Journal of Media Psychology, 1996, V. 2 (2), pp 4-12.
The purpose of the present study was to investigate how a class of perceptual variables the writer calls "victim reminders" affects public perception and judgment of guilt in criminal cases involving high profile celebrity defendants. Victim reminders may be defined as additional information, verbal, pictorial or textual, about victims in criminal cases which are designed to remind the public that real persons, not vague abstractions, were victimized. While the notion of victim reminders is relevant to all civil or criminal trials, it is particularly pertinent when considering criminal cases involving a celebrity defendant. In such instances victim reminders may be used to offset what may be called the "celebrity-defendant advantage."
What is the celebrity-defendant advantage in criminal cases? Celebrities are familiar rather than strangers to the public, even if such familiarity is only parasocial (Stever, 1994, Adams-Price & Greene, 1990) rather than first-hand and reciprocal. A parasocial relationship is one where a person develops a strong attachment to someone, most oftan a celebrity, he or she has never met or has only met very briefly. A presumption of familiarity takes hold. However what is usually known about the celebirty is merely a public image, not the more complex, fleshed-out knowledge generally found in one-on-one, reciprocal relationships. If a celebrity's public image is successfully "manufactured," this familiarity is often positive, providing a halo effect for future judgments and inferences about the celebrity regarding other favorable character traits (Mitroff and Bennis, 1989). Such a halo effect is particularly advantageous in high-profile criminal cases, e.g., those involving O. J. Simpson or Michael Jackson. With both Simpson and Jackson, the public has tended to view charges against them (homicide and child molestation, respectively) contextually, weighing the consistency of such charges against the celebrity's presumed familiarity and positive public persona, e.g., "O. J. is such a nice guy, he couldn't commit such a gruesome crime," or "Michael Jackson loves children, he wouldn't molest them." (Los Angeles Times, 1994, August).
With a non-celebrity defendant, on the other hand, the perceptual context
may be dramatically different and far from advantageous. There is
no positive public persona, no celebrity-nurtured incredulity factor to
bias judgments usually made on limited and frequently ambiguous or conflictual
information offered through the news media. In such low profile cases,
often little is publically known about a defendant. According to
Kassin & Wrightsman (1988) such a defendant's image rests upon the
particulars of the crime and whatever social stereotypes are activated
in areas such as physical attractiveness, race, social status, etc., e.g.,
He's an alcoholic, he probably did batter his wife.
By contrast, whatever negative demographic or physiognomic stereotype the celebrity-defendant theoretically engages may be offset by the public's presumed familiarity with the celebrity. As Allport (1954) noted, positive familiarity generally creates exceptions and offsets group or individual steretoypes. Thus, non-stereotypicality and presumed and positive familiarity can contribute to the celebrity-defendant advantage.
Trait and People Contexts
The meaning of a trait or behavior is partially determined by the context of other traits within which it is encountered. Celebrity may be seen as part of an individual's context of personality traits. Extending this reasoning, it seems likely that other people with whom one is phenomenally connected, via victim reminders, should also form what can be called a "people context." Adages such as "we are judged by the company we keep," or "birds of a feather flock together," are illustrations of "people contexts." Accordingly, both "people" and "trait" contexts are probable determinants of perceptions and judgments.
Depending on the people context (e.g., thinner or fatter people, more beautiful or homelier people), people can be made to look better or worse (Patkinis, Farquhar, Silbert & Hearst, 1989; Kenrick & Guiterres, 1980). In non-criminal circumstances, for example, politicians hope to enhance reputations and appeal to voters by connecting themselves to movie stars or athletes (Brownstein, 1990). Judgments of guilt or non-guilt in the eyes of the public or in a criminal proceeding may also be affected by contextual comparisons, and celebrity defendant attorneys may attempt to control the context (people, trait or both) within which their clients are being perceived and judged. One of those people contexts would be the victim or, in a murder trial, reminders of the victim. For example, if television-created celebrity defendants are accused of murder and the murder victim is portrayed as evil (e.g., Jose Menendez, the father of convicted murderers, Lyle and Erik Menendez, who was characterized during the trial as a sexually abusive, authoritarian father), the defendants' act of murder may be contextually softened, even made to appear justified. In other words, what might appear to be a hideous crime committed by brutal murderers may now be viewed as a crime of passion committed by otherwise admirable and likeable defendants.
