Evidentiary Trapdoors

I.     Introduction

Trap (n.):  “A stratagem for catching or tricking an unwary person. . . . A confining or undesirable circumstance from which escape or relief is difficult . . . . A trapdoor.” 1

As “[t]he government’s star witness” in the federal trial of David Hinkson for solicitation of murder, Elven Joe Swisher proudly displayed his Purple Heart lapel pin on his military uniform as he testified about his military exploits in the Korean War.2  He then confidently testified that the defendant had solicited him to commit the murders because of Swisher’s copious special operations experience while he was deployed.3 When pressed by defense counsel on cross-examination about the authenticity of his military credentials, Swisher dramatically produced a document from his pocket corroborating the details of his extensive service.4 

There was, of course, one problem with Swisher’s testimony:  none of it was true. Military documents subpoenaed by the defense revealed that Swisher lied about his military service to the court and the jury.5 But when the defense attorney later attempted to impeach Swisher’s testimony with documents provided by the Department of Defense, the prosecutor objected based on a novel interpretation of Federal Rule of Evidence (“FRE”) 608(b), which disallows the use of extrinsic evidence to prove collateral matters—such as a witness’s character for truthfulness—at trial.6 The court sustained the objection, the jury never learned of Swisher’s falsehoods, and the defendant was convicted on the solicitation charge.7

Similarly, in a case decided 15 years earlier, agents from the Drug Enforcement Administration (“DEA”) formulated a complex sting operation to arrest former officers who allegedly attempted to steal narcotics from drug dealers and resell them for profit.8 During the sting, DEA agents arrested Gilbert Musquiz in a Houston-area T.G.I. Friday’s parking lot as he was attempting to enter a known drug dealer’s car, allegedly to take the cocaine from the car.9 During the interval between Musquiz’s arrest (which included a search of his person and the intervening escape of another suspect) and the time the DEA agents later read him his Miranda warnings at the DEA office,10 Musquiz did not attempt to explain his actions to his arresting officers.11

At trial, Musquiz claimed his arrest was the result of a misunderstanding, and he further claimed that he approached the vehicle in the parking lot that day to collect evidence pursuant to a reward seeking information regarding Columbian drug traffickers.12 When the prosecutor informed the trial court that he intended to ask Musquiz on cross-examination why he did not provide this explanation when he was arrested, his attorney objected in a motion in limine, on the basis that the question constituted improper impeachment because Musquiz had invoked his Fifth Amendment right against self-incrimination.13 In response to the prosecutor’s argument that Musquiz’s right to remain silent attached only after Miranda warnings had been issued—and that recent case law interpreting Federal Rules of Evidence 607 and 613 treated a suspect’s silence as a tacit contradiction of his trial testimony—the court denied defense counsel’s motion in limine and allowed the questions.14 The defendant was convicted, and the conviction was upheld on appeal.15

Contrary to the intuitions of laypeople and law students alike—and despite being governed by nearly 70 rules across eleven different Articles of the Rules of Evidence16—the law of evidence is a landmine of textual imprecision, creative interpretation, and fervent disagreement among policymakers and among members of the judiciary. The foregoing vignette introduces the concept of the evidentiary trapdoor, a pervasive phenomenon within the Federal Rules of Evidence that arises out of this landmine. The evidentiary trapdoor has three primary components: (1) an application of a federal rule that is often contradicted by the rule’s plain language or taken advantage of due to an unintended ambiguity within the rule; (2) a court’s admission of evidence based on empirically untested beliefs that the evidence will heighten the fact finder’s ability to reach an accurate verdict; and (3) a novel application of the Federal Rule which creates unexpected harm to the proponent’s adversary, who is left unprepared.17

To the extent that legal scholars have questioned whether the admission of such evidence represents sound evidentiary policy, rule makers typically respond in two ways. First, they argue that if a novel interpretation of an evidentiary rule does not explicitly violate the rule, creative argumentation is encouraged in an adversarial trial setting.18 Second, they argue that the system should encourage the use of trapdoor evidence when the evidence benefits the legal tribunal by increasing the court’s ability to unearth the accurate facts of the dispute.19 Therefore, policymakers have left it to the individual advocate’s discretion in relying upon applications of evidentiary rules, which some critics deem to be inappropriate gamesmanship in representing a client.

