Don't miss our trial consultant responses at the end of this article: Julie Blackman, Ph.D., Beth Foley, Ph.D. and Kathy Kellermann, Ph.D.

Knowing how best to escape blame is essential. After an affair, it can mean the difference between marriage and divorce; after a mistake at work, it can mean the difference between employment and joblessness. Nowhere is escaping blame more important than in the courtroom, where reputations, fines and jail time all hang in the balance. For those accused of capital crimes, knowing how to avoid blame can literally mean the difference between life and death.

As a defense attorney, you are often tasked with helping clients escape blame. Obviously, the best strategy for escaping blame is to deny that your client had any sort of involvement in the wrongdoing, however, when faced with irrefutable evidence, that strategy is not an option. When it is clear that the defendant is guilty, defense teams often appeal to either the hero strategy or the victim strategy in hopes of reducing blame. They paint their clients either as upright citizens, with glowing records of previous good deeds, or as victims, with long lists of harms suffered at the hands of others. In a series of studies, we have explored which tactic – if either – is effective in reducing blame. When punishment hangs in the balance, is it better to be a hero or a victim?

Heroes or Victims

Two legal cases gained national attention in the summer of 2011 – the New Orleans Danziger Bridge Case and (more notably) the Casey Anthony trial. In the first case, the legal team framed the defendants as heroes, in the other, as a victim. Although both strategies are intuitively appealing, in the end only one strategy was effective in mitigating blame.

The Danziger Bridge Case took place in post-Katrina New Orleans, where five NOLA police officers were accused of killing two men, seriously injuring four others, and then orchestrating an elaborate cover-up scheme. Their defense attorneys employed the hero strategy, painting the police officers as “the good guys” in hopes of reducing blame. Despite the serious charges against them, attorneys argued that the police officers were not heartless killers, but rather the ones restoring order in a time of chaos and anarchy.

The intuitive logic behind the strategy is clear – if the police officers are the city’s heroes, how can they also be criminals? This strategy suggests that jurors employ a kind of moral calculus of good versus evil when judging defendants; the more good someone does, the harder it is to see them as blameworthy. This is certainly intuitively appealing, as there seems to a big psychological gap between heroes and villains, and so the more someone seems like a hero, the less they should appear to be a villain.

Indeed, there is some psychological research suggesting that good deeds might serve as a type of karmic insurance policy, safeguarding heroes from future harsh judgments. People generally believe that we live in a just world where good people ought to be rewarded and bad people punished (Lerner & Miller, 1978). Holding to this just world belief, it seems reasonable to expect that, in the courtroom, previously upright citizens should be rewarded with less blame for a crime.

It also seems reasonable that a person’s past good deeds would encourage a charitable interpretation of one’s transgressions. Consistent with this idea, people ascribe blame for specific acts on irrelevant [irrelevant in what sense?] bad character, blaming generally despicable people more harshly than others (Alicke, 1992). Juries are also more likely to grant clemency to good people, but not bad people who espouse ignorance of the law (Alter, Kernochan & Darley, 2007). If you know that the accused has a history of saving children, then maybe he committed a crime accidentally, or to help others. Despite the evidence in favor of the hero strategy, however, it ultimately failed the NOLA police officers, who were found guilty on all 25 counts.

Unlike the NOLA police, when Casey Anthony stood accused of murdering her three-year-old daughter and burying her in a neighbor’s backyard, her lawyers framed her as a victim. After hearing the alleged sexual abuse that Anthony suffered at the hands of her father and brother, it seemed more difficult for the jury to view her as a monstrous mother. Although the evidence against Anthony was compelling, she was acquitted of all charges.

Anthony’s trial is only the most recent example of famous cases that ended in an acquittal when the attorney used the victim strategy. In 1996, O.J. Simpson escaped a murder conviction partially because his lawyers painted him as a target of the LAPD’s racist leanings. A half decade earlier, Lyle and Erik Menendez were accused of murdering their wealthy parents. Although the brothers were found guilty of first degree-murder, being framed as casualties of their abusive father and overbearing mother helped them escape the death penalty. Why might the victim strategy work, when the arguably more intuitive hero strategy fails? Why do we seem to punish police officers, our national heroes, more harshly than victims? The answer stems from the underlying structure of morality.

Moral Typecasting

Moral and immoral behaviors typically require two parties, a dyad of moral agent and moral patient (Gray & Wegner, 2009). The moral agent is the doer of the moral action, while the moral patient is the recipient – for example, in a robbery, the thief is the moral agent and the victim is the moral patient. Without this dyadic structure, the moral nature of an action tends to disappear: without the victim, stealing is simply finding something; without the thief, a person has merely lost something. In the parlance of the law, moral agents represent mens rea – the presence of blameworthy intent – while moral patients represent actus reus– the presence of someone harmed by an action.

