Does the Stanford prison study help explain the effects of imprisonment? This week you will need to pick a side. First, you should post your answer to the board (use the format described in the instru

A Reply to Haney and Zimbardo More than 25 years ago, Haney, Banks, and Zimbardo (1973) tran&rmed the basement hallway of Stanford University’s Psychology Department into a make-believe prison block where a group of male student volunteers posed either as inmates or as guards. Some of the “guards” behaved badly and some of the students “begged to be released from the intense pains of less than a week of merely simulated imprisonment” (Haney & Zimbardo, 1998, pp. 709.) The experiment was therefore aborted after just six days and nights. Apparently many who read about the Stanford Prison Experiment (SPE), as this six day venture came to be called, agreed with the authors that it had demonstrated “the way in which social contexts can influence, alter, shape, and transform human behavior” @p-709-10). Based on studies of this kind, some of them Gedanken experiments as in the following quotation from Mischel’s influential textbook, many psychologists came to believe that social learning determines personality and that social context determines behavior.

“Imagine the enormous differences that would be found in the personalities of twins with identical genetic endowment ifthey were raised apart in two different ftilies.. . .Through social learning vast differences develop among people in their reactions to most stimuli they face in daily life.” (Mischel, 198 1, p, 3 11.) It was natural, therefore, to believe that crime is largely a consequence of criminogenic contexts that could be eliminated by social engineering.

It follows also that prisons, should they be necessary at all, provide an excellent opportunity for the rehabilitation of misdirected youth through the provision of healthy social learning and a more beneficent behavioral context.

Haney and Zimbardo (1998) devote most of their article to a regretfhl discussion of the five-fold increase since the early 1970s in the proportion of Americans serving time in prison, of the change in prison policy since then from rehabilitation to mere segregation, and of what they call “the racialization of prison pain.” The enormous recent increase in the rate of imprisonment of convicted offenders was not in response to a corresponding increase in the proportion of citizens victimized by violent crime, at least not according to the National Crime Victimization Survey. This increase in numbers of inmates is therefore attributed to an apparently willful refusal by “correctional administrators, politicians, policymakers, and judicial decision makers” to appreciate “most of the lessons that emerged from the SPE” (p. 718).. According to Haney and Zimbardo, these lessons are: 1. “SPE demonstrated the power of situations to overwhelm psychologically normal , healthy people and to elicit from them unexpectedly cruel.. .behavior” (p. 719).

2. “SPE also revealed how easily even a minimalized prison could become painful and powerful” (p. 719). 3. ‘?f situations matter and people can be transformed by them when they go into prisons, they matter equally, if not more, when they come out of prison” (p.720.). 4. “In prison, explanations of disciplinary infractions and violence [should] focus more on the context in which they transpired and less on the prisoners who engaged in them” (p.720).

5. “Good people with good intentions are not enough to create a good prison.. .the SPE and the perspective it advanced also suggest that prison change will come about only when those who are outside of this powerful situation are empowered to act on it.” 6. “Finally, the SPE implicitly argued for a more activist scholarship in which psychologists engage with the important social and policy questions of the day” (p. 721).

I agree with at least some of these rather vague prescriptions, although I am astonished by these authors’ claim that a handbook for prison reform (indeed, a basic text in enlightened criminology) can be harvested from watching a handful of college students role-playing guards and inmates for six days in a Stanford basement. But I disagree strongly with some of the more specific claims or assumptions made by Haney and Zimbardo. Personality is more Important than Context The situation& model still embraced by Haney and Zimbardo is wrong. The Gedanken experiment suggested by Mischel has now been conducted by Bouchard et al (1990) and the results were opposite to Mischel’s expectation. Identical twins separated in infancy and reared apart are as similar in personality as identical twins reared together, and that is very similar indeed (Tellegen, et al., 1998).About half of the variance (more than half of the stable variance) in basic traits of temperament or personality is associated with genetic differences between people.

Anyone familiar with the realities of prison life knows that some inmates are predictably violent and dangerous while some are predictably passive or tractable, We recently obtained scores on the Multidimensional Personality Inventory (Tellegen & Waller, 1994) from 67 inmates at Minnesota’s maximum security prison, Oak Park Heights’, men whose mean age was 32.We collected MPQs also from more than 850 male twins aged 30-3 (Lykken, in press). The men in our inmate sample had been convicted of serious crimes, usually murder. Because the MPQ is a self-administered inventory and requires high school reading skills, a considerable proportion of the inmate population could not be sampled but there is no reason to think that the participants differed temperamentally from the non-readers.