This Article advocates for expanding the academic conversation regarding trapdoor evidence and, specifically, thinking more critically about its admissibility. The current debate concerning trapdoor evidence focuses too myopically on doubtful empirical claims about the effect of trapdoor evidence on courts’ ability to reach accurate verdicts. This focus is unsatisfying for two reasons. First, it provides no basis for the empirical claim that trapdoor evidence increases a legal tribunal’s decisional accuracy, or even that it increases public perceptions of the tribunal’s accuracy. Second, it incorrectly assumes that increases in a tribunal’s decisional accuracy will lead to an increase in the public’s willingness to legitimize the tribunal, even if the tribunal derives its increased accuracy through procedures that the public deems unfair. Insights from psychological legitimacy theory and empirical literature on moral decision making suggest that governmental actors do not derive legitimacy from their ability to make accurate decisions alone. Instead, a more powerful predictor of whether the public will legitimize a legal tribunal is the ability of the court to produce decisions that result from fair procedures, regardless of whether those procedures always produce substantively accurate outcomes.20 Therefore, empirical evidence shows that, in most cases, trapdoor evidence does not increase the courts’ decisional accuracy, that laypeople deem it as unfair “gamesmanship,” and that it causes the public to delegitimize legal institutions that admit this type of evidence.21

This Article is the first to test these assertions using data from three original experiments. These experiments affirm that, consistent with the psychological literature on legitimacy and moral decision making, the admission of trapdoor evidence lowers the public’s perception of a legal tribunal’s legitimacy across several types of courtroom evidence, legal actors, and different types of trials.

This Article proceeds in several Parts. Part II outlines the law of evidence as it relates to the evidentiary trapdoor. Part III outlines the behavioral literature on legitimacy, gamesmanship, and moral decision making. Part IV reports the results from three original laboratory experiments, in which participants read a reenactment of several trials that involved the admission or exclusion of trapdoor evidence. Part V explores the implications of these findings, their limitations, and future directions.

II.     The Law of the Trapdoor

Part II of this Article defines the concept of the evidentiary trapdoor and briefly describes the context in which federal courts have allowed evidentiary trapdoors to flourish. It then provides examples of the phenomenon and justifications for its existence.

A.     Evidentiary Context

The law of evidence governs the manner in which advocates prove facts—and the inferences to be drawn from those facts—in civil and criminal trials.22 For most of this country’s development, the rules governing the flow of information to fact finders were a product of the common law.23 Until the mid-20th century, the rules that governed proof at trial were determined and modified “in fits and starts,” as individual cases worked their way through state and federal appellate courts.24 Although a select few influential state legislatures attempted to codify their state’s common law of evidence in the early 20th century,25 the federal government, led by Chief Justice Earl Warren, largely led the way toward a more uniform code in 1965, with the formation of an Advisory Committee to create the first draft of the Federal Rules of Evidence.26

Historically, the common law of evidence reflected judicial skepticism toward—and a marked distrust of—lay fact finders.27 The common law featured strict competency requirements for testifying witnesses,28 heightened standards for proving the authenticity of documents,29 a strict and complex bar against the use of second-hand evidence at trial,30 byzantine rules regarding the mechanics of impeachment and which witnesses an advocate could impeach,31 and confusing rules determining the conditions under which fact finders could rely on the opinions of experts.32 With this background, the codified Federal Rules of Evidence were designed with twin goals: (1) to reduce the procedural costs of litigation that existed under the common law; and (2) to increase the accuracy of decisions rendered by lay fact finders while providing procedural protections for the litigants.33 In attempting to achieve these goals, rule makers specifically sought to relax several features of the common law of evidence, including the rules governing competency, impeachment, and hearsay.34

