The Yale Law Journal

June 2016

Ideological Imbalance and the Peremptory Challenge

Joshua Revesz

Legal scholars, by and large, revile peremptory challenges. Allowing parties to unilaterally strike prospective jurors without explanation has been attacked as undemocratic,1 as prone to manipulation,2 as a potential First Amendment violation,3 and—most often of all—as racist.4 Judges5 and even prosecutors6 have spoken out against the procedure. And, although the Supreme Court sought in Batson v. Kentucky7 to limit the problems of peremptory challenges by constraining parties’ ability to strike jurors because of their race,8 Batson’s rule is decried as “almost surely a failure”9 and an “enforcement nightmare.”10

Studies suggest that, despite Batson and its subsequent cases, demographic profiling remains a principal strategy during voir dire.11 Yet some of the same studies show that, on net, the resulting demographic composition of juries is similar to that of their venires.12 As long as parties’ race- and gender-based peremptory challenges can “cancel each other out,”13 the argument goes, those challenges will not produce significant demographic disparities.14

This Comment contributes to these debates by noting a different disparity that peremptory challenges engender. It argues that peremptory challenges produce an ideological skew in juries. Because ideology and demographics (particularly race and gender) are asymmetrically correlated in the United States, demographic-based peremptory challenges alter juries’ ideological makeup. Specifically, because liberal jurors are easier to identify from demographic profiling than their conservative counterparts, the peremptory-challenge regime likely produces more conservative juries than would a system without those challenges. That bias disadvantages certain litigants, from tort plaintiffs to criminal defendants.

Part I of this Comment explores the intertwined roles of ideology and race in the jury-selection process. Part II formalizes this relationship, mapping a causal claim that peremptory challenges shift jury ideology rightward. It suggests that peremptory-challenge procedures produce juries that are considerably more conservative than a random sampling of Americans. This result reveals an additional axis of identity on which peremptory challenges discriminate, strengthening the case for curtailing or eliminating the practice.

I. ideology, race, and batson

Juror ideology matters. Studies show that juror ideology affects outcomes in both criminal15 and civil16 cases. Litigators agree: the American Association for Justice, formerly known as the Association of Trial Lawyers of America, provides strategies for handling “conservative” jurors;17 self-described “jury consultants” advise lawyers to change their strategies depending on the ideological makeup of the jury.18 Even courts, on occasion, acknowledge that ideology affects jury decision making.19

Because juror ideology can play an important role in the disposition of a case, litigators consider juror ideology when strategizing over peremptory challenges. Lawyers use peremptory challenges to strike jurors “thought to be inclined against their interests.”20 Peremptory challenges, in theory, “assur[e] the selection of a qualified and unbiased jury”21 by “eliminat[ing] extremes of partiality on both sides.”22 The Equal Protection Clause, however, constrains the use of the peremptory challenge. In Batson v. Kentucky, the Supreme Court barred the use of peremptory challenges to strike black jurors solely on the basis of their race.23 Yet Batson allows prosecutors to strike black jurors by asserting that they are striking them not on the basis of race, but on the basis of their perceived ideological bent. In other words, nothing prevents attorneys from striking prospective jurors based on their perceived ideologies, and nothing meaningfully impairs attorneys’ use of demographic stereotyping as a guide to jurors’ ideologies.

Post-Batson case law is full of examples confirming the importance of ideology to litigators during voir dire—and hinting that those attorneys’ perceptions of prospective jurors’ ideology are closely linked to those jurors’ race and gender. In multiple cases, in response to a Batson challenge, parties have (generally successfully) defended their decision to strike minority and women jurors on the basis of perceived ideology. In a civil suit between a black plaintiff and white defendants, for example, the defendants explained their decision to strike an unemployed black woman on the grounds that, based on those characteristics alone, she would be “an unduly liberal juror.”24

A simple model of juror psychology—that jurors tend to favor litigants with similar demographics25—cannot fully explain such strikes. In one case, the defendant belonged to a different race than the juror the prosecution challenged;26 in many cases, the challenged juror was a woman and the defendant was a man.27 These cases show the difficulty of separating demographics from ideology in assessing peremptory challenges: in each instance, prosecutors struck jurors not because the jurors shared the defendants’ characteristics, but rather because the prosecutors assumed—based largely on the jurors’ appearance—that the strikes would produce an ideologically favorable jury.