Although people are generally able to serve as both moral agents and patients – as both perpetrators and victims – in any specific moral act, people take on the role of either one or the other: the rapist is not the one raped; the thief is not the one who is stolen from. Research suggests that this tendency to be either the agent or the patient in a specific act is extended more broadly to perceptions of character, such that people see others as either agents or patients across moral acts in general. This phenomenon is called moral typecasting: just as we typecast actors into enduring Hollywood roles (think of Leonard Nimoy as Spock), so too do we typecast others into enduring moral roles, as either perpetual perpetrators or everlasting victims.

An analogy to visual perception can help capture the phenomenon of moral typecasting: when viewing the duck-rabbit illusion (Figure 1), at times you might see a duck whereas at other times you might see a rabbit. Despite knowing that the figure depicts both animals, we struggle to see the image as simultaneously being both animals. Likewise, s a result of moral typecasting, we perceive others as either the moral agent or the moral patient, despite knowing that people can technically be both victims and perpetrators.

Figure 1. Does this picture depict a duck or a rabbit?  Jastrow’s (1899) Duck Rabbit illusion.

At first blush, this distinction between moral agent and moral patient seems to be less important than the dichotomy we see between good and evil. From a young age, we split our social worlds into angelic heroes and nefarious villains and learn that heroes are praised, and villains punished. In moral typecasting, however, both the good and the bad agents are lumped in a single category – the moral agent. Since both heroes and villains do moral actions, we perceive them similarly – both are seen as responsible for their actions. In contrast, we perceive moral patients solely as the recipients of the moral action; we focus on the pain they suffer from villains or the help they need from heroes, and this perceived helplessness leads them to be seen as incapable of responsible action. More succinctly, moral typecasting suggests that a) victims should escape blame and b) heroes should earn blame at a level similar to villains.

The claims of moral typecasting seem to contradict both intuition and psychological research. In addition to the research discussed earlier about the supposed power of past good deeds, there is also ample research on “blaming the victim” (Lerner & Simmons, 1966). One of the best-known examples of blaming the victim involves survivors of rape. Upon hearing stories of rape, people often try to justify the assault by condemning the victim’s provocative clothing or flirtatious behavior. Blaming the victim stems from our belief in a just world. The flipside of believing in a just world where good people are rewarded is the belief that only those deserving harm will suffer (Lerner & Miller, 1978). Since we live in a just world, those who suffer somehow deserve their suffering. However, in “blaming the victim,” the victim is blamed solely for the act that casts him as a victim, and not for all subsequent behaviors. Typecasting, on the other hand, is more enduring and so extends attributions of blame beyond the harm that causes one to be cast as a victim. So, it is not how much blame a victim deserves for a rape, for instance, but instead how much blame a rape victim deserves for a subsequent murder.

Through a series of studies, we tested whether people’s judgments of blame follow the predictions of moral typecasting, rather than an intuitive distinction between good and bad. In particular, we tested whether highlighting victimization (the victim strategy) was better than highlighting virtue (the hero strategy).

Victimhood Versus Virtue

In the first experiment, we gave people a scenario concerning a man named George, who was initially seen as a hero, a victim, or neither before committing a misdeed. In particular, participants read that every week, George either donates $100 to charity (hero), gets $100 stolen from him by his cruel boss (victim), or spends $100 on groceries and other expenses (normal). George then takes $10 dropped from a woman’s purse, and we asked how much blame George deserves for this misdeed. Consistent with the moral typecasting hypothesis, we found that victim George received less blame and less punishment than both normal and hero George. We also found that hero George was given no less blame than normal George, suggesting that his previous good deeds did nothing for assignments of blame. Importantly, this occurred without specifying where George would be spending the money – it could be that hero George was a modern day Robin Hood and would be given the $10 to charity – but no one cut the hero a break. So while victims escaped blame, heroes did not.

Figure 2. The relative amount of blame assigned to heroes, victims, and neutral characters in the “George” case (Gray & Wegner, 2011).

In a second study, we examined the power of moral typecasting with a real-world workplace negligence scenario comparing heroes and victims. Participants read about two restaurant cooks – a “hero” cook who started a charity in college, and a “victim” cook who had been hit by a drunk driver while in college (though now fully recovered). The men ignore a woman’s request for a peanut-free salad, even though she is severely allergic to peanuts. The woman eats the salad, has a near deadly allergic reaction, and then threatens to sue the restaurant unless they fire one of the cooks. The question is who would participants choose to fire? Asking people to fire someone provides a strong test of moral typecasting, because while you might give less blame to a victim, presumably you would want to keep the hero on the staff so he or she could keep doing good. Consistent with typecasting, however, the hero cook was overwhelmingly assigned more blame than the victim, and – despite his good deeds – given the axe significantly more often.