Despite policymakers’ attempts to simplify the evidentiary rules, many of the complications from the common law—including the hearsay rule and its exceptions, the rules regarding privilege, and the rules governing the process of impeachment—remained under the codified Federal Rules of Evidence, while others—such as a rule purporting to ban the use of propensity evidence—became even more complex.35 The end result was a new federal evidentiary code that failed to improve problematic doctrines and introduced new challenges for litigants.36 It is perhaps unsurprising, then, that nearly every rule within the new federal evidence code required substantial clarification from the courts.37 Perhaps surprising, however, is the degree to which the resulting case law diverged from the spirit and plain language in nearly every major area of the evidentiary code.38 The deviations became so dramatic and widespread that the Advisory Committee to the Federal Rules of Evidence expressed serious concern over the extent and breadth of this divergence. The Committee explicitly warned the legal community of “red flags” in the plain reading of the Federal Rules that “might create a trap for the unwary.”39 In an attempt to dispel those red flags in the early 2000s, the Advisory Committee took the extraordinary step of directing its “Reporter to prepare [a] report in an effort to increase the awareness of counsel practicing in federal courts, as well as judges, about the possibilit[ies]” of such divergences in the Federal Rules.40

 B.     Trapdoor Types & Justifications

The Advisory Committee’s Report, however, did not attempt to locate every single judicial deviation from the text of the Federal Rules. These deviations can instead be categorized as two complementary trapdoors for advocates: (1) case law that flatly contradicts the text of a Federal Rule; and (2) case law that provides a non-obvious, significant development of the Rule’s text that the Advisory Committee did not contemplate.41

The most insidious evidentiary trapdoor involves case law that directly conflicts with the plain language of an evidentiary rule. A comprehensive list of these conflicts could fill an entire volume of a law review, but they can be located generally within the rules governing character evidence, impeachment procedures, scientific evidence, and hearsay.42 For example, FRE 407 prohibits the use of “subsequent remedial measures”—actions a defendant takes after an injury occurred that, a priori, would have made the injury less likely to occur—to prove a defendant’s culpability.43 The plain language of the Rule suggests that FRE 407 bars these actions by anyone that could have ameliorated the harm to the plaintiff.44 However, many courts have held that FRE 407 does not exclude such measures if they were undertaken by third parties45 or, more controversially, if the defendant undertook the measures at the behest of a third party.46 Courts that have deviated from the Rule’s plain language argue that the Rule’s traditional justification—encouraging defendants to undertake, on their own, post-injury repairs without the fear of creating “bad evidence”—does not apply in these circumstances47 and therefore suggest, implicitly in their decision to admit the evidence, that informing the jury of the defendant’s post-injury repair provides the jury with information that could increase the jury’s likelihood of reaching an accurate verdict.48

The courts employ similar reasoning in their treatment of FRE 704(b). FRE 704(b) forbids experts from testifying as to whether a defendant possesses the requisite mental state to commit a crime.49 Although the Rule states that no expert witnesses may do so,50 many courts have held that FRE 704(b) applies only to mental health experts.51 These courts state that the Rule does not apply to law enforcement experts who testify regarding a criminal defendant’s intent to sell narcotics, even though the Rule’s text makes no such distinction.52 Other courts have held that all expert witnesses in criminal trials are bound by FRE 704(b), however, they have similarly allowed law enforcement experts to sidestep FRE 704(b)’s prohibition by answering hypothetical questions about the defendant’s intent to distribute narcotics with factual scenarios that are identical to the facts of the matter being litigated.53 By deviating from the plain language of FRE 704(b) so plainly, these courts are implicitly ruling that an experienced law enforcement officer’s opinion regarding a defendant’s intent to distribute drugs is potentially more reliable than a psychiatrist’s determination of a defendant’s mental state; admitting the officer’s testimony therefore is valuable evidence that could increase the odds that a jury will reach an accurate verdict.54 

