child- hood do help lead to delinquency and other problems later on. It also underscores the need to try to prevent these problems from emerging in the first place.

Critique of Psychological Explanations Some psychologists criticize the "antipsychological bias" they see in sociological criminology (Andrews and Bonta 1999). While true to some extent, this charge is also too severe. Although sociologists certainly look beyond the individual, many

CHAPTER 5 / Explaining Crime: Emphasis on the Individual 135

of their structural explanations for crime rest on social-psychological states such as frustration and alienation (see Chapter 6). Many of the social process theories fa- vored by sociologists (see Chapter 7) also rest on psychological concepts such as learning and role-modeling.

In certain respects, psychological explanations complement sociological ones in explaining crime. By highlighting the importance of negative childhood expe- riences, especially within the family, for later delinquency and criminality, they

136 CHAPTER,^ / Explaining Crime: Emphasis o n the Individual

begin to fill in the smaller picture of crime that sociology's structural approach leaves empty. As we'll be seeing in the next two chapters, moreover, several psy- chological concepts play an important role in sociological theories of crime.

Despite the contributions that some psychological approaches make to our un- derstanding of crime, several issues remain (Curran and Renzetti 1994; Vold, Bernard, and Snipes 1998). First, psychological studies often use small, unrepre- sentative samples of offenders in prisons or mental institutions. Even if these of- fenders are psychologically different, the difference may be the result of their institutionalization and not the cause. Second, psychological studies generally dis- regard structural factors such as poverty and cannot easily account for variations in crime by group and location or for changes in crime rates.

Third, although they offer interesting statistical correlations, their causal order remains unclear. In this regard, the recent longitudinal studies of early childhood temperament and later delinquency have been very valuable, as their research de- sign allows us to conclude that temperament problems precede initial delinquen- cy, even if delinquency might later in turn affect temperament. The key is to discover why young children have temperament problems. Biologists think many of these are genetic in origin and trace them to biochemical defects. In a recent ex- ample of such research, two sets of scientists from the United States and Israel iden- tified a genetic basis for a personality trait they called "novelty-seeking." People with this trait tend to be extroverted, impulsive, and quick to anger (Angier 1996). We can expect that future research will seek to discover whether the gene the sci- entists found is disproportionately present in criminal subjects. For their part, as we have seen, many psychologists trace personality problems to early childhood experiences. While acknowledging the influence of these experiences, sociologists would stress that the most negative experiences occur amid poverty and other structural problems conducive to crime (see Chapter 6).

In an additional problem, psychologists of crime join their biological counter- parts in rarely studying crimes by white-collar offenders, even though these crimes may involve injury and death. That researchers and the public continue to be at- tracted to suggestions of psychological abnormality in common criminals may re- flect popular views of the poor and people of color more than the actual existence of abnormality.

Psychological approaches also suggest that crime and criminals are both psycho- logically abnormal. Normal people donf t commit crime; abnormal people do. Emile Durkheim (1962'.[1895]), one of the founders of sociology (Chapter I), wrote that crime and deviance are indeed normal, meaning that they occur in every healthy society because people will always violate the norms of any society. Building upon Durkheim's perspective, sociological criminology sees crime and deviance arising from normal social structures, institutions, and processes. Since psychological ex- planations feel that individuals have problems that lead them to commit crime, they suggest the way to reduce crime is to cure the few aberrant individuals who commit it. As we saw with eating disorders, a sociological perspective suggests there will always be other deviants to take their place given the social and struc- tural forces at work.

It may also be mistaken to view most criminals as psychologically abnormal. It's very possible to commit horrible violence and still be psychologically normal

CHAPTER 5 / Explaining Crime: Emphasis on the Individual + 137

in other respects. Studies after World War I1 of prison guards in Nazi concentra- tion camps found them to be good husbands and fathers who performed well on various psychological tests. Despite their apparent psychological normality, they were able to commit some of the worst crimes known to humanity.

Two famous psychological experiments are telling in this regard. The first took place in the early 1960s when Yale University psychologist Stanley Milgram re- cruited Yale students and residents of Bridgeport, Connecticut, to administer elec- tric shock in a learning experiment (Milgram 1974). Subjects were asked to apply electric shock to "learners" who performed poorly in word-pair tests. The inspi- ration for Milgram's experiment was the Nazi Holocaust, which was attributed to an "authoritarian personality" that was said to be the result of German culture, history, and socialization. Milgram planned to conduct his experiment in both Ger- many and the United States but was stunned to find his U.S. subjects all too ready to administer electric shock to learners who were screaming in pain and pleading for them to stop. Unknown to the subjects, no electric shock was actually used. The learners they saw were all actors, and the learners they heard over the loud- speaker were all part of a tape recording. Because it indicated that psychological- ly normal U.S. residents were quite capable of inflicting serious injury on innocent people, Milgram's experiment attracted wide attention and remains controversial to this day.

The other experiment was conducted at Stanford University, where psycholo- gist Philip Zimbardo constructed a mock prison in the basement of a psychology building (Zimbardo 1972). He recruited volunteers from male Stanford students and, after eliminating those with histories of illegal drug use or mental and emo- tional problems, randomly assigned the remainder to be either guards or prison- ers in the mock prison. The "prisoners" were "arrested" by real police, booked, and fingerprinted, and then taken to the mock prison. In the prison they wore uni- forms and stockings on their heads to remove their individuality, and were each given a number to replace their names.