These studies suggest that the victim strategy is optimal for escaping blame and punishment after a transgression, but sometimes the victim strategy is not an option. For example, a defendant might come from a privileged background and never have suffered real harm. In these cases, should one then turn to the hero strategy? Does being perceived as a hero earn one less blame than simply being perceived as a neutral character? If so, than the hero strategy might still have some utility in the courtroom.

In the next set of studies we compared the blame assigned to heroes to the blame assigned to neutral targets. Following-up on the workplace negligence study, we added a neutral cook to contrast with the hero and victim cooks, who – in contrast to being characterized by good deeds or harm – simply majored in communications in college and used to work at a local hardware store. When both the neutral cook and the victim cook caused a life-threatening allergic reaction, people chose to fire the neutral cook – victims win again. On the other hand, when the neutral cook was paired with the hero cook, the majority of people fired the hero cook (Figure 3). This is consistent with the idea of moral typecasting – a moral agent, whether good or evil, remains more capable of earning blame than your average person. In the courtroom, this suggests that it might be better not even to mention a history of good deeds once guilt has been established.

Figure 3. The relative amount of blame assigned to heroes, victims, and neutral characters in the workplace negligence case (Gray & Wegner, 2011).

In another study, we found that the punishment of heroes holds even when people judge those with a lifetime of good deeds. In the previously discussed dropped money case, we replaced the anonymous “George” with prototypic heroes, victims and neutral characters (Gray & Wegner, 2009), and a similar pattern emerged. People blame the Dalai Lama more than a high school teacher for keeping dropped money, and a high school teacher more than an orphan. If people assign blame solely by comparing tallies of good versus evil, the Dali Lama should clearly earn the least amount of blame; however, consistent with the moral typecasting hypothesis, living a lifetime of good deeds actually earned him more blame, and only the victim escaped blame relative to a neutral target.

The power of the victim strategy extends not only to explicit judgments but also to memories of immorality. We presented people with a detailed story of a businessman’s morning (get up, brush teeth, put on grey suit, eat bagel, drive to train, board train, etc…) which included one moral transgression: as with “George,” this man sees a woman in front of him drop $10 and he picks it up and doesn’t give it back. Importantly, a single sentence of background was provided at the beginning of this story, casting him as a hero (once worked for Habitat for Humanity), a victim (once hit by a drunk driver), or your average person (once worked at a hardware store).

After reading this story, participants had a five minute break, and then we asked people to recall five things from the story; we were interested in whether people recalled the moral misdeed. Consistent with typecasting, we found that significantly fewer people remembered the moral transgression when it was enacted by a victim, relative to the average and heroic person. Also consistent with typecasting, the moral misdeed was recalled earlier in the list of remembered details for the hero relative to the average person, suggesting that it was more salient for the good doer. Thus, people seem to forget the wrongdoings of victims but never forget those of previous good doers, providing one more piece of evidence in favor of victimhood over virtue as a way to avoid blame.

Boundary Conditions

Through a number of studies, we found evidence in favor of moral typecasting, which advocates the victim strategy over the hero strategy. Despite these data, there are important cases in which victim strategy may be unlikely to work. First, if it is employed without real harms to back it up, then the defendant might seem like a malingerer, and earn even more blame (i.e., not only are they guilty, they’re also dishonest). Even without perceptions of dishonesty, harms suffered by a defendant must be perceived to be extreme enough to others; just imagine a privileged defendant complaining of some minor harm in front of a jury of less well-off peers – these complaints will likely prompt more hostility than sympathy.

The victim card can also backfire if people are perceived as responsible for their misfortune, for although people typically respond to tragedies with sympathy, they generally feel resentment when someone appears to have brought suffering upon him or herself (Weiner, 1980). For example, we sympathize with AIDs patients when they contracted the disease through a blood transfusion, but resent those who contracted it through sexual behaviors (Weiner, 1995). If people perceived the defendant as responsible for his past victimization, then the blame for that victimization can increase the blame in the current case.

The victim strategy also has limits, as even the most prototypical victims can seem responsible if the crime is particularity egregious. For example, children are strongly perceived as moral patients – sensitive to harm but not responsible for causing it – but when two British children tortured and murdered toddler James Bulger in 1993, they were perceived as moral agents, and tried as adults. In general then, typecasting suggests that the more horrific the crime, the harder it will be to perceive the defendant as a victim.