Evidentiary trapdoors take another form as well: an application of the rule that does not directly conflict with the its plain language, but provides a non-obvious development of the rule that the Advisory Committee did not intend. FRE 803(3), the state of mind exception to the hearsay rule, provides an example of such an application. FRE 803(3) allows fact finders to consider then-current statements of a declarant’s intent to commit a future act, however, the Rule is silent regarding whether an advocate can use these statements also to prove the acts of a third party.55 Courts have split in this situation, but many courts have held that, for the purpose of providing the fact finder with a complete picture of the underlying events at issue, statements made pursuant to FRE 803(3) can prove not only the future actions of the person who uttered those statements, but also the actions of third parties implicated by those statements.56

A final example comes from FRE 410, which purportedly disallows the use of statements that defendants make in plea bargaining negotiations as evidence against them if they withdraw their plea and elect to go to trial.57 The Rule is designed to allow open communication between the government and a criminal defendant during the plea bargaining stage of the pre-trial process, but the Rule is silent regarding under what conditions, if any, a defendant may waive the protections afforded under FRE 410.58 Although courts have declared that similar protections afforded under other Federal Rules are not waivable,59 many courts have held that in the context of FRE 410, a defendant is held to her decision to waive the Rule’s protections.60 The foregoing examples are representative of the continuing uncertainties and ambiguities in the application of the Federal Rules of Evidence.

The Federal Rules of Evidence were created to reduce the procedural uncertainty that existed under the evidentiary common law.61 However, in many respects, the Rules failed to reduce uncertainty and created additional complexity. In fashioning an evidentiary code with language that was both too specific and too general,62 the Advisory Committee unwittingly created an environment for evidentiary trapdoors to flourish. Surprisingly, and despite the Advisory Committee’s recent warnings, legal scholars largely defend the presence of these trapdoors on the theory that they achieve an important goal of the legal system: aiding the fact finder in discovering the truth underlying the legal dispute.

The empirically unproven starting assumption that the Rules reduce complexity and aid in fact finding has caused many trial scholars to take one of two favorable positions toward evidentiary trapdoors. Some scholars approach the issue from the perspective of libertarian ethics.63 They argue that, because the use of evidentiary trapdoors is not expressly forbidden within the Federal Rules of Evidence, an advocate’s decision to avail herself of trapdoor evidence should be respected.64 Other scholars advocate for a stronger, deontological position. They argue that an advocate has a moral and ethical duty to avail herself of every strategic advantage within the evidentiary and procedural rules to ensure a victory for her client.65 Those who subscribe to the deontological view characterize an advocate’s failure to avail herself of trapdoor evidence as a moral and ethical shortcoming eligible for sanction.66

The debate regarding an advocate’s use of trapdoor evidence should be expanded for two related reasons. First, the debate assumes that trapdoor evidence produces greater decisional accuracy in the courts, and is therefore a net benefit to the legal system; a contention unsupported by the empirical evidence. Second, the discussion fails to account for the psychology that underlies laypeople’s perceptions of moral decision making, which strongly suggests that laypeople do not view these strategic decisions the same way that legal scholars do. Rather, laypeople perceive an advocate’s use of trapdoor evidence as at best unfair gamesmanship, and at worst morally unethical. Moreover, the psychological literature on legitimacy theory suggests that people closely associate violations of morality with violations of fair process, which causes them to delegitimize not only the transgressing actor, but also the entity that allows the actor to transgress.67 The remainder of this Article explains these psychological doctrines, provides empirical evidence of their validity, and discusses the implications for an expanded debate regarding the use and availability of trapdoor evidence.