Figure 1 shows the profiles of the 22 inmates scoring highest and the 22 scoring lowest on the aggression scale of the MPQ, plotted using the data from the non-criminal ’ I am indebted to Dr.Kmeth Carlson at Oak Park Heights Correctional Facility for collecting these data and sharing them with me. 4 male twins as norms.The most aggressive inmates are deviant also on most of the other MPQ scales; they are more than one SD below the normal mean on well being, achievement, and social closeness, the traits that comprise the Positive Emotionality super-factor of the MPQ. The aggressive inmates are more than one SD above the mean on stress reaction and on alienation which, with aggression, comprise the Negative Emotionality super-factor.And they are more than one SD below the normal mean on control (vs. impulsiveness), and on traditionalism, two of the traits that comprise the Constraint super-factor. The non-aggressive inmates, on the other hand, yield essentially normal profiles except for that low score on aggression and an elevation on harm avoidance.In spite of their confinement in the same ‘Painful and powerful” prison environment, these men show great variability, one from another, not only in personality but also in their tendencies to make or to stay out of trouble in that environment. INMATE PERSONALITY PROFILES90 High vetw~s Low on MPQ Aggression WBSP ACH SC SRAL AGG COHA TRAD I+ High -e-* Low I Figure 1. Mean MPQ profiles of inmates in a maximum security prison who scored highest or lowest on aggression.A T-score of 50 represents the mean for some 850 non-criminal males aged 30-33; a T-score of 70 is two SDS above the normal mean etc.. Modern Prisons are not Places of Unremitting Pain Because the six day SPE “had painful, even traumatic consequences for the prisoners [Stanford students pretending to be inmates] against whom it was directed” (p. 719), Haney and Zimbardo concluded that real prisons must have devastating psychological 5 effects upon real inmates serving long sentences.

Perhaps because they are situationists, rather than trait psychologists, they neglected our extraordinary human capacity to adapt to circumstances, good or bad.

Suh, Diener, & Fujita (1966) have shown that both positive and negative life experiences have usually lost their effect on subjective well being after six months.

A year after either winning the lottery or being permanently crippled in an accident, most people experience about the same average level of happiness that they felt before that event. In a study I did long ago in another Minnesota prison (Lykken, 1957), one inmate, the pitcher on the prison baseball team, had been paroled the previous fall. He made it back in time for the spring baseball season by the expedient of breaking the display window of a jewelry store and then leisurely collecting rings and watches until arrested on the spot.. He admitted he was happier back in prison than he’d been on the outside. INMATE PERSONALITY PROFILES High versus Low on MPQ Well Being70-- pl*. ____.__._______....*_-._____.._.___.__..___._._.__..------ .’ ,p_____*.. ___..__._._.._.--_-_-----..-..--.. v-- --s, d 20 II II II I I I 1 ’ WE3SP ACH SC SR AL AGG COHA TRAD + High -Ed Low Figure 2. MPQ profiles of the 22 men who scored highest on Well Being, and the 22 who scored lowest, among 67 inmates of a maximum security prison. The mean expected release date for our sample of Oak Park Heights inmates is the year 2030 yet, after having been there for an average period of 37 months, many of them appear to have become well-adjusted to prison life and many are surpisingly happy.

Figure 2 shows that, while the lowest-scoring third professed considerable pain and alienation, the upper-third scored higher on well being than three-fourths of our 850 non- criminal young men.Oak Park Heights is a modern prison, well run and reasonably safe 4 because the staff, rather than the inmates, are in control. The well adjusted inmates can take classes, learn skills, find peaceful ways to pass the time.I would not wish to be incarcerated at Oak Park Heights, not even if1 was made pitcher of the baseball team, but at least I could get a lot of reading done. NCVS vs. UCR DATA Violent Crimes “V -_._..__..__.___--._-_....-. I1905 Year - UCR Data -*- NCVS Data Figure 3. Trends since 1973 in violent crime as revealed by the National Crime Victimization Survey versus the FBI’s Uniform Crime Reports.The Epidemic of Imprisonment is due to an Epidemic of Crime. The National Crime Victimization Survey, on which Haney and Zimbardo rely, has summarized annually since 1973 the reports of more than 90,000 Americans over age 12 concerning whether they have been the victims of specified crimes during the past year.