We are not suggesting that the power of the victim strategy is necessarily rational. Certainly, we are not suggesting that everyone stop doing charity work and helping out in their communities to take on the mantle of victimhood. Indeed, being a victim has psychological costs associated with it (Tait & Silver, 1989; Bergeron, 2005) and good deeds have a host of benefits; engaging in moral behavior not only helps society, but boosts the mood of do-gooders (Dunn, Aknin, & Norton, 2008) and can even make them physically stronger (Gray, 2010). Even in courts of law, being a hero can be advantageous if it is unclear whether one has actually committed the crime, because highlighting one’s good deeds leads to a more charitable interpretation of the situation (Alicke, 1994). However, the benefits of heroism disappear once guilt is assured, so one must weigh the chances of complete acquittal afforded by the hero strategy against the robust power of victimhood.


Popular imagination holds that at the gates of Heaven, St. Peter has a record of each person’s life events. If good deeds outweigh the bad, then that person earns their Heavenly wings; if bad deeds outweigh the good, then that person is condemned to eternal perdition. As compelling as this metaphor may be, research suggests that in cases where blame is at stake, people seldom count good deeds against bad. If anything, previous good deeds actually make you more blameworthy for moral transgressions because they cast you as a moral agent – and therefore as ultimately responsible. Instead, it seems that the best route to escaping blame lies through victimhood, because it is so difficult to see moral patients as moral agents. This victim strategy may not always work, but when guilt is clear, it consistently trumps the hero strategy.

There is no doubt that society needs heroes, but when the life of a defendant hangs in the balance, it seems you should leave stories of valiant heroism at the door and instead focus on the harms suffered at the hands of others. For once, victory goes to the victims.

Kurt Gray is an Assistant Professor of Psychology at the University of Maryland, College Park. He completed his Ph.D. at Harvard University. He investigates the psychological basis of moral judgments and their real-world consequences, in domains ranging from animal rights to toture. He is currently the principal investigator of the Mind Perception and Morality Lab.

Chelsea Schein is completing her Ph.D. in social psychology at the University of Maryland, College Park. She studies moral decision making, moral behaviors, and mind perception, and currently works with Dr. Kurt Gray in the Mind Perception and Morality Lab.


Alicke, M. D. (1992). Culpable causation. Journal of Personality and Social Psychology, 63(3), 368-378.

Alicke, M. D. (1994). Evidential and extra-evidential evaluations of social conduct. Journal of Social Behavior & Personality, 9(4), 591-615.

Alter, A. L., Kernochan, J., & Darley, J.M. (2007). Morality influences how people apply the ignorance of law defense. Law & Society Review, 41(4), 819-864. doi:10.1111/j.1540-5893.2007.00327.x

Bergeron, G. M. (2005). Don't Call Me A Victim: Faith, Hope & Sexual Abuse in the Catholic Church. Lowell, MA: Arc Angel Publishing.

Dunn, E., Aknin, L. B., & Norton, M. I. (2008). Spending money on others promotes happiness. Science, 319(5870), 1687-1688. doi:10.1126/science.1150952

Gray, K. (2010). Moral transformation. Social Psychological and Personality Science, 1(3), 253 -258. doi:10.1177/1948550610367686

Gray, K., & Wegner, D. M. (2009). Moral typecasting: Divergent perceptions of moral agents and moral patients. Journal of Personality and Social Psychology, 96(3), 505-520. doi:10.1037/a0013748

Gray, K., & Wegner, D. M. (2011). To escape blame, don’t be a hero–Be a victim. Journal of Experimental Social Psychology, 47(2), 516-519. doi:10.1016/j.jesp.2010.12.012

Jastrow, J.(1899). The mind's eye. Popular Science Monthly, 54, 299-312.

Lerner, M. J., & Miller, D. T. (1978). Just world research and the attribution process: Looking back and ahead. Psychological Bulletin, 85(5), 1030-1051. doi:10.1037/0033-2909.85.5.1030

Lerner, M. J., & Simmons, C. H. (1966). Observer’s reaction to the “innocent victim”: Compassion or rejection? Journal of Personality and Social Psychology, 4(2), 203-210. doi:10.1037/h0023562

Tait, R., & Silver, R. L. (1989). Coming to terms with major negative life events. In J.S. Uleman & J. A. Bargh (Eds.), Unintended thought: The limits of awareness, intention and control. New York: Guilford.

Weiner, B. (1980). A cognitive (attribution)-emotion-action model of motivated behavior: An analysis of judgments of help-giving. Journal of Personality and Social Psychology, 39(2), 186-200. doi:10.1037/0022-3514.39.2.186

Weiner, B. (1995). Judgments of Responsibility: A Foundation for a Theory of Conduct of Social Conduct. New York, NY: Guilford Press.

Response by Julie Blackman, Ph.D.

Julie Blackman, Ph.D. is a Vice President at DOAR Litigation Consulting in New York City. She is a social psychologist who has worked as a trial strategy consultant in criminal and civil matters for about 30 years.