Pretrial publicity plays an important role in portraying to the public (and prospective jurors) the images of the celebrity defendant and his or her victims. Research evidence clearly indicates exposure to trial-related news stories shapes public opinion (Kassin and Wrightsman, 1988). Such shaping leads the public or actual or prospective jurors to form opinions prematurely, thereby possibly prejudicing their reactions to evidence presented in trial.
Ordinarily, in highly-publicized crimes, public opinion tends to align itself with the prosecution (e.g., Carroll, Kerr, Alfini, Weaver, MacConn, & Feldman, 1986) But, given the media circus of pretrial publicity surrounding a high-profile or celebrity case (e.g., O.J. Simpson, Alec Baldwin), an additional extra-legal contaminating factor may be cause for concern: Celebrity status of the the alleged criminal may overshadow many otherwise compelling aspects of the case, even the existence of murder victims. This may be especially so in those cases where only one of the parties, the defendant in particular, is a celebrity.
From a "people context" perspective, however, reputation is only part of the picture. In the Simpson case, for example, before and during the trial the defendant was in plain sight of the public, in a position to speak in his own behalf, his public persona possibly enhancing the credibility of his self-defense. For the victims, it was a different story. Neither Nicole Brown Simpson nor Ronald Goldman, the murder victims in the Simpson case, could speak in their own behalf. This is obviously true in all murder cases. But all murder cases do not have a celebrity defendant whose mass appeal and mass support are anchoring points against which evidence is evaluated. In the mind of the public, then, how does the physical absence, non-celebrity status, and absence of pre-existing public affections for Nicole Brown Simpson and Ronald Goldman rival the physical presence and halo effects bathing the reputation of their alleged murderer, O.J. Simpson? This is one of the questions the present study sought to explore.
A recent meta-analysis of the welter of literature on the "Beautiful is good" stereotype (Eagly, Ashmore, Makhijani, & Longo,1991) concluded that the positive stereotype or halo effect of attractiveness derives its power from context. We may view the positive celebrity stereotype of Simpson as an equivalent to the "beautiful is good" stereotype and part of the celebrity-defendant advantage. But, if Simpson's murderer-incongruent hero or good-guy image is contextually assaulted with compelling reminders of his alleged victims, this might alter the "people context" within which Simpson is evaluated. All other things being equal, then, the celebrity-defendant advantage in the court of public opinion would be diminished and the tendency to perceive Simpson as innocent of the crimes, simply because of a halo effect, might be reduced.
Race and Judgment
But, the Simpson murder case is somewhat more complicated than merely highlighting the celebrity defendant advantage. The fact that Simpson is an African-American and that his alleged victims were white raised the question of whether the race of defendants and of victims would enter into the judgmental equation, particularly in Los Angeles, the site of so much recent racial strife and controversy. Experimental studies (e.g., Mazzella & Feingold, 1994, Pfeifer & Ogloff, 1991) have demonstrated that jurors' perceptions and interpretations of trial testimony are significantly influenced by the race of the defendant. Other research shows that jurors consider the victim's race as well as the defendant's when assigning a verdict (Hyams, Leinart, Rowe, & Rogers, 1993, Hans & Vidmar, 1986). Additional research shows that jurors are quite sensitive to the racial similarity or dissimilarity between themselves, the defendant, and the victim (Towson & Zanna, 1983, Ugwuegbu, 1979). The same might be true for the public's perceptions. Indeed, polls before the Simpson trial showed that African Americans were less inclined than other racial groups to believe Simpson was guilty (L. A. Times, 1994, August).
In order to test the hypothesis that the public's judgments about the possible guilt of O. J. Simpson are affected by the presence of victim reminders and that racial factors enter into the judgmental equation, the study reported below was undertaken. Victim reminders took either of two forms: (1) a brief written description of victims, or (2) pictures of the victims. It was predicted that victim reminders would increase the tendency to judge O.J. Simpson as guilty. It was further predicted that pictorial representation of victims would be more effective as victim reminders than a simple written description and thereby produce greater tendencies toward a judgment of guilt.
