Skip to content
Search resources Open menu

Jurors and Jury Decision-Making

Nullification and Compromise Verdicts

Published

Snap Shot

  • The question: How often do jurors and court-martial members return verdicts based on their views of the law or the penalty rather than the evidence, and how do lesser-offense options shape compromise verdicts?
  • What the research shows: Studies of real trials, based on 1950s judge-jury comparisons and a later partial replication, suggest that nullification is uncommon and that evidence strength is the main driver of verdicts, though jurors’ views of the law’s fairness were associated with hung juries. In mock trials, explicit nullification instructions shifted verdicts in both directions depending on the case and increased the influence of jurors’ attitudes and emotions. Offering lesser offenses changed the distribution of verdicts in early individual-participant experiments, though a reanalysis qualified some conclusions, and research on penalty severity is mixed.
  • What it does not show: The research does not show that any particular verdict was a nullification or a compromise, and it does not show that nullification favors one side. Most experiments use euthanasia or similar scenarios, mock jurors, and older designs.
  • The law: Juries have the power to acquit against the evidence but no right to be told so (Sparf; Dougherty; Hardy), and inconsistent verdicts generally stand (Dunn; Powell). Lesser-included-offense instructions guard against all-or-nothing choices (Keeble; Beck; Schad). Arizona allows jurors to consider a lesser offense after reasonable efforts fail (LeBlanc); courts-martial vote on the charged offense first (R.C.M. 921).
  • Issues for each side: Prosecutors can draw on the research showing that evidence strength is the main driver of verdicts, while recognizing that jurors’ views of the law’s fairness as applied can affect outcomes; defense counsel can draw on lesser-offense law to avoid all-or-nothing choices, while recognizing that courts generally do not permit nullification arguments and that a lesser option can also produce conviction.

Jury nullification refers to a verdict of acquittal returned despite evidence that would legally support conviction, because the jury believes that the law, its application to the defendant, or the likely punishment is unjust. Because a general verdict of not guilty cannot be reviewed and jurors cannot be punished for their verdicts, juries have the power to nullify, although courts have consistently held that they have no right to be told of that power. A related phenomenon is the compromise verdict, in which jurors who disagree about the most serious charge settle on a lesser offense or on conviction of some counts and acquittal of others. Both raise the same question: how far do verdicts reflect jurors’ values and the deliberative process, rather than the evidence and the law as instructed?

These issues reach court in several ways: in requests for nullification instructions or argument, in voir dire about attitudes toward a law or its penalties, in motions to remove a juror who has said he or she will not follow the law, in disputes over lesser-included-offense instructions and the order in which jurors may consider them, and in challenges to inconsistent verdicts. This paper summarizes research on how often nullification occurs, the effect of nullification instructions, the influence of penalty knowledge, and compromise among offense options, followed by the governing federal, military, and Arizona law.

The research applies in both directions. Jurors’ values can produce leniency in a sympathetic case and harshness in an unsympathetic one, and compromise can produce a conviction on a lesser charge that some jurors would not otherwise have supported, or an acquittal on a greater charge that some would have supported.

How Often Juries Depart From the Evidence

The foundational evidence comes from The American Jury, in which trial judges reported how they would have decided 3,576 criminal jury trials in the 1950s. Judges and juries agreed in roughly three quarters of cases, and when they disagreed the jury was usually more lenient. The authors attributed most of the disagreement to different interpretations of the reasonable-doubt standard and also pointed to jurors’ sentiments about the defendant, which they believed influenced juries when the judge regarded the case as close (Kalven & Zeisel, 1966, as cited in Eisenberg et al., 2005). A related idea was later tested as the liberation hypothesis (Devine et al., 2009). A partial replication with felony trials from four jurisdictions, including Maricopa County, found a similar level of agreement, with disagreements concentrated in cases of middling evidentiary strength (Eisenberg et al., 2005). On this evidence, which is decades old and describes groups of trials rather than any particular jury, departures from the evidence appear uncommon and to occur mainly at the margins.