As Kurt Lewin, the father of social psychology, said, “There is nothing so practical as a good theory.” Clear, practical implications for the courtroom flow from Gray & Schein’s attention to theoretical work (and empirical results) on moral typecasting. For any of us who consult to attorneys who represent defendants in criminal matters, there is much of value to consider here.

Gray & Schein’s article adroitly combines theory, research and practice. It is good to see a well-developed line of research bear directly on key trial strategy decisions. And, the take-away message is clear: those likely to be found guilty of crimes will fare better if their experiences as victims are emphasized. A history of good deeds will not exonerate them and indeed may make their misdeeds more memorable and more blameworthy. So, criminal defense attorneys are well-advised to feature their clients’ suffering rather than their clients’ benevolent acts. Victim-perpetrators fare better, say Gray and Schein, than hero-perpetrators.

The fact that the findings are somewhat counter-intuitive enhances their appeal since they lead to a more nuanced trial position that one’s courtroom adversary is less likely to discern and refute effectively. One might have imagined that heroes would retain some of their shine when they face criminal charges. The research suggests otherwise. It is, by the way, better to be “normal” than to be a hero when facing criminal charges say Gray & Schein.

This research and the resulting advice are particularly interesting to me as they are largely consistent with my observation that character witness testimony at criminal trials is poorly received. In countless mock trials, we have seen mock jurors decry the efforts of character witnesses. Their testimony is seen as disconnected from the charges and as a blatant effort to distract the jurors from the matters at hand. Complimentary descriptions of criminal defendants are seen as smoke screens and, if anything, seem to weaken rather than strengthen the defendant’s case. I have seen defendants convicted at trial in the wake of testimony designed to make them seem like generous, caring, and helpful members of society.

Gray & Schein also note that despite their recommendation that victimhood be emphasized over virtue, sometimes there is little victimhood to point to. Particularly in securities fraud or other white-collar crime cases, defendants are often well-to-do and unlikely to be convincing victims, except for their arrest and prosecution. Not surprisingly, perhaps, defense attorneys often point to an overzealous prosecutor in an intuitive effort to reclaim victimhood for their clients. While such defendants may not have been victims before, criminal defense attorneys may argue that they are being victimized right now, before the jury’s eyes, by an unjust prosecution.

I have advised defense attorneys not to rely too heavily on the idea that the prosecutor is overzealous. Most jurors like and respect the government and may be moved to resist a defense that suggests that the government has cast its net too wide or has overstepped in its efforts to convict the defendant. It is hard to put the government on trial. It is as if, psychologically speaking, the defendant has taken on its own burden of proof – a great and legally unnecessary burden for the defense since as a legal matter the burden of proof rests entirely with the prosecution. Even so, in light of Gray & Schein’s research, accusing the government of victimizing the defendant may be the best course of action for an attorney representing a defendant with little pre-indictment victimization.

I would encourage Gray & Schein to continue their line of research in this direction. Does casting a defendant as victimized by the prosecution work to soften the jurors’ sense of the guilt of defendants who have lived good lives, free of victimization, before the instant case? How does blaming the government for victimizing the defendant work in a case where the defendant is already more easily seen as a victim (e.g., a homeless person who is charged with a crime)? Does it increase the likelihood of acquittal or lead to a less severe sentence if the defense blames the government along with other factors in a victim-defendant’s life?

Theory and research yield grounded recommendations that promote more effective trial strategies. This article by Gray & Schein represents a significant advance in insight with regard to the relationships among moral typecasting, the perception of victims (and heroes) and trial strategy.

Response by Beth Foley, Ph.D.

Beth Foley is a founder partner in the litigation consulting firm Zagnoli McEvoy Foley. She has been providing practical solutions to trial teams since 1990 and has been studying and teaching communication and persuasion since 1987.

This research confirms what I have been observing for several years and adds nuance to case strategizing. However, the challenge is applying this research to jury trials because there are many variables that affect juror decision-making. I definitely think the characteristics of the defendant are important, but not more important than the themes, the evidence, the verdict questions and the jurors’ life experiences. I’ve seen jurors have negative views of a defendant and still find in his favor. In the case of Casey Anthony, it’s unclear if the strategy of portraying her as a victim of child sexual abuse made a difference. Trying to make police officers out to be heroes was probably a risky strategy to begin with because the line between “good cop and bad cop” has been blurry for a long time. Jurors’ personal experiences with law enforcement in New Orleans likely trumped the “hero” theme. Still, there are practical take-aways from Gray and Schein’s research for both criminal and civil cases and for corporate leaders and braggadocios witnesses too. Most important is the notion that the “hero” strategy may be riskier than we realized, especially in today’s cynical climate. It’s one thing if a strategy isn’t persuasive, it’s another thing if it backfires and makes the punishment worse. Jurors are increasingly more suspicious – and for good reason. It’s a common occurrence in our society for someone to be a hero one day and villain the next. Heroes and villains are sometimes one in the same. There is no clearer example of this than disgraced Penn State football coach Joe Paterno. The cautionary message from the Gray and Shein research is that previous good deeds actually make you more blameworthy for moral transgressions because you are cast as a moral agent. This all suggests that when the stakes are high, you are safer to position in a neutral light instead of a “heroic.”