Based on previously cited research, it was expected that Blacks would be less inclined to perceive Simpson as guilty than Whites. To determine whether it is minority status or racial identification which affects judgments of guilt, Hispanics were included in the subject population as a comparison point for both Blacks and Whites.
Other aspects of the pretrial publicity and the appropriate sentencing of O. J. Simpson, were he convicted of the charges, were also explored. Previous research has shown that women, Blacks, the poor and certain religious groups are more likely to oppose the death penalty than their demographic alternatives (e.g., Cowan, Thompson & Ellsworth, 1984). Such groups are also less cynical about the insanity defense (Thompson, Cowan, Ellsworth & Harrington, 1984). Therefore it was predicted that Blacks in the present study would be the least supportive of the death penalty in general and as regards Simpson in particular, and more supportive of an insanity defense.
PROCEDURE AND METHODOLOGY
On two separate Wednesdays in July, 1994, a 5-item opinion survey was administered to a sample of student and non-student respondents at the California State University, Los Angeles campus. The data was collected by a racially mixed group of 14 students (nine females, five males) in the author's experimental psychology class. The students were briefly trained in how to do interviews in the field. The respondents were not randomly selected; they were given the survey in the order they were encountered at various locales on the campus. The use of diverse locations to gather respondents was done to avoid biasing the sample with like-minded congregants at any particular campus location. There were equal numbers of male and female Whites, Blacks and Hispanics in each sample and in each of the three experimental conditions described below.
The data collected on the second Wednesday was done as a check on the representativeness of the sample on the first Wednesday. Analyses revealed that there were neither significant nor substantive differences between the two samples, either in terms of demographic characteristics or dependent variable measure outcomes. Consequently, data from the two samples were combined for subsequent analyses. This yielded a total of 254 participants.
The mean age of the sample was 26.3 years. For males and females respectively it was was 26.9 and 25.5 and for respondents in each of the three vitcim reminder conditions (Basic, Written and Pictorial, respectively) it was 26.9, 25.7 and 25.9. The mean ages respectively for Blacks, Hispanics and Whites were 27, 24.8 and 26.9. None of these age differences was statistically significant..
Respondents were assigned to one of three victim reminders conditions. One condition (Basic) simply asked the respondents a series of questions about the Simpson case with no reference to the victims other than their names. A second condition (Written Description) added to the basic survey some information about Nicole Brown Simpson and Ronald Goldman regarding their ages and the number of children born to the Simpsons during their marriage. A third condition (Pictorial Description) added to the basic survey side-by-side, 1.5" x 3" black and white news photos of O. J. Simpson, Nicole Brown Simpson and Ronald Goldman. These were all bust shots and, like the Written Description, were positioned at the top of the survey page preceding the survey questions. Except for question 4, dealing with selecting one of three possible sentences for Simpson if he were found guilty of the charge of murder, all questions could be answered on a six point Likert-type scale ranging from Strongly Agree to Strongly Disagree.
The respondents were approached, their cooperation was solicited and they were given one of the three surveys to fill out. The interviewers stood away from the respondent while he or she answered the questionnaire in private. This was done to minimize any response biasing effects the surveillance of an interviewer might have. Upon completion, the questionnaire was placed in an envelope and the respondents were thanked for their cooperation. Out of respondent view, the interviewer noted on the survey the respondent's race. This procedure was followed in order to avoid sensitizing the respondents to possible race-related survey concerns which may have influenced their responses or cooperation.
Unless otherwise indicated, all items were subjected to 3 x 3 ANOVAS, with 3 levels of Race (White, Hispanic and Black) and 3 levels (Conditions) of victim reminders (Basic or no reminders, Written reminders, and Pictorial reminders). Table 1 contains the mean scores on the five survey items for all Racial groups and Conditions of victim reminders.
Question 1: The Issue of Guilt or Non-Guilt
Responses to the statement "O. J. Simpson committed the murders of Nicole Brown Simpson and Ronald Goldman," showed significant differences as a function of Condition, F (2, 245) = 4.97, p < .01, and as a function of Race, F (2, 245) =13.0 , p < .001. The interaction between Race and Condition was non-significant, F (4, 245) = .77, ns.