The Batson majority overlooked this phenomenon. Its logic rested on what Neil Gotanda calls “formal-race analysis,” which emphasizes a shallow racial equality by presuming that “racial classifications are unconnected to social status or historical experience.”28 As Gotanda notes, Batson’s argument that “[a] person’s race simply ‘is unrelated to his fitness as a juror’”29 “invokes that unconnectedness of a juror’s formal-race classification to any other personal attributes which might relate to jury duty.”30 But, in reality, race—and many other observable characteristics—are correlated with a range of experiences and attitudes.31 As Frederick Schauer observed, Batson fails to prohibit “the use of race as a generalization . . . where the generalization is statistically legitimate[.]”32 Thus, Batson does not stand in the way of attorneys’ use of race as a proxy for ideology.

In the election-law context, an analogous difficulty has engendered the “party-race problem.” Legislation and redistricting schemes that seek to weaken Democrats have deleterious effects on minority voters.33 When minority plaintiffs—or the Department of Justice on behalf of minority voters—bring challenges under the Voting Rights Act, Republican legislatures respond that they sought to disadvantage Democrats, not minorities.34 As long as election manipulation for partisan advantage is legal, it will be impossible to separate these claims. Similarly, as long as peremptory challenges for ideological advantage are permissible, it will be impossible to administer peremptory challenges in a racially neutral manner.

The connection between race and ideology, therefore, poses twin dangers. First, ideology may serve as a pretext for racially motivated strikes.35 Second, ideologically motivated strikes, legal under Batson, may nevertheless be based on demographic profiling. Part II explores the ramifications of this second point: it demonstrates that the use of demographic profiling to strike liberal jurors creates juries that are more conservative than the general population.

II. peremptory challenges and ideological bias

Independent from the argument that demographic profiling to discriminate by ideology is merely a pretext for race discrimination, demographic profiling for ideological discrimination poses additional threats to overall jury impartiality. Specifically, because ideology and demographics are asymmetrically correlated, the use of demographic profiling in peremptory challenges is likely to result in juries that are more ideologically conservative than the venire population. For this reason, litigants whose positions typically attract liberal support are hurt by the peremptory-challenge system.

A. Stereotyping for Ideological Challenges

Attorneys cannot try their cases during voir dire. They must therefore use heuristics—what then-Justice Rehnquist called “seat-of-the-pants instincts”36—to assess prospective jurors’ desirability. “Yet ‘seat-of-the-pants instincts’ may often be just another term for racial prejudice”37: because attorneys have no concrete information on which to rely, they often turn to stereotyping.

Indeed, evidence suggests that attorneys lodge peremptory challenges based on a small number of easily discernible characteristics. In voir dire, “questioning jurors is principally a matter of getting basic background information from which [attorneys] infer likely beliefs and attitudes relevant to the case being tried.”38 Stephen Adler observes that “the legal profession has perpetuated a vast, mostly secretive lore concerning what sorts of people make what sorts of jurors, and these classifications usually weigh heavily towards ethnic, class, and racial stereotypes.”39 Experiments show that perceptions of race40 and gender41 better predict peremptory challenges than all other information provided. Justice Breyer has remarked that “the use of race- and gender-based stereotypes in the jury-selection process seems better organized and more systematized than ever before.”42

To be sure, the opportunities for questioning that voir dire provides might mitigate this effect: if attorneys can learn more about potential jurors, they can base their challenges on “beliefs and attitudes” relevant to the case at hand. But, as Charles Ogletree noted, the “often restrictive nature” of voir dire makes it difficult for attorneys to acquire relevant, nonstereotypical information about prospective jurors.43 In many jurisdictions, including in the federal courts, judges may conduct voir dire and need not provide opportunities for litigants to ask questions.44 A 1994 survey revealed that only fifty-four percent of federal judges allowed attorney questioning during voir dire.45 There is no reason to believe that figure has changed dramatically in subsequent decades.