III.     The Psychology of the Trapdoor

This Article argues that, instead of increasing the likelihood of accurate verdicts, the use of trapdoor evidence lowers the public’s perceptions of the fairness of trials, which manifests in the public’s decreased willingness to legitimize legal tribunals that rely on trapdoor evidence. This Part examines the relevant psychology literature that is later used in support of three original experiments that test this hypothesis empirically. It proceeds in two Parts. First, it defines the concept of psychological legitimacy, explains its relationship to decision-making accuracy and procedural fairness, and describes the conditions under which the public legitimizes and delegitimizes legal actors. Second, it draws on the moral decision-making literature, as it relates to the psychology of cheating behavior and gamesmanship, to explain the mechanism by which the public delegitimizes trials that contain trapdoor evidence.

A.     Psychological Legitimacy

Legitimacy is an amorphous concept with implications for all aspects of our daily life, including the manner in which we are governed.68 Legitimacy as a political theory can be traced as far back as the Enlightenment, when moral philosopher John Locke opined that “the government is not legitimate unless it is carried on with the consent of the governed.”69 

Building on Locke’s famous pronouncement, political theorists describe the concept of legitimacy as the status and acceptance that governed people confer onto their governors’ institutions and conduct based on the belief that those actions constitute an appropriate use of power.70 According to German sociologist Max Weber, the governed confer legitimacy onto legal actors via an alignment of values between the political actors—that is, through public trust that the government will act in the interests of the governed—and not through the government’s coercion or force.71 Therefore, to the extent that a misalignment develops between the values of the governed and the actions of the government, political legitimacy is endangered.72

Numerous interdisciplinary scholars have attempted to explain the theories that underlie people’s willingness to legitimize governmental action. Broadly speaking, these theories fall into two camps. The first camp is often referred to as either “output,” “instrumental,” or “goal-attainment” legitimacy. This theory posits that legitimacy is derived almost entirely from substantive outcomes either for society at large or, more specifically, for the individuals affected by the governmental action.73 Thus, under this theory, legitimacy is a function of social exchange, insofar as exchanges and interactions with governmental actors resulting in a positive distribution of goods to the governed create a greater willingness among the governed to legitimize the governmental action.74

In contrast, a second theory of psychological legitimacy is referred to as “substantive” or “relational” legitimacy. In contrast to the instrumental, goal-oriented model, this model posits a relational, equity-based manner in which governmental actors attain popular legitimacy.75 The theory suggests that a government attains legitimacy through its procedural responsiveness to the concerns of its citizens by allowing them to meaningfully participate in the governmental process. Legal psychologist Tom R. Tyler’s group value model provides empirical support for this theory of legitimacy.76 As Professor Tyler and I have written elsewhere:

The relational [group value] model [of legitimacy] argues that people value the [governmental actor’s] use of fair procedures because those procedures carry messages of status and inclusion which reinforce people’s identification with legal institutions and authorities and support their feelings of inclusion and status in the community. This then leads to high self-worth and favorable self-esteem. When people can present their concerns to judicial authorities and feel that those authorities consider and take account of their concerns, people’s identification with law and legal authorities is strengthened.77

These theories predict different outcomes when a governmental actor uses what the governed perceive to be fair procedures that reach an unfavorable substantive outcome. The instrumental, goal-attainment theory predicts that governmental actions that lead to substantive unfairness vis-à-vis the governed will result in the potential loss of the actor’s popular legitimacy.78 The relational, social value model predicts, however, that citizens will tolerate some substantively unfair outcomes while continuing to recognize the legitimacy of the governmental actor, provided that the process by which the governmental actor made her decision is considered fair and just.79 Although substantive outcomes are not irrelevant, a wealth of social psychology evidence on the phenomenon of “procedural justice” supports the relational model of governmental legitimacy.80

Scholars recently have begun to apply psychological theories of legitimacy specifically to the public’s perceptions of legal regimes.  In the United States, most of the scholarship to date has focused on the legitimacy of the Supreme Court.81 Legal psychologists, however, have begun investigating the factors that cause people to perceive county, state, and federal district courts as worthy of legitimacy. For example, Professors John Thibaut and Laurens Walker pioneered this movement by hypothesizing that people perceive the courts as having two distinct but related goals: (1) to get to the truth of a legal matter (that is, to correctly find the facts that underlie the dispute), and (2) to do so in a manner that the public deems to be fair and just.82  Recent empirical work suggests that these goals can be mapped onto a Cartesian plane that signifies the degree of legitimacy that is concomitant with the level of decisional accuracy and procedural fairness that the public perceives a legal tribunal to provide:83