These reports are from members of a stratified sample of families interviewed either in person or by telephone by (mostly female) employees of the U.S. Census Bureau.The Uniform Crime Reports, compiled by the FBI since 1929, summa&ecrimes actually reported to the police.As Figure 3 reveals, these two methods of measuring violent crime tell very different stories. The FBI data indicate an increase since 1973 of 54% (a peak increase of 73%) while the NCVS data indicate an actual decrease in violent crime of 15%. NCVS interviewers do not contact transients, people who are in hospital or in jail, nor do they venture into the more dangerous regions of the inner city.The NCVS tells us about middle-class crime while the UCR includes the rapid rise that ghetto crime has been displaying since the early 1960s. 7 It is not true, as Haney and Zimbardo would have us believe, that our current high rate of imprisonment is due merely to punitive courts and longer sentences.The reason that we have so many more men in prison now than in 1960 is that the crime rate now is several times higher than in the 1960s. And these are not only drug crimes. As Figure 4 reveals, the number of violent crimes reported to the police, divided by the U.S.

population, is currently about four times the rate in 1960. Many more crimes are reported, more arrests made, and many more men are convicted of violent crimes than 40 years ago and, fortunately for the rest of us, many of these violent criminals have been at least temporarily segregated in state or federal prisons.VIOLENT CRIMES vs.N of INMATES Rates per 100,000 U.S. Population -50% ! I I ,I 1 I 1 I1960 19651970 19751980 19851990 1995 2000 Year I- CiRIMES -a-* INMATES I Figure 4.

The increase since 1960 in the rates of violent crimes reported to the police and in the proportion of the U.S. male population serving terms in state or federal prisons.

Figure 4 reveals that the increase in the rate of imprisonment actually lagged the increase in the crime rate, beginning its acceleration only about 1980. The figure also displays the much-heralded dip in violent crime that has occurred since about 1993. The most likely explanation for this modest decline is the fact that 1.3 million potential perpetrators, compared to about 200,000 in 1970, are now behind bars. Because the typical prison inmate committed some 12 serious crimes during the year prior to his last 8 arrest (Bhunstein, Cohen, & Farrington, 1988), takii a million such men off the streets and into prison is bound to yield at least a temporary diminution in the crime rate. Haney and Zimbardo consider it “barbaric’ that we have so many men in prison.

While it is not a satisfactory solution to our crime problem, I believe with most Americans that sequestering violent crimmals is preferable to just turning them loose. Rehabilitation Does Not Work If Haney and Zimbardo are correct in what they think they learned from the SPE, everyone-including prison inmates--should respond to sociahzed environments in a socialized manner.

By creating such conditions in our prisons then, afler perhaps a fairly short period of acclimation and habituation, formerly unsocialized inmates should become accustomed to behaving lie law-abiding citizens and be ready for release.

As Haney and Zimbardo point out, it would be necessary also to provide socialized environments for these parolees to return to, adequate jobs, housing in good neighborhoods, and the like.

And there is no doubt that some inmates, after serving their time even in our current unenlightened prisons, manage to remain within the law (or at least unapprehended) after their release. Some inmates, after all, are reasonably normal, socialized persons who were unfortunate enough to be too strongly tempted; some, indeed, were actually innocent of the crimes for which they were convicted.

Because recidivism is only frequent and not inevitable, one may be led to believe that some criminals are rehabilitated even by the present system and, therefore, that many more might similarly benefit from a more enlightened correctional system.

The fact is, however, that Haney and Zimbardo cannot point to a single convincing study indicating that prison reforms designed to augment rehabilitation have ever been successful. I am not about to claim that every reasonable method has been tried. In fact, I should be very interested to see what would happen if each new inmate were to learn that his future supervisor, teacher, and disciplinarian was to be a distant computer, “John,” with whom he could communicate by means of a very sturdy keyboard and monitor inset into the wall of his cell.The computer would provide programmed learning tasks appropriate to the inmate’s ability and interests.By doing what the patient but implacable computer required, the inmate could earn more palatable food, TV time, access to a telephone, and other privileges. Only after he had achieved appropriate basic educational goals, and had demonstrated his willingness to live by the computer’s rules, would an inmate begin to be allowed to mix with other inmates and take further steps in demonstrating his improved level of socialization.