Victims, Heroes or Normal: What’s a Criminal Defendant to Do?

Response by Kathy Kellermann, Ph.D.

Kathy Kellermann, Ph.D. is President of ComCon Kathy Kellermann Communication Consulting, a trial and jury consulting firm based in Los Angeles, California. ComCon works on civil and criminal cases in both federal and state courts, and supports the free Jury Research Update blog.

At the start of a trial, jurors think of very few criminal defendants as victims and even fewer as heroes. Even more problematic, jurors think of very few criminal defendants as “normal” or “like themselves.” At the start of trial, many jurors already believe criminal defendants are bad people, probably guilty of having committed the charged offense(s). Many jurors objectify defendants as being “unlike” themselves and “not normal people.” Prosecutors use these juror predispositions and cast defendants as members of disliked out-groups many people believe are deviant, such as gang members, skin-heads, white supremacists, terrorists, gypsies, Muslims, and even attorneys, tax cheats and manipulators.

In my experience, re-humanizing and “normalizing” a defendant often must precede casting the defendant in the role of a moral agent, or any other role. Jurors are often surprised to learn that a gang member graduated from high school, held a job, and has a girlfriend, making the defendant neither a hero nor a victim, but “normal” or “typical,” rather than hated. Jurors are often surprised to learn that a white supremacist had no prior criminal record, takes care of his mother, and can feel remorse, making the defendant neither a hero nor a victim, but “normal” or “typical,” rather than hated. Jurors are often surprised to learn that an attorney has a wife and family, and lives in a middle-class neighborhood, making the defendant neither a hero nor a victim, but “normal” or “typical”, rather than hated. This process of humanization is not to make the defendant either a victim or a hero, but to make the person “normal” or “typical,” rather than hated; a person who is a member of similar groups as jurors, rather than a member only of disliked and deviant out-groups.

Does this process of humanization matter? Yes, and even if the humanization is based on seemingly trivial characteristics. For example, Zukier and Jennings (1983-1984) examined the influence of diagnostic and non-diagnostic information on judgments of guilt in a murder trial. One group of jurors considered only information that was diagnostic of guilt (i.e., direct evidence). A second group of jurors considered this same information and also received non-diagnostic information that was “typical” about a defendant’s height and vision: these jurors learned the defendant was of average height and had average vision. A third group of jurors considered the same diagnostic information, and received “atypical” information about the defendant’s height and vision: the defendant was extremely tall and had extremely good vision. “Atypical” diagnostic information had no effect on jurors’ verdicts. However, jurors given “typical” non-diagnostic information were more likely to acquit the defendant. The researchers concluded “extremeness in one category (a defendant’s height and vision) is related to “extremeness” in another category (a defendant’s likely guilt) and that “typicality” in one category (a defendant’s height and vision) is related to “typicality” in another category (a defendant’s innocence).

Humanizing a defendant is a strategy that is usually only effective when the evidence for guilt is weak or ambiguous. Smith, Stasson and Hawkes (1998) found that only when there was a small amount of diagnostic information pointing toward guilt did non-diagnostic character information of “typicality” (i.e., showing the defendant was not the sort of person who would be likely to commit the alleged crime) reduce guilty verdicts. The impact of character evidence was greatest when evidence diagnostic of guilt comprised only 20% of the total evidence presented to jurors. Jurors tend to follow the evidence when it is clear, independent of how a defendant is framed by the defense or the moral role in which a defendant is placed.

After a defendant is humanized, a decision can then be made as to whether to cast the defendant as a victim or a hero or in some other role (e.g., insane). If the role of victim is selected, the nature of the victimization matters. First, some excuses are more compelling than others to jurors. Heath and colleagues (2001) compared excuses that were self-inflicted to those that were not self-inflicted. Jurors judged a defendant who gave the excuse of Cocaine Dependency Disorder (a highly self-inflicted condition) as guiltier than a defendant who gave the excuse of PTSD (Post Traumatic Stress Disorder, a condition inflicted by others). Excuses involving self-inflicted conditions are less persuasive than excuses involving conditions inflicted by others.

Additionally, some jurors are more compelled by excuses than other jurors. Some people believe we live in a just world, while others do not. Some jurors believe bad things happen to good people, but others believe that people are in control of their destiny. Some jurors believe all accidents have an explanation though others believe accidents are accidents. These differences between jurors can affect which excuses jurors find persuasive.

