Don't miss our trial consultant responses at the end of this article: Holly G. VanLeuven, MA and Katherine James, MFA.

A Psychological Perspective on the Risks and Benefits of Positive Character Evidence

In criminal trials, defense attorneys can call witnesses to testify about a defendant’s good character. Although the Federal Rules of Evidence technically limit these character witnesses to describing general personality traits, in practice many judges allow such witnesses to support their claims with examples of specific good deeds (Uviller, 1993). Once a guilty verdict has been reached, the defense can introduce evidence of the defendant’s prior good deeds to seek a mitigated sentence. For example, during his recent sentencing for fraud, a former CEO submitted over 60 letters from friends and acquaintances enumerating even the most mundane of his positive behaviors, such as serving coffee to his housekeeper (Lee, 2011). Although people do sometimes judge transgressors more favorably in light of their good deeds (e.g., Birnbaum, 1973), positive character evidence has not consistently been shown to exert a strong influence on the kinds of decisions with which jurors are typically faced (Hunt & Budesheim, 2004; Maeder & Hunt, 2011). Moreover, introducing such evidence during a trial can be risky because, under the Federal Rules of Evidence, it allows the prosecution to counter with negative character evidence that would not otherwise have been admissible (Ross, 2004). It is thus important to understand the situations in which a defendant’s history of good deeds is most likely to positively – or negatively – impact judgments of him or her. The present article describes research that examines when and why a history of good deeds can make a defendant seem more like a hero – or more like a hypocrite.

To illustrate how one’s history of good deeds can have different effects on judgments of one’s subsequent wrongdoing, consider two verdicts rendered in the court of public opinion. First consider reactions to allegations that Martin Luther King, Jr. committed adultery (Abernathy, 1989). Because of King’s status as a moral paragon, many people might be likely to give him a “pass” and withhold their condemnation. His contributions to the civil rights movement might seem to outweigh his extramarital dalliances. By contrast, consider the public’s reaction to former New York governor Eliot Spitzer, who in 2008 was found to have committed adultery with prostitutes. While no moral paragon, Spitzer did have a virtuous history: As district attorney, he cracked down on sex trafficking, which no doubt reduced its number of victims. In light of these good deeds, was the public willing to excuse Spitzer for soliciting prostitutes? On the contrary, he was widely condemned as a hypocrite (Hakim & Santos, 2008). How can we explain these opposing reactions to King and Spitzer? Why does a history of good deeds only sometimes excuse transgressions?

One of the many differences between King and Spitzer is that King’s alleged infidelity was unrelated to his prior good deeds. That is, there is nothing inherently contradictory about advancing civil rights and committing adultery. By contrast, Spitzer’s infidelity was closely related to his prior good deeds. Someone who fights prostitution and then solicits prostitutes himself appears to be “saying one thing but doing another” or failing to “practice what he preaches” – defining features of hypocrisy (Barden, Rucker, & Petty, 2005; Stone & Fernandez, 2008). Spitzer’s fight against prostitution seems less sincere and perhaps more politically strategic when one learns of his subsequent crime. Not only do people dislike hypocrites (see Gilbert & Jones, 1986), they also feel a certain glee when hypocrites are punished (Smith, Powell, Combs, & Schurtz, 2009). This analysis suggests that when you have clearly done something wrong, good deeds can help you get off the hook – but only if those good deeds are unrelated to your transgression. Otherwise, hypocrisy will prevent you from getting a pass.

Spitzer and King represent interesting case studies, but numerous other factors could explain the public’s different reactions to them. To investigate the impact of good deeds on judgments of subsequent transgressions more systematically, my collaborator Benoît Monin and I conducted a series of experiments (Effron & Monin, 2010). In a first experiment, research participants (students at Stanford University) read a fictional newspaper article about a high school principal who had committed a blatant act of sexual harassment: While dining at a local restaurant, he had made provocative comments to a waitress, touched her in an unwelcome way, and offered her money for sex. Not surprisingly, participants condemned the principal and his behavior, and thought that he should resign from his job. 

A second group of participants read about the same act of harassment, but first they learned that the principal had a history of virtuous behavior that was unrelated to his transgression: He had worked tirelessly for years to reduce illegal drug use among students at his school. Participants in this second group were less condemning of the principal and his behavior towards the waitress, and were less eager for him to resign. In other words, the principal was more likely to be let off the hook for the same act of sexual harassment when he had done an unrelated good deed. Participants acted as if his transgression were at least somewhat balanced out by his prior acts of virtue.