A study of hung juries in 372 felony trials in four large urban courts, including Maricopa County, examined the role of nullification directly. Using surveys of judges, attorneys, and jurors, the researchers found that evidentiary characteristics were the most important factors in both acquittals and hung juries, and that the interpersonal dynamics of deliberation also mattered. Jurors’ concerns about the fairness of the law as applied were a significant predictor of hung juries, although jurors generally rated the fairness of the law and of the outcome favorably. The authors concluded that nullification is unlikely to play a dominant role in most cases (Hannaford-Agor & Hans, 2003). The study measured jurors’ perceptions after trial; it could not determine whether any particular jury hung because some jurors refused to apply the law.

Nullification Instructions and Arguments

Because courts do not tell juries of their power to nullify, most research on nullification is experimental, comparing mock juries given standard instructions with those given nullification instructions. In an early study, 45 six-person mock juries heard audiotaped trials of a murder during a robbery, a vehicular homicide by a drunk driver, and a nurse’s mercy killing of a terminally ill cancer patient, with the evidence of guilt clear in each. Juries given explicit nullification instructions spent less time discussing the evidence and more time discussing personal experiences. In the mercy-killing case they returned more not-guilty verdicts, but in the drunk-driving case they returned fewer not-guilty verdicts and harsher verdicts. Instructions made no difference in the robbery-murder case, and a milder instruction that mentioned nullification did not change verdicts compared with standard instructions (Horowitz, 1985). Nullification instructions, on this limited evidence from a small number of scenarios, do not simply produce leniency; they amplify jurors’ reactions to the particular case, in either direction.

Later studies tested why. In an online experiment comparing a murder for profit with a mercy killing, information that affected jurors’ liking for the victim influenced verdicts and emotional reactions only when jurors had received nullification instructions in the mercy-killing case; the information did not change how jurors processed the evidence (Horowitz et al., 2006). In a mock euthanasia trial, nullification instructions strengthened the effect of jurors’ pretrial attitudes about euthanasia on their verdicts by increasing their reliance on moral outrage toward the defendant, while outrage at the law itself predicted lenient verdicts with or without nullification instructions (Peter-Hagene & Ratliff, 2021). Another euthanasia study found that acquittals were most likely when jurors received nullification instructions and the defendant was the decedent’s spouse (Scott et al., 2023). These studies used a narrow range of cases, mostly mercy killings, in which many people view the law as morally contested.

Jurors also differ in their general willingness to rely on their own sense of justice. In two studies, jurors who scored high on a measure of nullification beliefs, or who received instructions encouraging reliance on their own sense of justice, were more influenced by evidence that had been ruled inadmissible (Cook et al., 2004). Attitudes of this kind are a proper subject of inquiry in voir dire, as discussed in the hub’s paper on voir dire, but no measure of nullification beliefs has been validated for excusing an individual juror.

The research does not show how often nullification occurs in particular kinds of cases today. Commentators have debated whether nullification can be justified as a response to perceived unfairness in the enforcement of particular laws, including in relation to race (Butler, 1995), and others respond that it invites inconsistent and potentially biased outcomes. The empirical studies summarized here do not resolve that debate; they show that nullification instructions increase the influence of jurors’ attitudes and emotions, whatever their direction.

Penalty Knowledge and Conviction

Jurors in most American criminal trials are not told the penalty, and they are instructed not to consider it. One reason is the concern that jurors who learn of a severe penalty will require more evidence to convict or will acquit despite the evidence. An early experiment suggested that mock jurors were less willing to convict when the prescribed penalty was severe (Kerr, 1978). A later series of four experiments, with penalties ranging from a few years’ imprisonment to life, found no indication that mock jurors were less likely to convict under harsher penalties, so long as the evidence required for conviction was held constant (Freedman et al., 1994). In a study of students judging an academic cheating case, conviction votes were lowest when a severe penalty was controlled by an authority and participants believed their votes had real consequences (Kaplan & Krupa, 1986). The research is mixed, the designs vary, and the studies are old; it does not establish how jurors in modern trials respond when they know or suspect a mandatory or severe penalty.