For example, younger jurors are more persuaded than older jurors by excuses based on conditions inflicted by others. Higgins and colleagues (2007) investigated whether younger or older jurors were more persuaded by excuses. Jurors learned about a defendant in an assault case who either gave an excuse of Cocaine Dependency Disorder or PTSD. Younger and older jurors equally (and highly) believed the defendant with Cocaine Dependency Disorder was responsible for the assault. However, older jurors felt the defendant with PTSD was more responsible for the assault than did younger jurors – they were more certain of their verdicts, as well.

As another example, jurors who support the death penalty are less compelled by mitigating factors of child abuse and alcohol abuse than jurors who are opposed to the death penalty. In the sentencing phase of capital cases, jurors frequently hear about a defendant’s history of child abuse and alcohol abuse, both of which are offered by the defense as mitigating factors. Stevenson and colleagues (2010) investigated whether jurors used a defendant’s child abuse and alcohol abuse as mitigating factors, aggravating factors, or ignored the evidence. Over 370 death-qualified jurors watched a sentencing hearing related to a murder in the course of an armed robbery. Jurors learned from the prosecution that the defendant had a history of six prior convictions, two of which were armed robberies, and a psychiatrist testified that the defendant was dangerous and likely to commit future crimes. Jurors learned from the defense that the defendant was physically abused by his father when he was a child, had been an alcoholic since age 12, was intoxicated at the time of the crime, and the times when he has been violent are when he was drunk. Thirty-four juries deliberated on the appropriate sentence. In deliberations, jurors discounted child abuse and alcohol abuse as mitigating factors. While a higher proportion of jurors’ comments about child abuse were oriented toward mitigation (33%) than aggravation (7%), an even higher proportion of jurors’ comments (44%) argued to ignore child abuse as a mitigating factor. Similarly, jurors made more mitigating (22%) than aggravating (18%) comments about alcohol abuse, and even more comments about ignoring alcohol abuse as a mitigating factor (26%). The more strongly a juror supported the death penalty, the more he or she argued to discount child abuse and alcohol abuse as mitigating factors, and use them as aggravators.

The importance of focusing on the jurors – their experiences and attitudes – matters not only for an effective positioning of a defendant as a victim, but also as a hero.

Many jurors hold people in a hero role to a higher standard. I have worked on cases involving allegations of sexual abuse by a minister and election fraud by a judge. In both cases, we were concerned about jurors who might hold the defendant to a higher standard than the typical defendant because of the defendant’s “everyday” role. And we saw our task to be to “normalize” the defendant. In the case of the election fraud, we decided to cast the judge as a “victim” of the District Attorney’s office, because he had beaten a District Attorney in the judicial race, a rare event in the venue where this case occurred. The judge was gay, and in voir dire we focused extensively on jurors’ attitudes about the judge being gay. In so doing, we implicitly provided jurors with an explanation for why the charges were filed against the judge and we cast the judge into a victim role. We chose to cast the judge as a victim to counter the higher standard by which jurors might otherwise judge the judge.

However, cases exist in which certain types of jurors hold heroes to a lesser or different standard. Consider the authoritarian juror, that is, the juror who is very conventional, traditional, and submissive to authority; the juror who identifies with power figures and has a desire to punish violators of norms and social values. Authoritarian jurors in criminal trials tend to be prosecution-oriented and conviction prone because they respond well to the fact that prosecutors represent the legitimate authority of the state and society, an authority that endeavors to punish violators of the law.

Authoritarian jurors, however, are not always desirable for the prosecution. A number of exceptions exist to the pro-prosecution orientation of authoritarian jurors:

  1. Cases involving “crimes of obedience,” such as the Oliver North case or William Calley’s trial for the My Lai massacre in Vietnam, gain a sympathetic ear with authoritarian jurors. In these instances, authoritarian jurors are less likely to convict than non-authoritarian jurors because they can sympathize with a defendant who obeys orders.
  2. Cases involving allegations of misconduct by law enforcement personnel (e.g., excessive force or police brutality) are likely to enlist authoritarians’ sympathy for the criminal defendant in the absence of other compelling issues. Authoritarian jurors are receptive to claims by law enforcement personnel that they were responding to the level of force necessary in the situation at hand. The fact that the defendants, being police officers, and the potential for other police to testify in favor of these defendants would play into authoritarian sympathies. Of no minor importance is the face that the alleged victim is likely to be someone who is thought to have broken the law, a fact that does not escape the attention of the authoritarian juror.

Said differently, authoritarian jurors are more lenient in their verdicts for offenders who are similar to themselves, presuming the evidence is not clear for conviction.