A third group of participants read that the principal had a history of virtuous behavior that was closely related to his misdeed: Before committing sexual harassment himself, he had worked tirelessly for years to reduce sexual harassment among his students. Under these circumstances, participants reacted to the principal much like the public reacted to Eliot Spitzer. Instead of showing any inclination to reduce their condemnation of him, they labeled him as a hypocrite and dismissed his prior good deeds as insincere. Given this negative reaction, the principal would have been better off if his history of fighting sexual harassment had never come to light. 

We found complementary results using a different version of the newspaper article – one that described how the principal had committed a drug-related offense rather than an act of sexual harassment. Participants were relatively happy to let the principal off the hook for his drug offense when he had fought sexual harassment at his school (an unrelated good deed), but not when he had fought drug use (a related good deed). Once again, we see that a virtuous track record can help get one off the hook for an unrelated transgression, but will make one look like a hypocrite for committing a related transgression.

Note that participants in this study judged how much the principal should be condemned for wrongdoing that he had clearly committed. In this sense, participants were in a similar position to judges determining a sentence after a guilty verdict has already been rendered, or jurors during a trial who have no doubt that the defendant is guilty. Given the results of the experiment, defense attorneys should hesitate before introducing positive character evidence that seems closely related to a crime for which their client has already been convicted or that their client clearly committed. Rather than mitigating a sentence or softening judgments, good deeds that are related to blatant wrongdoing may invite attributions of hypocrisy (see top row of Table 1). 

Table 1: Effects of prior good deeds on judgments of an alleged wrongdoer

There are, however, some situations in which a virtuous history that is closely related to subsequent wrongdoing can be an asset to a defendant rather than a liability. Specifically, when a defendant’s guilt is uncertain, introducing positive character evidence may be less likely to raise the specter of hypocrisy and may, in fact, benefit the defendant. To illustrate, imagine a white corporate manager who has been accused of racial discrimination because, of the seven employees eligible for advancement, he promoted only the five who were white, and held back the two who were black. Although this decision falls suspiciously along racial lines, the manager claims that the more meritorious employees just happened to be white. Would the manager seem like a hypocrite if one knew that he had previously fought to reduce racial discrimination at his company? Probably not, even though these good deeds are closely related to his alleged wrongdoing. Instead, these good deeds make his promotion decision seem less likely to have been racially motivated. Because he is not clearly guilty, a history of virtuous behavior that is closely related to his alleged wrongdoing will likely lead people to give him the benefit of the doubt. By contrast, if he were clearly guilty of racial discrimination (e.g., if he had been recorded saying that he refused to promote racial minorities), then a history of opposing racial discrimination would likely make him look like a hypocrite.

We found evidence for this idea in a second experiment (Effron & Monin, 2010, Study 2). Research participants read a fictional newspaper article reporting a situation like the one described above: a white manager was accused of racial discrimination, but there was ambiguity surrounding whether he was guilty. Participants who had previously learned that the manager had a history of opposing sexual harassment (an unrelated good deed) condemned him less and were less supportive of taking legal action against him than participants who had been told nothing about the manager’s history. As we saw in our first study, these unrelated good deeds seemed to balance out any wrongdoing that the manager might have committed, but they did not make the allegations seem any less likely to be true.

A different group of participants who had previously learned that the manager had a history of opposing racial discrimination (a closely related good deed) expressed even less condemnation and were even less supportive of taking legal action. Instead of merely balancing out any wrongdoing that the manager many have committed, these related good deeds made participants more confident that the manager had not committed any wrongdoing at all. Rather than seeming hypocritical, the manager’s history of opposing racial discrimination made participants less likely to believe that he was guilty of committing such discrimination himself.

By contrast, a history of fighting racial discrimination did make the manager seem hypocritical to a different group of participants who read a version of the newspaper article in which the manager was clearly guilty of racial discrimination. In this version, participants were told that the media had uncovered emails in which the manager had written that the black employees’ race made them “unsuitable for management.” Because his history of fighting racial discrimination directly contradicted behavior that was clearly discriminatory, participants accused him of hypocrisy and urged legal action against him. 

These results suggest that introducing positive character evidence that is closely related to a client’s alleged wrongdoing can sometimes be an effective strategy for the defense. If the jury is unsure about the defendant’s guilt, this closely related character evidence could increase the odds of an exculpating verdict. But if the jury is already convinced that the defendant is guilty, this strategy could backfire by sparking perceptions of hypocrisy and increasing the jury’s willingness to condemn the defendant (see bottom row of Table 1). 