Victim Reminder Conditions
In scoring this item, the higher the score the greater the guilt judgment of Simpson. As Table 1a indicates, for all groups, regardless of Race, those who had victim reminders were more likely to judge Simpson guilty of the alleged charges of murder than those without such reminders. This difference increased from Written to Pictorial reminders of victims. Results of a Bonferroni post-hoc paired comparison procedure, however, revealed that, while the increases between Basic (no reminders) and Written reminders and between Basic and Pictorial reminders were significant (p < .05 and p < 01 respectively), the difference between the Written and Pictorial reminders, though in the predicted direction of increased guilt, was non-significant. The evidence thus failed to support the prediction that Pictorial reminders were more powerful than Written but did support the hypothesis that victim reminders significantly impacted on judgments of Simpson's guilt.
Though there was a significant main effect for Race, further analysis revealed that, while Blacks showed an increase in guilt judgments from Basic to both Victim Reminder conditions, they were consistently lower than both Hispanics and Whites on the magnitude of such judgments. Only Whites, moreover, showed any appreciable increase in guilt judgments between Written (M = 4.2) and Pictorial reminders (M = 4.6). For Hispanics, on the other hand, while victim reminders increased judgments of guilt, the greatest increase occurred for respondents who were given Written reminders of victims rather than for those who were given the Pictorial reminders. This is essentially a reverse of what obtained for Whites and Blacks. The reason for this is unclear and results warrant further study.
The other main difference between the races concerned the absolute degree of judged guilt or non-guilt of Simpson. Given the weighting of the two poles on this item, Agree = 6, Disagree = 1, anything above a mean score of 3 would lean in the direction of guilt while anything below a mean score of 3 would lean in the direction of non-guilt. The average score for Blacks in the Basic condition was 2.6 whereas their mean scores in the Written and Pictorial reminder conditions were 3.3 and 3.4 respectively. This indicates that, in the Basic condition, Blacks leaned toward a judgment of non-guilt while Hispanics and Whites leaned toward a judgment of guilt in all three conditions. Summing respondent scores across all Conditions, Blacks adjudged Simpson to be less guilty than either Hispanics or Whites (Blacks, M = 3.1, Hispanics, M = 3.7, Whites, M = 4.2).
Question 2: Insanity Defense
As Table 1b indicates, responses to the statement "If O. J. Simpson did commit the murders, he was probably temporarily insane at the time of the murders," revealed no statistically significant differences either in terms of Race, F (2, 245) = 1.67, ns, or Condition, F (2, 245) =.35, ns. But, the pattern of responses was similar to that seen regarding the judgment of Guilt. While all races tended to disbelieve the insanity defense (the higher the score the greater the disagreement), Blacks and Whites were further apart on the issue with Whites (M = 3.9) dismissing the insanity defense more than Blacks ( M = 3.4). As with the issue of Guilt, Hispanics (M = 3.7) fell in the middle of the other two races on this issue. But, clearly, the prediction that Blacks would be supportive of the insanity defense in the case of O.J. Simpson, was not supported.
Question 3: Death Penalty In General
The higher the score on this item, the greater the support for the death penalty in general. As Table 1c reveals, the issue of the death penalty did find modest support from all respondent groups when they were asked to indicate whether or not they support or oppose the death penalty. While there was no main effect for Condition, F (2, 245) = .32, ns, there was a significant effect for Race, F (2, 245) = 5.25, p < .01. Blacks were significantly less supportive of the death penalty in general than either Hispanics or Whites (Blacks, M = 3.4, Hispanics, M = 4.1, Whites, M = 4.2), thereby supporting one of the studyís hypotheses.
Question 4: Death Penalty for Simpson
Respondents were asked what sentence Simpson should receive if he were convicted of the crimes for which he was charged: a lengthy prison sentence, a life sentence or the death penalty. As the means in Table 1d show and a Chi-square analysis revealed, there was no effect for Condition, X2 = .82, df =4, but a marginally significant overall effect for Race, X2 = 8.56, df =4, p < .07. Further analyses revealed that Blacks differed significantly from both Whites, X2 = 4.69, df =2, p < .05 and Hispanics, X2 = 7.54, df =2, p < .05, but that Hispanics and Whites were non significantly different from each other, X2 = .78, ns. While all groups preferred a life sentence to either a death or lengthy prison sentence, Blacks were more opposed to the death sentence than either Hispanics or Whites, again supporting one of the studyís hypotheses.