The hero role has limits – more than the victim role, I believe. Nonetheless, the hero role can be effective under certain circumstance such as (a) the “right kind” of people are on the jury (e.g., authoritarians), (b) the “right kind” of defendants are on trial (e.g., police officers), and (c) the evidence for guilt is weak or ambiguous. A given law enforcement officer might be convicted using a hero strategy because the evidence of guilt is strong, the officers may not have been cast as “typical” officers, and/or the jurors are not the “right kind” of jurors for that case and defendant. Another law enforcement officer might be acquitted because all of these needed conditions for the hero role are met.

I am intrigued by the importance of the victim role versus the hero role, and believe that generally the victim role is an easier role in which to cast defendants than the hero role. Very few people, and so very few defendants, have led a perfect life, free from any criticism or bad acts. Character witnesses offering general positive information about a defendant’s personality and good character can be impeached through cross-examination that allows prosecutors to introduce specific bad acts in which a defendant has engaged. Hunt and Budesheim (2004) found that positive character evidence did not reduce guilt perceptions or decisions to convict, and when a character witness was cross-examined with examples of a defendant’s previous specific bad acts, jurors’ impressions of the defendant were more negative, guilt perceptions higher, and conviction decisions more likely than when no information at all was provided about the defendant’s character. The hero role is dangerous for many criminal defendants.

However, the victim role also has limits, however, and so cannot be seen as the only other choice of a role in which to cast a defendant. These limits include (a) excuses are differentially compelling to jurors, (b) who is on the jury affects how acceptable the excuses are, and (c) the evidence for guilt must be ambiguous. For me, the victim role will be effective only if the excuse(s) offered are acceptable to jurors, the “right kind” of jurors are on the jury, the evidence for guilt is not clear, and the defendant is humanized so as not to be placed by jurors into disliked out-groups.

Defendants adopting a victim role are frequently convicted, as can be seen in cases involving battered women. Russell and Melillo (2006) found that defendants most likely to receive verdicts of not guilty matched jurors’ expectations about a prototypical battered woman[i] and had passive response histories of never aggressing against the abusive husband. Defendants most likely to receive guilty verdicts deviated from jurors’ expectations about the prototypical battered woman (e.g., in her 50s, rested, strong, working, no children, blaming husband, not appearing fearful, etc.) and had an active response history of having fought back on previous occasions. Further, male jurors were more likely than female jurors to render guilty verdicts. An effective victim role for a battered woman defense demands “typicality,” the “right kind” of defendant, and the “right kind” of juror.

My first thoughts in most criminal cases are how to humanize a person who I anticipate most jurors will dislike, and how to lower the standard by which jurors judge the defendant. When a victim role serves this purpose, I advise it. When a hero role serves this purpose, I advise that. Many times, I do not have a choice of role for the defendant – it is pre-decided by the case facts, the jurors in the venue, or by other factors.


Heath, W.P., Grannemann, B. D., Peacock, M.A., & Dulyx, J. (2001). Effects of considering who and why the defendant attacked. Journal of Applied Social Psychology, 31, pp. 860-887.

Higgins, P. L., Heath, W.P., & Granneann, B. D. (2007). How type of excuse defense, mock juror age, and defendant age affect mock jurors' decisions.Journal of Social Psychology, 147, pp. 371-392.

Hunt, J. S. & Budesheim, T. L. (2004). How jurors use and misuse character evidence. Journal of Applied Psychology, 2,; pp. 347-361.

Russell, B. L. (1999). Attitudes toward battered women who kill: Prototypes and counterfactual thinking in judgments of culpability. Dissertation Abstracts International, 60, p. 8B.

Russell, B. L.. & Melillo, L. S. (2006). Attitudes toward battered women who kill: Defendant typicality and judgments of culpability. Criminal Justice and Behavior, 33, pp. 219-241.

Smith, H. D., Stasson, M. F., & Hawkes, W. G. (1998-1999). Dilution in legal decision making: Effect of non-diagnostic information in relation to amount of diagnostic evidence. Current Psychology, 17, pp. 333-345.

Stevenson, M.C., Bottoms, B.L., & Diamond, S.S.S. (2010). Jurors' discussions of a defendant's history of child abuse and alcohol abuse in capital sentencing deliberations. Psychology, Public Policy, and Law, 16, pp. 1-38.

Zukier, H. & Jennings, D. L. (1983-1984). Nondiagnosticity and typicality effects in prediction. Social Cognition, 2, pp. 187-198.

[i] Russell (1999) found that jurors expect a battered woman to exhibit certain physical, social, behavioral and psychological characteristics:physically being young, frayed, weary, fragile, and thin (135 lbs), socially having little to no interaction with others, two children and financially dependent on her husband, behaviorally hiding signs of abuse, making excuses for her husband’s behavior, and trying to please her husband, psychologically guilt-ridden, confused, depressed and fearful.