Summary and Implications

The results of our studies suggest that people are often willing to soften their judgments of a defendant in light of his or her history of good deeds – but whether and how much depends jointly on two factors which are summarized in Table 1: (a) how clearly guilty the defendant is of wrongdoing, and (b) how closely related people think the defendant’s good deeds are to the alleged wrongdoing.

When a defendant is clearly guilty (e.g., during a trial in which the evidence against the defendant is overwhelming, or during sentencing after a guilty verdict has already been rendered), emphasizing that he or she has a history of good deeds can be risky for the defense (see top row of Table 1). On the one hand, if the good deeds are perceived as being unrelated to the alleged wrongdoing, then this strategy will likely mitigate condemnation. For example, if Eliot Spitzer had had a history of fighting illegal drugs, then the public might not have responded with such outrage when they learned that he paid for sex. On the other hand, if the good deeds are perceived as being related to the alleged wrongdoing, then highlighting them can backfire: instead of mitigating condemnation, they can make the defendant seem like a hypocrite. This, of course, was the sort of public opinion that Spitzer actually faced when he paid for sex after fighting against prostitution.

When people are uncertain whether a defendant is guilty, there is less risk in emphasizing his or her history of good deeds (see bottom row of Table 1). In such situations, good deeds that seem unrelated to allegations may not convince many people that the defendant is innocent, but these good deeds can still mitigate condemnation by seeming to balance out any wrongdoing that may have occurred. By contrast, good deeds that seem related to the wrongdoing can convince people of the defendant’s innocence, and can thus dramatically reduce condemnation without seeming hypocritical. To use the language of evidence law, if jurors are unsure about the defendant’s guilt, they will treat good deeds that are related to the alleged crime as propensity evidence. If it had been unclear whether Spitzer was guilty of soliciting prostitutes, then his history of fighting prostitution would probably have convinced some people that he was innocent.

Some caveats should be kept in mind when applying our research findings. First, the fictional individuals in our studies performed good deeds before they allegedly committed transgressions. We might have found different results if they had performed good deeds only after the allegations against them had come to light. On the one hand, good deeds performed after a transgression could seem like a sincere act of atonement, lessening any perceptions of hypocrisy (Barden et al., 2005). On the other hand, good deeds performed after a transgression could seem like an insincere strategy for winning support and therefore magnify perceptions of hypocrisy. A second caveat is that the extent to which good deeds are related to a transgression is to some degree subjective. For example, is it hypocritical to condemn others for sexual harassment and then to commit racial discrimination? Our research participants tended to think not – but we suspect that the fact that harassment and discrimination both involve abuses of power, for example, would allow a clever prosecutor to convince jurors that it does. 


When deciding during a trial or during sentencing whether to introduce positive character evidence, defense attorneys should consider how closely related the good deeds will seem to the defendants’ wrongdoing, and how confident the judge and jury are about the defendant’s guilt. Otherwise, attempts to make a defendant look like a hero could backfire, and instead make him or her look like a hypocrite. 

Daniel A. Effron is a visiting assistant professor and postdoctoral fellow at the Kellogg School of Management at Northwestern University.  He recently received his PhD in social psychology from Stanford University.  Dr. Effron’s research interests focus on how people judge moral transgressions, and on the psychological processes that allow people to act in morally questionable ways without compunction.  Information about his research may be found at


Abernathy, R. D. (1989). And the walls came tumbling down. New York: Harper & Row.

Barden, J., Rucker, D. D., & Petty, R. E. (2005). "Saying one thing and doing another": Examining the impact of event order on hypocrisy judgments of others. Personality and Social Psychology Bulletin, 31, 1463-1474.

Birnbaum, M. H. (1973). Morality judgment: Test of an averaging model with differential weights. Journal of Experimental Psychology, 99, 395-399.

Effron, D. A., & Monin, B. (2010). Letting people off the hook: When do good deeds excuse transgressions? Personality and Social Psychology Bulletin, 36, 1618 –1634.

Gilbert, D. T., & Jones, E. E. (1986). Exemplification: The self-presentation of moral character. Journal of Personality, 54, 593-593.

Hakim, D., & Santos, F. (2008). G.O.P. Consultant says he reported Spitzer trysts in 2007. The New York Times. Retrieved from

Hunt, J. S., & Budesheim, T. L. (2004). How jurors use and misuse character evidence. Journal of Applied Psychology, 89, 347-361.