Compromise and Lesser-Offense Options

The options a jury is given affect how its members distribute their verdicts. In an early experiment, mock jurors who could choose only between conviction of the most serious offense and acquittal acquitted more often than those who were also offered lesser offenses (Vidmar, 1972); a reanalysis of the data qualified some of the original conclusions (Larntz, 1975). Research on decision-making more generally has shown a compromise effect, in which an option is chosen more often when it falls between two alternatives than when it is the most extreme choice offered. Experiments using legal scenarios found that adding a more serious or a less serious charge changed how often participants chose an intermediate offense such as manslaughter (Kelman et al., 1996). These studies used individual participants without deliberation.

Within deliberating juries, compromise also arises from the process of reaching agreement. Juries that disagree about the most serious charge may settle on a lesser offense or divide their verdicts across counts. The hub’s paper on deliberation reviews research showing that the first-ballot majority usually prevails and that factions favoring acquittal tend to be more influential than equal factions favoring conviction; compromise on a lesser charge is one way such disagreements are resolved. The research does not show whether a particular split or lesser-offense verdict reflects compromise or a considered judgment that the evidence proved only the lesser offense.

The legal concern runs in both directions. Without a lesser option, jurors convinced that the defendant did something wrong but doubtful about an element of the greater offense may convict of the greater offense rather than acquit; with a lesser option, jurors who would have acquitted may agree to convict of the lesser offense to end deliberations. The experimental evidence supports the general premise that options matter, but it does not show which effect predominates in real trials.

Limits of the Evidence

Most nullification experiments use mock jurors, written or audiotaped trials, and a narrow set of morally contested scenarios, especially mercy killing, so they may not generalize to drug, firearm, sexual offense, or military cases. Several foundational studies are decades old. The real-trial studies rely on judges’ and jurors’ after-the-fact reports and cannot identify individual nullifying verdicts. The studies reviewed here did not report differences by juror gender, so the research does not show whether men and women respond differently to nullification instructions or lesser-offense options. No study reviewed here examined court-martial members, who vote by secret written ballot under a three-fourths rule and consider lesser offenses in a fixed order.

Above all, the research describes group tendencies. It cannot show that a particular verdict was a nullification or a compromise, and an inconsistent or lesser verdict may reflect a jury’s reasoned view of the evidence.

Questions to Ask About This Research

When Horowitz (1985) is cited to show that nullification instructions produce leniency, ask what exactly was found, since the instructions produced more acquittals in the mercy-killing case but harsher verdicts in the drunk-driving case and no change in the robbery-murder case. When the finding that concerns about the law’s fairness predicted hung juries (Hannaford-Agor & Hans, 2003) is cited to show that nullification is common, ask how large the effect was and whether the authors drew that conclusion, since they found evidence strength most important and nullification unlikely to dominate. When Kerr (1978) is cited to show that penalty knowledge reduces convictions, ask whether the finding has held up, since Freedman et al. (1994) found no such effect across four experiments. When the euthanasia studies (Peter-Hagene & Ratliff, 2021; Scott et al., 2023) are applied to another type of case, ask who was studied and whether a morally contested mercy-killing scenario applies. These questions are explained in “Reading Research Claims: A Guide for Attorneys” on this site.

Nullification, Compromise Verdicts, and the Law

Federal and Military Standards

In Sparf v. United States (1895), the Supreme Court held that federal juries must take the law from the court, and courts since have held that a defendant is not entitled to an instruction informing jurors of their power to nullify. United States v. Dougherty (1972) acknowledged that the jury has that power but upheld the refusal to tell jurors of it. A juror who is shown to be refusing to follow the law may be removed during deliberations, but in United States v. Thomas (1997) the Second Circuit held that a deliberating juror accused of refusing to follow the law may not be removed if the record shows any possibility that the juror is unpersuaded by the evidence, a rule that the same court later limited to alleged nullification (United States v. Spruill, 2015). Jurors are generally not told the consequences of their verdict; in Shannon v. United States (1994), the Court held that a federal jury need not be instructed on the consequences of a verdict of not guilty by reason of insanity, reasoning that jurors should reach their verdict without regard to what sentence might be imposed.