Question 5: Pretrial Publicity
Finally, on the issue of pretrial publicity, respondents were asked whether and to what degree they agreed or disagreed with the following statement: "Whether it favors the prosecution or favors the defense (O. J. Simpson), the effect of pretrial publicity will prevent justice from being truly served in the trial of O. J. Simpson." A higher score indicates a greater belief in the pretrial publicity impact on the trial outcome. There was no significant effect for Race on the issue of publicity effects, F (2, 245) = 1.5, ns, but there was a significant effect for Condition (victim reminders), F (2, 245) = 5.21, p < .001. As cell means in Table 1e reveal, although there was a general tendency for respondents to see the pretrial publicity as working against a fair trial, those respondents with victim reminders were far more likely to think the publicity would exert such an effect than those in the Basic condition. Hispanics believed in the pretrial publicity effect to a lesser degree than either Blacks or Whites (Blacks, M = 4.4, Hispanics, M = 4.1, Whites, M = 4.6).
Since the respondents were not questioned as to the direction of pretrial publicity bias, only whether there might be a bias, the precise meaning of the results on this question is obscure. The fact that there was an effect for Condition but not for Race, however, may mean that the perceived pretrial bias is in the direction of favoring O. J. Simpson and the position of the defense. Nothing else would plausibly explain why there is an increase in the belief in the biasing effects of pretrial publicity as function of victim reminders.
Effects of Gender
As regards the effects of gender, the only significant differences to emerge were that females were less supportive of the death penalty (M = 3.56) than males (M = 4.2), t (252) = 2.82, p < .01 and that females were more extreme than males, X2 = 7.25, df = 2, p < .03, in favoring sentencing O. J. Simpson to life in prison (M = 48% and M = 37% for females and males respectively) as compared to sentencing him to the death penalty. This is probably related to their lower preference for the death sentence in general.
The results of the present study clearly indicate that the presence of victim reminders affects the publicís opinion of O. J. Simpson's guilt in the murders of Nicole Brown Simpson and Ronald Goldman. Results also indicate that, while all races are affected by victim reminders, Blacks and Whites are more polarized on the issue than Hispanics. This is true not only on the issue of Simpson's guilt but also on the viability of the insanity defense, support of the death penalty and the biasing effects of pretrial publicity. These results support the study's central hypotheses and suggest that racial identification with the defendant rather than minority status per se is a determining factor in public opinion regarding guilt in this celebrity case.
As regards the insanity defense, it is interesting that, similar to the findings of Thompson, et al (1984), in the present sample people who disagreed with the insanity defense for Simpson tended to recommend a more severe sentence if Simpson is convicted (r = .29, p < .05). However, when results are broken down according to race, the correlation is a substantial .64 for Blacks but only .25 for Whites while for Hispanics the correlation is near zero (r = .02). This suggests that the demographic relationship that Thompson et al (op cit) found was, in the present study, substantial only for a minority group (Blacks) that typically doesnít support the death penalty. Presumably, when Blacks believed that Simpson was responsible for the murders in the absence of mitigating emotional circumstances, they are more inclined to feel he deserves a severe sentence than are other racial groups.
The results of the present study support the hypothesis that the public's judgment of the guilt of celebrity defendants in cases such as O. J. Simpsonís, is indeed affected by victim reminders. However, given the significant effect of race, another element must be entered into the equation, namely the identification of the judge with the person being judged. In other words, it would seem that Blacks' identification with Simpson was part of the phenomenological field in which judgments about Simpson were being rendered. Whether there was a contrast effect for Whites is unclear but certainly possible. If one uses the Hispanic sample population as a base line of judgment then, in all but one instance, Whites and Blacks were polarized as to judgments about Simpson and related aspects of the case such as insanity defense and punishment. Of course, Black mistrust of the judicial and police systems in Los Angeles may provide an alternative explanation for the observed greater Black reticence to view Simpson as guilty as compared with other racial groups. The present research design did not explore that alternative dimension.