Even when they may ask questions, attorneys might find stereotypes inescapable. Empirical work suggests that attorneys engage in “biased hypothesis testing” when asking questions during voir dire—that is, their questions merely confirm their preexisting hypotheses about prospective jurors’ preferences.46 Despite scholars’ advocacy to expand voir dire to allow attorneys to make better-informed, nondiscriminatory decisions,47 attorney questioning might nevertheless be infected by the stereotypes discussed above. Reid Hastie has made a similar point, observing that increased attorney questioning does not assist attorneys in making judgments about prospective jurors.48 As a result, more attorney questioning may not meaningfully expand the number of characteristics on which attorneys lodge challenges.

Accordingly—at least as long as demographics are the primary basis for peremptory challenges—an attorney’s success at peremptory challenges will depend on how ably she can use demographic characteristics to predict the potential juror’s suitability. This process depends, in turn, on the alignment between demographic characteristics and jury voting patterns. If jurors’ demographic features correlate strongly with their likely votes, the lawyer’s job will be easier. If they do not, her peremptory challenges will add little value.

That match need not be symmetrical. Consider a case where all blue-eyed jurors are likely to side with the plaintiff but where brown-eyed voters will likely split evenly. In such a case, the peremptory challenge evidently advantages the defendant: the defendant’s attorney can lodge challenges in utter surety, while the plaintiff’s lawyer must use guesswork.

B. Stereotyping Liberals and Conservatives

Importantly, American demographics manifest this sort of asymmetry. Some group identities—including those based on race and gender, two of the features on which peremptory challenges are often based—are highly correlated with membership in the Democratic Party, a strong indicator of a liberal ideology. On the other hand, the groups that tend to support the Republican Party and display conservative ideologies do not have such strong identity-ideology correlations. This fact could be historically contingent, or a reflection of a real difference between ideological coalitions.49 In either event, it has ramifications for the jury-selection process, as it advantages litigants whose ideal juror is conservative while harming those who seek a more liberal jury.

To measure this effect, this Comment calculates the likelihood that an attorney would profile a juror as conservative or liberal, based on the intersection between the jurors’ race and gender. It bases its analysis on 2012 data from the American National Election Study, which asks a nationwide sample of respondents a battery of political questions every two years.50 This Comment uses the survey’s data on partisan affiliation as a proxy for ideology, rather than looking at data on self-described ideology itself, for three reasons. First, there is a strong correlation between partisan identity and ideology: a Pew Research Center study found that “92% of Republicans are to the right of the median Democrat, and 94% of Democrats are to the left of the median Republican.”51 Second, self-identification of ideology is notoriously unstable, incoherent, and difficult to define,52 while partisan identification is easy to understand and to code. Third, studies of other actors in the judicial system similarly use political party as an ideological proxy.53

Table 1 shows the results of this methodology. It presents party-affiliation data, as a proxy for ideology, by race and gender. It sorts these demographic groups by ideological skew: at the top, black women are overwhelmingly Democratic; at the bottom, white men are the most Republican-leaning demographic.

The Table shows a stark divide. Majority-liberal demographic groups are overwhelmingly Democratic, while majority-conservative groups are more mixed. Only about 2.5% of blacks and 17% of Hispanics affiliate with the Republican Party, while whites are significantly more balanced.

Table 1.

politics by demographic54


The implications for jury selection are clear. If a party is seeking to strike liberal jurors, demographic profiling makes it easy: a black juror is a sure strike, and Hispanic jurors are also overwhelmingly likely to be left-of-center. If a party strikes a random member of a majority-liberal group (essentially, a racial or ethnic minority), that party will eliminate a Democratic juror fifty-nine percent of the time, while its action will backfire by eliminating a Republican only ten percent of the time.56 And of course, if the party focuses only on random black prospective jurors—precisely the people Batson is designed to protect—it will succeed even more often.