103 3 Sevier Fig. Decisional Accuracy

Empirical studies confirm Thibaut and Walker’s hypothesis, finding, perhaps counterintuitively, that people often perceive a reciprocal relationship between a legal tribunal’s ability to get to the truth of a dispute (i.e., the tribunal’s “decisional accuracy,”) and the fairness of the proceedings that the tribunal uses to produce its decision (i.e., the tribunal’s “procedural justice”).84

Recently, Professor Tom R. Tyler and I extended these findings by examining which factor has the greatest influence on legitimacy of trial court decisions in the eyes of the governed—the perceived accuracy of the legal tribunal or the procedural justice that it provides.85 We hypothesized that, although a court’s decisional accuracy would be relevant to public perceptions of the courts’ legitimacy, the perceived fairness of the proceedings would have a significantly greater impact.86 In forming this hypothesis, we relied on Tyler’s seminal work on legal legitimacy, Why People Obey the Law. In that work, Tyler explained that the adversarial system is particularly strong at providing relational benefits to litigants by allowing them to have a voice in the proceedings, to select their own counsel and call their own witnesses, and to cross-examine those who accuse them of wrongdoing.87 We hypothesized that those relational benefits would be more salient to the public than the accuracy benefits that the adversary system may also provide.88

Our previous study revealed that procedural justice was in fact a far greater predictor of people’s willingness to legitimize the courts than the perceived decisional accuracy that the courts provide.89 This suggests that psychological, relational signals that are produced from the court’s willingness to listen to its litigants and provide them a fair hearing are a greater predictor of the public’s willingness to legitimize the courts than is the court’s decisional accuracy, either as an independent predictor of perceptions of legitimacy or as a mediator of the effect of procedural justice.90 Decisional accuracy was not irrelevant, however. The ability of the court to get at the truth of a legal proceeding had an independent impact—albeit a weaker one—on perceptions of legitimacy, and it also served as a link between people’s perceptions of the fairness and legitimacy.91 Specifically, people believed that fair procedures beget greater decisional accuracy, which leads people to legitimize the tribunal.92 Nonetheless, we found a significantly stronger effect of a direct path between people’s perceptions of the trial’s procedural justice (that is, its fairness), and its legitimacy. 

We see this philosophical tug-of-war between the goals of decisional accuracy and procedural justice within the law of evidence as well. Many legal scholars posit that the Federal Rules of Evidence exist primarily to balance the competing concerns of producing accurate verdicts while ensuring that the process used to obtain those verdicts is fair and just.93 Precisely because the Federal Rules of Evidence serve these dueling concerns, the Rules are littered with opportunities for litigants to avail themselves of non-intuitive applications of the Rules—ostensibly for the purpose of improving the factfinders’ decisional accuracy—in a manner that damages the public’s perception of the fairness of the proceedings. If the public’s perceptions of the legal tribunal’s legitimacy are tethered more closely to the perceived fairness of the tribunal’s procedures than to the tribunal’s perceived substantive accuracy, such accuracy enhancing uses of the Federal Rules will decrease the public’s sense of the tribunal’s legitimacy. To examine the psychological mechanisms behind this phenomenon, this Article next reports research on moral decision making in the context of perceived gamesmanship.