Inconsistent verdicts generally stand. In Dunn v. United States (1932), the Court upheld a conviction on one count despite acquittal on another count resting on the same evidence, and in United States v. Powell (1984) it reaffirmed that rule, observing that inconsistency may reflect lenity, compromise, or mistake, and that a court cannot know which. Lesser-included-offense instructions are governed by Federal Rule of Criminal Procedure 31(c). In Keeble v. United States (1973), the Court held that a defendant is entitled to such an instruction when the evidence would permit a jury rationally to find him guilty of the lesser offense and acquit of the greater, reasoning that a jury forced to choose between conviction and acquittal may resolve doubts in favor of conviction. Beck v. Alabama (1980) held that a death sentence may not be imposed after a jury was denied the option of a lesser included noncapital offense supported by the evidence, and Schad v. Arizona (1991), an Arizona case, held that Beck was satisfied when the jury was offered second-degree murder. Schad also held, in a plurality opinion, that jurors need not agree on whether a first-degree murder was premeditated or a felony murder under Arizona’s statute. In United States v. Tsanas (1978), the Second Circuit described the risk that jurors may reach a compromise verdict on a lesser offense to avoid a mistrial, held either sequential-consideration instruction legally permissible, and held that the defendant is entitled to elect the alternative on timely request.

In courts-martial, members serve the jury’s function. The Court of Appeals for the Armed Forces held in United States v. Hardy (1997) that there is no right to nullification and that the military judge did not err in declining to instruct members on a power to disregard the law, while also holding that the judge may not direct a verdict of guilty and noting that protecting deliberations outweighs an occasional disregard of the law. In United States v. Smith (1988), the Court of Military Appeals upheld a military judge’s refusal to let defense counsel ask members whether they knew that the charged offense carried a mandatory life sentence, where the purpose was nullification. Article 79, UCMJ, permits findings of guilty of a lesser included offense, and United States v. Jones (2010) adopted an elements test for identifying such offenses. Under R.C.M. 921(c), members vote by secret written ballot, a finding of guilty requires at least three-fourths of the members present, and members may not vote on a lesser included offense until a finding of not guilty of the charged offense has been reached, after which they vote on included offenses in order of severity.

Arizona

Arizona’s pattern instructions tell jurors that they must follow the law whether they agree with it or not (RAJI Preliminary Criminal 2), that they must not be influenced by sympathy or prejudice (Standard Criminal 1), and that they must not consider the possible punishment, which is left to the judge (Standard Criminal 5). In State v. LeBlanc (1996), the Arizona Supreme Court rejected a requirement that jurors unanimously acquit of the greater offense before considering a lesser one, and adopted an instruction allowing jurors to consider the lesser offense if, after full and careful consideration, they cannot agree on the greater; the pattern lesser-included-offense instruction follows LeBlanc (Standard Criminal 36). A defendant is entitled to a lesser-included-offense instruction when the evidence supports it, and in State v. Wall (2006) the court held that the instruction may not be refused on the ground that the defendant pursued an all-or-nothing defense. When a jury reports an impasse, Arizona Rule of Criminal Procedure 22.4 allows the court to ask how it can assist deliberations.

The Arizona and military approaches differ in a way the research bears on. The LeBlanc instruction is designed to reduce deadlock without requiring jurors to acquit of the greater offense first, while the military’s ordered voting requires a finding of not guilty of the charged offense before members consider a lesser one. Research directly comparing these orderings is sparse, and the compromise research suggests that the structure of the choice can affect outcomes in either direction.