By the end of the first day of the experiment, the "guards" were already treat- ing the prisoners harshly, verbally abusing them and forcing them to stand at at- tention hours on end, do calisthenics, and the like. Within another day or so the prisoners refused to come out of their cells (revamped laboratory rooms). The rebellion prompted further abuse by the guards. At the end of a few more days, one of the prisoners suffered a nervous breakdown and had to be convinced he wasnf t really a prisoner. With more prisoners suffering similar symptoms, Zim- bardo decided to end his experiment prematurely. Zimbardo later wrote that the behavior of both the guards and the prisoners was pathological in many ways but could not be due to preexisting problems in the individuals comprising both groups, since he screened them for such problems and then randomly assigned them to the groups. Instead the behavior was due to the structural conditions and role expectations of the mock prison experience that led normal people to be- have unacceptably.

All the major theories discussed in this chapter focus on the individual. Classical and neoclassical perspectives (rational choice) assume that individuals commit crime when they decide the potential gains outweigh the potential costs. Biological and

138 + CHAPTER 5 / Explaining Crime: Emphasis on the Individual

psychological jheories attribute crime to individual biological or psychological at- tributes. \

Ultimatels your view of the world influences the value you find in these ex- planations. If you think that people are responsible for their own behavior, then you'll probabb prefer rational choice views. If you think that social problems arise from individual faults and problems, then you will probably pre- fer biological dnd/or psychological theories. If instead you think that social rob- lems derive primarily from problems in the larger social structure, then you will probably prefer sociologicalexplanations. While there's certainly room for more than one way to understand criminality,, the explanation we adopt has impor- tant implications for efforts to reduce crime.

Rational choice views attribute crime to the choices individuals make about their own behavior. Crime policies based on these views aim to affect these choic- es by making punishment more certain and more severe. It's unclear, however, whether the 'decision making of potential criminals follows the rational choice model, as well as whether increasing the certainty and severity of punishment c ~ s reduce crime rates significantly. The world of rational choice and deterrence the- ory is, further, largely devoid of social inequality and social structure. To the ex- tent these social facts generate criminality, the rational choice model ignores important sources of crime.

Biological and psychological explanations both ultimately locate the gene- sis of crime inside the individual. They suffer from common methodological problems, including small sample sizes, difficulties in distinguishing offend- ers from nonoffenders, and, despite some recent longitudinal studies, ambigu- ity in causal order. Hence, these interpretations have not yet established a strong role for biological or psychological factors in criminality. Several explanations in the two disciplines also minimize the importance of social and structural fac- tors for delinquency and crime. Even if the-biological evidence were more con- clusive, sociologists would continue to be troubled by its implications for social policy on crime. Several psychological approaches are more compatible with a sociological framework but still miss the larger crime picture on which sociol- ogy focuses.

Historically, biological research and, in its work on intelligence, psychologi- cal research have had damaging consequences for women, the poor, and people of color. Recent writings on intelligence and heredity indicate that views on the inferior intelligence of certain g o u p s have not yet lost favor. Sensitivity to the dangers of these and similar views demands that biological and psychological ev- idence of criminality be interpreted cautiously.

Despite the limitations of biological and psychological research from a soci- ological standpoint, this research has been valuable in stressing the importance of early childhood for later delinquency and criminality. Certain approaches in both fields focus on childhood medical and psychological problems stemming from poor prenatal health and nutrition and inadequate parenting. These ap- proaches suggest that programs focusing on families at risk for both sets of prob- lems can achieve significant crime reduction.

Because this risk is greatest for families living in poverty, this line of biolog- ical and psychological work complements sociological attention to the crimino- genic effects of poverty and other structural conditions in the social environment that we'll explore in the next chapter. Chapter 7 examines social process theories in sociology that stress negative childhood social experiences, which again are more common for families living in poverty. Notwithstanding their differences,

CHAPTER 5 / Explaining Crime: Emphasis on the Individual + 139

contemporary efforts in biology psycholpgy, and sociology thus all underscore the crime-reduction potential of well-designed and well-funded efforts that ad- dress the causes and consequences of poverty.

I

abnormality atavism classical school concordance

deterrence theory discordance

ego Enlightenment heredity id

IQ moral development

neurotransmitter

personality phrenology positivism premenstrual syndrome psychoanalytic rational choice theory

somatology superego temperament testosterone

1. How were criminals treated during the Age of Reason? How may the Clas- sical School of Criminology be considered a reaction to this treatment?

2. What concerns do many sociologists and criminologists have about bio- logical explanations of criminal behavior? How valid do you think these concerns are?

3. What does the research on intelligence and crime tell us? What are the methodological and other critiques of this line of research?

4. The question of abnormality versus normality lies at the heart of much psy- chological research on criminal behavior. Do you think many criminals are psychologically abnormal?

The text discusses a good deal of psychological research on crime and delinquen- cy. The Web site of the American Psychological Association (APA) includes infor- mation from its Public Policy Office on youth violence. To access this information, go to http://www.apa.org/ppo/violence.html. Scroll down until you reach the sec- tion entitled Are some children just prone to violence? What does this section say about whether violent behavior is inherited?

140 + CHAPTER 5 / Explaining Crime: Emphasis on the Individual