But I am not so na’ive as to claim, as Haney and Zimbardo seem to believe that, even with unlimited resources and control, I (or they) could turn Oak Park Heights into a prison with single-digit recidivism rates.A young person who has managed to reach his or her late teens almost wholly unsocialkred is likely to remain a danger to society for life. Like our talent for language, our human proclivities for socialization require to be 9 elicited, shaped, and reinforced in childhood or they may be forever lost. As Judge C.D.

Gill (1994) has wisely observed, ‘The place to fight crime is in the cradle.” The Black:White Ratio of Prison Inmates Reflects the Black:White Ratio of Criminal Perpetrators. Haney and Zimbardo, in referring to “the ratialization of prison pain,” seem to attribute the fact that nearly half of the prison inmates in the U.S. are A&an Americans to racist bias on the part of the police and the courts.They even exaggerate the discrepancy by saying, “although they represent only 6% of the general U.S. population, African American men constitute 48% of those confined to state prisons” (p. 714.) With similar logic one might say that, since men constitute only about 45% of the population of Norway, the fact that 95% of Norwegian prison inmates are male is evidence of gender bias. In another place (Lykken, 1995), I have offered a more reasonable and, I believe, a more constructive explanation for the racial discrepancy in American prisons:

Although one might suspect that the criminal justice system is quicker to arrest and to convict Black than White suspects, reports by victims of the race of the person who robbed or assaulted them correspond closely to the proportions of Blacks and Whites arrested for such crimes (J. Wilson & Herrnstein, 1985).

“In 1988, in the nation’s 75 most populous urban counties, Blacks were 20% of the general population but 54% of all murder victims and 62% of all defendants” (DiIulio, 1994). In Little Rock, Arkansas, victims of more than 80% of the violent crimes (97% of Black victims) reported during 1991 identified the assailant as Black (Uyttebrouck, 1993). Although AtZcan Americans make up only one-eighth of the population of the United States (one-third of the population of Little Rock), one gets the impression that many more than one- eighth of the perpetrators of the violent crimes that we read about daily or see reported on the television are Black and this impression is correct; in 1991, Blacks accounted for 32% of U.S. property crime and 45% of violent crime (FBI, 1992).

In 1965, when the Black illegitimacy rate had climbed to about 25%, Daniel Patrick Moynihan wrote his famous memorandum on the break up of the Black family, predicting much of the social dislocation that has since come to pass (see Rainwater & Yancey, 1967).

White illegitimacy, which was only about 5% in 1965, has now exceeded the level that, in the Black community, presaged all those dire consequences.

In his analysis of 1980 data from 150 U.S. cities, Sampson (1987) found that the strongest predictor of both homicide and robbery by Black juveniles was the local percentage of Black households headed by females. Sampson also found a similar relationship within the White community, where the percentage of families headed by females was a strong predictor of both juvenile and adult offending.

Figure 5 shows that White illegitimacy in the U.S. is rapidly catching up to the Black rate, which seems now to have reached its asymptote. 10 U.S. ILLEGITIMACY RATES~~___________-__.____.__.___.____.~_._____._._._____..___.___..__.___.__ . ..____ _ __._.____.. 70______ ___.___.___._-_.__-----__.__.___.___._.____--__-___.____.___.___.___.___.__.__.._.-__---_----.--. I- Black Women ---g White Women I Figure 5. Birth rate forunmarried women by race: United States, 1970-92. Modified from Vital and Health Statistics, Series 21: Data on Natality, Marriage, and Divorce, No.

53, DHHS Publication No. (PHS) 96-193 1, Figure 4, p. 4.

It can be shown that, computed separately for Blacks and Whites, a youngster reared without the resident participation of the biological father is about 7 times more likely in consequence to become delinquent and then criminal (Lykken, 1997). Fatherless rearing results, across racial lines, in similarly increased risk for child abuse, for teenage runaway, for school dropout, teenage pregnancy, and welfare dependency. Summary While I agree with Haney and Zimbardo that psychologists should try to play a stronger and more constructive role in advising those responsible for social policy, I am persuaded that the vague and politically correct nostrums that they recommend cannot be helpful. We have too many men (and increasing numbers of women) in prison because we have too much crime.

We have too much crime because an ever-increasing proportion of our children are reaching adolescence essentially unsocialized.

Some of these youngsters can be described as psychopaths, meaning that their innate temperaments from early childhood made them very difficult to manage, too difficult for the average parent. 11 But crime has increased far too rapidly to be attributable to dysgenic factors. Most of these troublesome youth are what I call sociopaths, meaning that their rearing environment failed to elicit, shape, and reinforce their inherent human capacity to develop an effective conscience as well as their instincts of empathy and altruism and social responsibility.