Special Issues in Legal Practice

The principles described above apply with equal force to every party in a legal proceeding. The science operates the same way regardless of who is asking the questions, and the research summarized here does not by itself favor either side. What differs between the parties is not the science but the vantage point — the particular places where these issues tend to surface, and the particular errors each party is best positioned to guard against. The two summaries below identify those recurring issues. They are offered for orientation, not as litigation strategy; in every case the controlling question is what the evidence in that case will, and will not, support, and the expert’s role is to inform the trier of fact rather than to advocate for either side.

Special Issues for Prosecutors

The research supports the position that evidence strength is the main determinant of verdicts, that nullification is unlikely to dominate in most cases, and that explicit nullification instructions, which courts do not give, increase the influence of attitudes and emotions on verdicts, in either direction; the studies did not measure verdict accuracy. Voir dire focused on whether jurors can apply the law as instructed, including laws some jurors may view as unfair, is consistent with the research showing that such attitudes vary among individuals, although no measure of them has been validated for excusing a particular juror.

The research also counsels caution. Jurors’ views of the fairness of the law as applied were associated with hung juries in one study (Hannaford-Agor & Hans, 2003), a perception that may bear on outcomes even without any nullification argument. Lesser-included-offense instructions supported by the evidence must be given, and in Arizona they may not be refused because the defense pursued an all-or-nothing strategy; the research on decision options suggests that the structure of the choice can affect outcomes in either direction.

Special Issues for Defense

The research supports requests for lesser-included-offense instructions where the evidence permits, consistent with Keeble and Beck, and it supports voir dire into attitudes that bear on whether jurors can fairly consider the defense. It is consistent with the observation in Powell that an inconsistent or split verdict may reflect compromise, lenity, or mistake, but it cannot show which occurred in a particular case, and the law rarely allows such verdicts to be set aside.

The research also counsels caution. Courts generally do not permit argument urging nullification, and in courts-martial Hardy upheld the refusal of a nullification instruction and Smith upheld a judge’s restriction of voir dire aimed at nullification. Nullification instructions in experiments produced harsher as well as more lenient verdicts, and the evidence that penalty knowledge reduces convictions is mixed. The availability of a lesser offense can lead to conviction of that offense by jurors who might otherwise have acquitted.

This paper is for educational purposes only and is not legal advice. The author is a psychologist, not an attorney; summaries of law are general and must be verified against current authority or with a licensed attorney.

References

Butler, P. (1995). Racially based jury nullification: Black power in the criminal justice system. Yale Law Journal, 105(3), 677–725. https://doi.org/10.2307/797197

Cook, A., Arndt, J., & Lieberman, J. D. (2004). Firing back at the backfire effect: The influence of mortality salience and nullification beliefs on reactions to inadmissible evidence. Law and Human Behavior, 28(4), 389–410. https://doi.org/10.1023/B:LAHU.0000039332.21386.f4

Devine, D. J., Buddenbaum, J., Houp, S., Studebaker, N., & Stolle, D. P. (2009). Strength of evidence, extraevidentiary influence, and the liberation hypothesis: Data from the field. Law and Human Behavior, 33(2), 136–148. https://doi.org/10.1007/s10979-008-9144-x

Eisenberg, T., Hannaford-Agor, P. L., Hans, V. P., Waters, N. L., Munsterman, G. T., Schwab, S. J., & Wells, M. T. (2005). Judge-jury agreement in criminal cases: A partial replication of Kalven and Zeisel’s The American Jury. Journal of Empirical Legal Studies, 2(1), 171–207. https://doi.org/10.1111/j.1740-1461.2005.00035.x

Freedman, J. L., Krismer, K., MacDonald, J. E., & Cunningham, J. A. (1994). Severity of penalty, seriousness of the charge, and mock jurors’ verdicts. Law and Human Behavior, 18(2), 189–202. https://doi.org/10.1007/BF01499015

Hannaford-Agor, P. L., & Hans, V. P. (2003). Nullification at work? A glimpse from the National Center for State Courts study of hung juries. Chicago-Kent Law Review, 78(3), 1249–1277.