B.     The Moral Psychology of Gamesmanship

Moral psychology is a burgeoning academic field with myriad implications for the law. As I have described it elsewhere: 

Moral psychology is an interdisciplinary field that blends social psychology and philosophy to determine the evolutionary origins, personality traits, and social cues that predict how laypeople reason about fairness and justice. Although early scholars—including Aristotle, Plato, and Socrates—interpreted the field to refer to moral development over the human lifespan, modern scholars interpret the field to include an array of topics, including the philosophy of the mind, the creation and development of ethical codes, the notion of free will, moral sensitivity, moral action, moral identity, emotional forecasting (as it relates to appraisals of situational fairness), and, most importantly, moral judgment.

The most striking feature of morality research involves the counterintuitive nature of moral judgment. The [relevant research] strongly suggests that such assessments are not based on deliberative reflection at all; rather they are the result of intuitive, instantaneous, emotional reactions to social stimuli. Moreover, although this model of moral decision-making concedes that these initial moral appraisals—at least to some degree—can be altered through deliberation with others, laypeople often do not realize that their moral judgments are the result of subconscious, affective initial appraisals, which are then rationalized through self-serving, post-hoc justifications.94

In light of this research, it is perhaps unsurprising that our moral values influence our evaluative judgments toward the appropriateness of gamesmanship behavior, which the psychological literature on cooperation defines as non-intuitive (albeit allowable) strategies for gaining an advantage over a competitor in a contest for physical or psychological resources.95 Somewhat surprisingly, however, the research on gamesmanship is still in its infancy, and only a handful of studies have been conducted examining its effects. There is, however, a wealth of literature on cheating behavior. Cheating behavior is substantially different from gamesmanship in several important ways,96 but the two phenomena share similar psychological features,97 such that both are relevant to the discussion of psychological legitimacy. This Article therefore first examines the literature on cheating behavior and then compares and contrasts it with the (smaller) literature on gamesmanship.

Cheating behavior encompasses a conscious disregard not just for the letter of a social rule that the cheater breaks, but also for its underlying spirit.98 The empirical literature suggests that, in a variety of social contexts ranging from paying one’s taxes,99 to an individual’s academic studies,100 to sportsmanship,101 and in romantic relationships,102 people view the act of cheating as a transgression of agreed-upon social norms worthy of punishment.103 People perceive the nature of the transgression as not only the receipt of a substantively unfair advantage, but also as the receipt of a competitive advantage in a procedurally unfair manner.104

Psychologists believe that the moral opprobrium that people impose upon cheating behavior has evolutionary roots.105 A recent study argued that punishing those who skirt social rules serves a crucial cohesiveness function for our species’ evolutionary survival.106 The authors conceive of cheating behavior as a free rider problem, in which an individual takes advantage of the generosity of others, contributes nothing to the social environment in return, and enjoys more gains in evolutionary fitness than those who do not cheat.107 This is due to the fact that cheaters do not bear the costs of cooperation that non-cheaters bear.108 Punishing cheaters therefore has dual evolutionary benefits: (1) it promotes cooperative behavior, which benefits the species over time; and (2) it reduces or eliminates the fitness advantage that unpunished cheaters possess over time.109

Although moral opprobrium toward cheating behavior may be evolutionary in nature, it is not universal. Attitudes toward cheating often are context-specific even among those who believe that cheating should be swiftly and severely punished. Psychology research suggests that some aspects of an individual’s level of societal power can influence not only her willingness to cheat, but also her tolerance for the cheating of others. For example, researchers examined the attitudes of wealthy and non-wealthy individuals and found that socially powerful individuals tend to engage in rule breaking at a greater rate than less powerful people do, while simultaneously being more punitive toward rules transgressions by others.110

Other mediating factors, such as the degree of sympathy for the cheater and the person who suffers the harm,111 and the social context in which the cheating occurs112 affect our tolerance for such behavior. For example, cheating that is perceived to be for a greater good, against an unfair, and often powerful, individual or entity is considered a minor offense.113 With minor stakes for the involved parties, people’s moral tolerance for cheating thereby increased.114

Against that background, the psychological literature on attitudes toward gamesmanship is far less developed. Recall that cheating behavior involves two types of violations: (1) a violation of the letter of a social rule; and (2) a violation of the rule’s spirit.  Gamesmanship, by contrast, involves only a violation of the spirit of a social rule; it explicitly does not involve a violation of its letter.115 It is a more open question, therefore, whether the extent and degree of moral opprobrium that people levy against cheating behavior is also levied against gaming behavior, because it is unclear that people perceive gamesmanship as a moral transgression in the same way. 