Lee, P. G. (2011). White-collar criminal hopes good deeds will lead to lighter sentence. The Wall Street Journal Law Blog. Retrieved June 22, 2012, from

Maeder, E. M., & Hunt, J. S. (2011). Talking about a black man: The influence of defendant and character witness race on jurors' use of character evidence. Behavioral Sciences and the Law, 29, 608-620.

Ross, J. (2004). "He looks guilty": Reforming good character evidence to undercut the presumption of guilt. University of Pittsburgh Law Review, 65, 227-279.

Smith, R. H., Powell, C. A. J., Combs, D. J. Y., & Schurtz, D. R. (2009). Exploring the when and why of schadenfreude. Social and Personality Psychology Compass, 3, 530-546.

Stone, J., & Fernandez, N. C. (2008). To practice what we preach: The use of hypocrisy and cognitive dissonance to motivate behavior change. Social and Personality Psychology Compass, 2, 1024-1051.

Uviller, H. R. (1993). Credence, character, and the rules of evidence: Seeing through the liar's tale. Duke Law Journal, 42, 776-839.

We asked two trial consultants to respond to Daniel Effron's article.

Holly G. VanLeuven, MA, has been a practicing Trial Consultant since 1972. She is president of Genesis Group in Scottsdale, Arizona.

Holly VanLeuven responds:

HERO OR HYPOCRITE? A Continuum of Likely Judgment

Effron’s study is based upon responses by college students to a series of fictional newspaper accounts describing a variety of transgressions along with a variety of types of good deeds attributed to the alleged transgressor.  I haven’t seen the statistics from Dr. Effron’s study.  His conclusions do, however, coincide with conventional wisdom and my personal experience.

How and to whom will these conclusions be useful? At the very least, a review of this research gives the signal to trial consultants and members of the litigating team to seriously consider how to deal with positive and negative character evidence.  

Risks and benefits of character references deserve a prominent place on every trial strategy check-list. What will be the impact of a parade of people who have had personal experience with the client’s good and/or bad acts?  Is it better to have none, some or many people?  Is it better to try to match character witnesses to the demographics of the jury…or doesn’t it matter?  If the client is well-known to be a person of ill-repute, is it smart to acknowledge that up front, going on to say that in spite of past deeds, the law requires that the accused have a fair trial to determine his/her culpability for the charges at hand.  If the client has been seen in the past as an honorable person…a priest, a teacher, a Sandusky…can that possibly mitigate an overwhelming body of evidence against him/her or is it instead seen as lack of integrity from the beginning, earning the client the dreaded “Hypocrite” label?  The great danger of failing to develop the best strategy is in compromising the jury’s perception of the accused as innocent until proven guilty.

These  questions, and especially the answers to them, are very tricky and will vary from case to case.  The important thing is to ask the questions and to KNOW they are tricky.  With that knowledge, the trial team can carefully develop the answers and test the answers by means of talking it through with experts and people whose judgment you trust, doing focus groups or mock trials, as well as simply applying your own knowledge and experience to the issues.

Katherine James, MFA is a trial consultant based in Culver City, CA. Her specialization is live communication skills.

Katherine James responds:

What a fascinating theory that Effron puts forward. I wonder if the results would change if, instead of dealing only with students, his study subjects included people of many generations and diverse backgrounds to more closely represent real-world jury panels.

The concept of “atonement” – making up for past sins – has some pretty deep roots in human experience. For example, making a racist remark in the workplace, having it pointed out and then working to end racist comments in that very workplace seems to be in keeping with “atonement.” If the purpose of punishing someone is to get that person to pay for his or her sins, well, the payment has been made.

However, setting up a program to end racism in the workplace, then committing a racist act in the workplace is hypocritical. My experience tells me that younger jurors are less forgiving than older jurors in this regard. The older we are, the more likely we are to have learned from our previous mistakes. We realize we are works in progress, and unless the transgression is heinous, we are more apt to forgive than a younger, more idealistic juror. 

I am certainly not the only trial consultant who has experienced a major backfire of inundating the jurors with the “good deeds” of a corporation when working to defend a lawsuit. No one wants to hear how you helped the local youth basketball league if you did it with the money that you earned while ripping off someone else’s intellectual property. 

Although Effron’s conclusions seem logical, a study that involves not only students but a well-rounded and diverse group of people would do more to reassure me of the accuracy of his research. 

Perhaps Effron and a dedicated group at ASTC could further develop this theory through research. It certainly is a question at the heart of both civil and criminal cases.