All but the most difficult children become adequately socialized in the extended-family environments of traditional societies that most resemble the environment of human evolutionary adaptation in which our ancestors evolved their innate talent for social living.

Such traditional societies have very little crime.

Crime rates increased when the child-rearing practices of modern societies deviated from those to which our species had become adapted and we began to entrust the responsibility of socializing children almost entirely to young parental couples most of whom are untrained and inexperienced.

Now more than a third of all American infants are being raised without even the help and support of a resident biological father. I believe that this social revolution, which began earlier among African Americans but the White community is catching up, is the root cause of our current epidemic of crime and other social pathology.

If Haney and Zimbardo wish to give useful advice to policy makers, I suggest that they forget about the SPE and consult instead Jack Westman’s important book Licensing Parents. REFERENCES Blumstein, A., Cohen, J., & Farrington, D.P. (1988). Criminal career research: Its value for criminology. Criminology, 26, l-37.

Bouchard, T.J. Jr., Lykken, D.T., McGue, M., Segal, N.L., & Tellegen, A. (1990).

Sources of human psychological differences: The Minnesota Study of Twins Reared Apart. Science, 250,223-228. DiIulio, J.D. Jr. (1994, Fall). The question of Black crime. The Public Interest, 3-32. Eunkook Suh, Ed Diener, and Frank Fujita (1996).

Events and subjective well-being:

Only recent events matter. Journal of Personality and Social Psychology, 70, 1091-l 102. FBI (1992). Unijbrm crime reports for the United States, 1991. Washington, DC: U.S.

Department of Justice.

FBI (1993). Age-speczjk arrest rates and race-spec$c arrest rates for selected oflenses, 19651992, Washington, DC: U.S. Department of Justice.

Gill, C. D. (1994). In the Foreword to J. C. Westman, Licensing parents. New YorkPlenum @viii) Haney, C., Banks, W. & Zimbardo, P. (1973). Interpersonal dynamics in a simulated prison International Journal of Criminology and and Penology., I, 69-97. Haney, C. & Zimbardo, P. (1998). The past and future of U.S. Prison policy: Twenty- five years after the Standord Prison Experiment. American Psychologist, 53,709- 727.. Lykken, D. T. (1995). The AntisociaZ Personalities. Hillsdale, NJ. Erlbaum.

Lykken, D.T. (1997). Factory of crime. Psychological Inquiry, 8, 261-270. 12 Lykken, D.T. (in press).The causes and costs of crime and a controversial cure. Journal of Personality. Mischel, W. (198 1). Introduction to personality (3”‘. Ed). New York: Holt, Rinehart & Winston National Center for Health Statistics (1993a). Vital statistics of the United States, 1989, Vo2.L Nat&y. DHHS Pub. No. (PHS) 93-l 100. Washington, DC: U.S.

Government Printing Office.

National Center for Health Statistics (1993b). Vital statistics of the United States, 1989, Vol. IIL Marriage and divorce. DHHS Pub. No. (PHS) 93-1103. Washington, DC:

U.S. Government Printing Office.

Rainwater, L., & Yancey, W. (1967). The Moynihan Report and the politics of controversy. Cambridge, MA: The MIT Press.

Sampson, RJ. (1987). Urban black violence: The effect of male joblessness and family disruption American Journal of Sociology, 93(2), 348-382.

Tellegen, A., Lykken, D.T., Bouchard, T.J., Jr., Wilcox, K., Segal, N. & Rich, S. (1988).

Personality similarity in twins reared apart and together. Journal of Personality and Social Psychology, 54, 1031-1039.

Tellegen, A., & Wailer, N. (1994). Exploring personality through test construction:

Development of the Multidimensional Personality Questionnaire. In S.R Briggs, & J.M. Cheek (Eds): Persona&v Measures: Development and Evaluation (Vol 1, pp.133-161). Greenwich, CN: JAI Press. Uyttebrouck, 0. (1993, April 14). Police study links blacks to 80% of violent crimes. Arkansas Democrat Gazette, p. 1. Westman, J. (1994). Licensing Parents: Can We Prevent Parental Abuse And Neglect? New York: Plenum Wilson, J.Q., & Herrnstein, R.J. (1985). Crime and human nature, New York: Simon, & Schuster. 13