By contrast, a party seeking to avoid conservative jurors has a much harder task: if it picks a member of a majority-Republican demographic at random, it will succeed only thirty-four percent of the time and will eliminate a Democratic juror on a full twenty-six percent of strikes. This disparity can be illustrated in terms of expected value: an attempted strike of a liberal will remove 0.49 more Democratic jurors than Republican jurors from the panel; an attempted challenge to a conservative prospective juror will remove only 0.08 more Republican jurors than Democrats.57 In other words, peremptory challenges targeting liberals are six times more effective, based on no data other than raw demographics, than strikes targeting conservatives.58

C. Peremptory Challenges’ Right-Leaning Skew

This result distorts juries’ ideologies. A simulation of ten thousand juries suggests that, if parties in a case pursue an ideologically driven peremptory-challenge strategy based on demographic stereotyping, peremptory challenges dramatically decrease the number of liberal jurors and increase the number of conservative ones.59

Without peremptory challenges,60 a twelve-person jury mirroring nationwide demographics as measured using ANES data would contain 4.6 Democrats, 3.2 Republicans, and 4.2 members of neither party. (Traditionally, more Americans identify as Democrats than as Republicans.)61 In a federal civil case (each party receives three peremptory challenges)62 in which parties pursue an ideologically driven peremptory-challenge strategy based solely on demographic stereotyping, the resulting jury would contain 4.2 Democrats, 3.4 Republicans, and 4.3 members of neither party. In the same circumstances in a California civil case (six peremptory challenges a side),63 the resulting jury would contain 3.9 Democrats, 3.7 Republicans, and 4.4 members of neither party. In the same circumstances in a federal capital case (twenty peremptory challenges a side),64 the resulting jury would contain 3.4 Democrats, 4.1 Republicans, and 4.6 members of neither party. Table 2 reports these results.

Those respective scenarios have 10%, 15%, and 26% fewer Democrats and 6%, 16%, and 21% more Republicans than the no-challenge baseline. In all cases, peremptory challenges have the expected effect of striking jurors on the political extremes and increasing the representation of centrist jurors.65 However, the effect is ideologically uneven: while homogenizing the jury to an extent, peremptory challenges also skew it rightward.

Table 2.

simulation of ideological peremptory challenges (ten thousand simulations)


This stylized simulation tells a troubling story. If—as studies show66—lawyers use race- and gender-based stereotypes as predictors of jurors’ suitability, and if—as evidence suggests67—perceived suitability is closely linked to juror ideology, then the peremptory-challenge procedure skews jury outcomes. It advantages litigants whose success relies on a conservative-minded jury, while disadvantaging those who seek a liberal jury.

This problem is endemic to the current peremptory-challenge regime. It could be partly ameliorated, to be sure, by weakening attorneys’ reliance on demographics during voir dire, either by increasing attorney questioning68 (if attorneys seeking to strike conservative jurors could ask about gun ownership, for example, it would be easier for them to lodge ideological challenges)69 or by requiring jurors to disclose ideologically salient information on their questionnaires.70 But those ideas are not total solutions. In the election context, for example, political scientists have shown that voters perceive black candidates as more liberal than white candidates with identical policy positions,71 suggesting that even hard data cannot overcome demographics in determining perceptions of ideology. The connection between peremptory challenges and ideological bias is thus baked into the peremptory-challenge procedure.


This Comment shows how the use of peremptory challenges may skew juries ideologically to the right. The data presented add to the robust literature on the other ways in which unfettered peremptory challenges bias the trial process.72 They also strengthen the case for legislative action to limit or abolish the peremptory challenge. The peremptory challenge is not a constitutional right,73 but rather is codified in federal statute74 as well as in the laws of all fifty states.75 Yet, as debate has raged in courtrooms76 and in the media,77 legislatures have stayed silent. Only a few legislatures have considered proposals to curtail peremptory challenges;78 no state appears to have restricted its peremptory-challenge practice since Batson.

Legislatures are slow to act on racial and criminal justice issues.79 But they frequently alter procedural rules to advance their preferred ideological outcomes in the courts.80 Legislative reform of peremptory challenges may be more easily accomplished, therefore, by stressing the ideological bias of peremptory challenges and its ramifications in civil cases.81 Considering peremptory challenges as a problem of politics as well as a problem of racial justice may strengthen the case for self-interested legislative action.