Finally, the impact of victim reminders on how the public judges the guilt of a celebrity defendant such as O. J. Simpson and how actual jurors may judge Simpson can only be the subject of conjecture. The array of information available to a jury and the context of their judgment are dramatically different than for that of the public. Moreover, whether the presence of victim reminders has similar effects on public opinion or juridical decisions in other celebrity or non-celebrity criminal cases warrants further investigation. The fact that a year after this study was conducted, O.J. Simpson was acquitted of the murder charges in a trial fraught with allegations of racism, evidence tampering and by a jury dominated by Blacks suggests that, as with many research studies which look at the potentially significant impact of social or perceptual variables on judgments, juridical or otherwise, reality is far sloppier and difficult to control than researchers might wish. Consequently, extrapolating from research results to real world outcomes is rather hazardous. Whether the race of the jurors was a determing effect on the Simpson trial verdict, as was suggested by the results of the present study, or whether other aspects of the trial were of overriding significance, is a question which will probably forever elude a definitive answer.
Allport, G. (1954). The nature of prejudice. Reading, MA: Addison-Wesley.
Asch, S. E. (1946). Forming impressions of personality. Journal of Abnormal and Social Psychology, 41, 258-290.
Asch, S. E., and Zukier, H. (1984). Thinking about persons. Journal of Personality and Social Psychology, 46, 1230-1240.
Brownstein, R. (1990). The power and the glitter. New York: Pantheon.
Carroll, J. S., Kerr, N. L., Alfini, J. J., Weaver, F. M., MacCoun,
R. J & Feldman, V. (1986). Free press and fair trial: The role
research. Law and Human Behavior, 10, 187-201.
Cowan, C. L., Thompson, W. C., & Ellsworth, P. C. (1984).
The effect of death qualifications on jurors' predisposition to convict
and on the
quality of deliberation. Law and Human Behavior,. 8, 53-79.
Eagly, A. H., Ashmore, R. D., Makhijani, M. G., & Longo, L. C. (1991).
What is beautiful is good, but ... : A meta-analytic review of research
on the physical attractiveness stereotype. Psychological Bulletin, 110, 109-128.
Hans, V., P. & Vidmar, N. (1986). Judging the jury. New York: Plenum Press.
Hyams, R. W., Leinart, M., Rowe, S., & Rogers, W. (1993).
Acquaintance Rape: The effect of race of defendant and race of victim on
juror decision. The Journal of Social Psychology, 133, 627-634.
Kassin, Saul, M. & Wrightsman, Lawrence, S. (1988). The
American jury on trial: Psychological perspectives. New York:
Kenrick, D. T., & Gutierres, S. E. (1980). Contrast effects
in judgments of attractiveness: When beauty becomes a social problem.
Personality and Social Psychology, 38, 131-140.
Los Angeles Times (1994), August 23, Section A, p. 23.
Mazzella, R. & Feingold, A. (1994). The effects of physical
attractiveness, race, socioeconomic status, and gender o f defendants and
on judgments of mock jurors: A meta-analysis. Journal of Applied Social Psychology, 24, 1315-1344.
Mitroff, D. & Bennis, W. (1989). The Unreality Industry: The
deliberate manufacturing of falsehood and what it is doing to our lives.
N.Y.: Birch Lane Press.
Pfeifer, J. E., & Ogloff, J. R. (1991). Ambiguity and guilt
determination: A modern racism perspective. Journal of Applied
Psychology, 21, 1713-1735.
Pratkanis, A. R., Farquhar, P. H., Silbert, S., & Hearst. J. (1989).
Decoys produce contrast effects and alter choice probabilities. University
California, Santa Cruz. Cited in Aronson, E. (1995). The Social Animal (7th ed.) New York: W. H. Freeman.
Stever, Gayle, S. (1994, August). Para-social attachments:
motivational antecedents. Paper presented at the meeting of the
Psychological Association, Los Angeles.
Thompson, W. C., Cowan, C. L., Ellsworth, P. C., & Harrington, J.
C. (1984). Death penalty attitudes and conviction proneness: The
translation of attitudes into verdicts. Law and Human Behavior, 8, 95-113.
Towson, S. M. J., & Zanna, M. P. (1983). Retaliation against sexual
assault: Self-defense or public duty? Psychology of Women Quarterly,
Ugwuegbu, D. C. E. (1979). Racial and evidential factors in juror attribution
of legal responsibility. Journal of Experimental Social
Psychology, 15, 133-146.
Back to Index Page