Horowitz, I. A. (1985). The effect of jury nullification instruction on verdicts and jury functioning in criminal trials. Law and Human Behavior, 9(1), 25–36. https://doi.org/10.1007/BF01044287

Horowitz, I. A., Kerr, N. L., Park, E. S., & Gockel, C. (2006). Chaos in the courtroom reconsidered: Emotional bias and juror nullification. Law and Human Behavior, 30(2), 163–181. https://doi.org/10.1007/s10979-006-9028-x

Kaplan, M. F., & Krupa, S. (1986). Severe penalties under the control of others can reduce guilt verdicts. Law and Psychology Review, 10, 1–18.

Kelman, M., Rottenstreich, Y., & Tversky, A. (1996). Context-dependence in legal decision making. Journal of Legal Studies, 25(2), 287–318. https://doi.org/10.1086/467979

Kerr, N. L. (1978). Severity of prescribed penalty and mock jurors’ verdicts. Journal of Personality and Social Psychology, 36(12), 1431–1442. https://doi.org/10.1037/0022-3514.36.12.1431

Larntz, K. (1975). Reanalysis of Vidmar’s data on the effects of decision alternatives on verdicts of simulated jurors. Journal of Personality and Social Psychology, 31(1), 123–125. https://doi.org/10.1037/h0076237

Peter-Hagene, L. C., & Ratliff, C. L. (2021). When jurors’ moral judgments result in jury nullification: Moral outrage at the law as a mediator of euthanasia attitudes on verdicts. Psychiatry, Psychology and Law, 28(1), 27–49. https://doi.org/10.1080/13218719.2020.1751741

Scott, J. D., Bell, D., Barry, B., & Edlund, J. E. (2023). Juror decision making and euthanasia: Exploring the role of jury nullification, manner of death, and defendant-decedent relationship. Psychological Reports, 126(6), 3052–3070. https://doi.org/10.1177/00332941221093244

Vidmar, N. (1972). Effects of decision alternatives on the verdicts and social perceptions of simulated jurors. Journal of Personality and Social Psychology, 22(2), 211–218. https://doi.org/10.1037/h0032605

Legal Authorities

Beck v. Alabama, 447 U.S. 625 (1980).

Dunn v. United States, 284 U.S. 390 (1932).

Keeble v. United States, 412 U.S. 205 (1973).

Schad v. Arizona, 501 U.S. 624 (1991) (plurality opinion of Souter, J., on the juror-agreement holding; Scalia, J., concurring in part and in the judgment).

Shannon v. United States, 512 U.S. 573 (1994).

Sparf v. United States, 156 U.S. 51 (1895).

State v. LeBlanc, 186 Ariz. 437 (1996).

State v. Wall, 212 Ariz. 1 (2006).

United States v. Dougherty, 473 F.2d 1113 (D.C. Cir. 1972).

United States v. Hardy, 46 M.J. 67 (C.A.A.F. 1997).

United States v. Jones, 68 M.J. 465 (C.A.A.F. 2010).

United States v. Powell, 469 U.S. 57 (1984).

United States v. Smith, 27 M.J. 25 (C.M.A. 1988).

United States v. Spruill, 808 F.3d 585 (2d Cir. 2015).

United States v. Thomas, 116 F.3d 606 (2d Cir. 1997).

United States v. Tsanas, 572 F.2d 340 (2d Cir. 1978).

Article 79, UCMJ, 10 U.S.C. § 879.

Fed. R. Crim. P. 31(c).

Ariz. R. Crim. P. 22.4.

Manual for Courts-Martial, United States, R.C.M. 921(c).

Revised Arizona Jury Instructions (Criminal), Preliminary Criminal 2 and Standard Criminal 1, 5, and 36 (6th ed. 2022).

Related topic papers: Deliberation, Unanimity, and Court-Martial Verdicts · Jury Instructions: Comprehension and Effectiveness · Voir Dire and the Detection of Juror Bias · Command Climate, Unlawful Command Influence, and Court-Martial Members · Closing Argument and Emotion

Related topics from Arizona Forensic Services: Courts-Martial