Asked another way, is cheating immoral because of the violation of the letter of the social rule, or does the moral opprobrium attach also because of the violation of the rule’s spirit?  If people perceive only the former to be immoral, does that mean that they believe that others may avail themselves of loopholes in social rules to gain advantage over competitors so long as the letter of the rule is not violated? Psychology research suggests that, all else equal, people tend to have negative attitudes toward a social actor’s use of loopholes, particularly in legal settings.116 Research also suggests that many of the factors that lead people to punish cheating behavior also lead them to disavow the use of gamesmanship, including, for example, the harm caused to the individual who is the object of the gamesmanship and to society at large.117 However, gamesmanship research in other contexts, such as sports psychology, reveals complex attitudes toward gamesmanship. Although people disapprove of cheating in sports, they are much more ambivalent toward gamesmanship. For example, a recent study of gaming and cheating in youth sports found that children’s moral tolerance for gamesmanship was heavily influenced by the attitude of individual participant’s coach toward gamesmanship.118 

Put another way, the extent to which people accept gamesmanship depends on the context in which the gamesmanship occurs, for example, in the sports arena119 versus in the courtroom, and the reinforcement structures in place, for example, the attitude toward gamesmanship exhibited by a superior who is in a position to reward or punish the individual actor, such as a sports coach, a senior partner, or a district attorney.120

Recall that moral opprobrium toward cheating behavior does not result merely because a cheater receives an advantage compared to her competitor.121 The disapproval results, in part, because people deem the process by which the advantage was procured unfair.122 People might perceive that exploiting loopholes and exceptions in the governing rules, although not technically cheating, still violates notions of fair process. If so, then in certain legal contexts, gamesmanship may be viewed as akin to cheating vis-à-vis people’s perceptions of procedural justice.123

It is this phenomenon—popular perceptions of procedural justice—that bridges people’s moral attitudes toward gaming behavior with their willingness to legitimize (1) the actor who engages in gamesmanship and
(2) the social setting that allows the actor to do so. These findings have important implications for the rules of evidence and the perceived legitimacy of the legal system in which those rules operate. How laypeople view an attorneys’ use of trapdoors within the Federal Rules of Evidence, which are designed to enhance the legal tribunal’s decisional accuracy, is an empirically open question. On the one hand, even if the use of trapdoor evidence constitutes gamesmanship, the psychological literature suggests that gamesmanship is not always viewed as morally equivalent to cheating and, to the extent that the use of trapdoor evidence carries a moral stigma, the empirical literature indicates that people might be less punitive if the trapdoor evidence is perceived to be used for “a greater good,” such as decisional accuracy. On the other hand, to the extent that gamesmanship and cheating behavior both violate public perceptions of procedural fairness, the literature suggests that this will negatively impact people’s willingness to legitimize trials that allow trapdoor evidence. The original experiments that follow examine these competing hypotheses in determining the relationship between the use of trapdoor evidence and the public’s willingness to legitimize legal tribunals.

IV.     Three Experiments

This Article now reports the results from three original experiments. These experiments were designed to test several experimental questions, including: (1) whether the use of accuracy-enhancing loopholes in the evidentiary rules enhances public perceptions of the trial’s accuracy and whether the use of the evidence causes people to see the trial as more—or less—legitimate; and (2) to the extent that evidentiary trapdoors have little or no effect on perceptions of accuracy and are harmful to the public’s perception of the trial’s legitimacy, how robust are those findings? Specifically, are they true regardless of the type of trial or the identity of the party that uses them?

A.     Study 1: Evidentiary Trapdoors

In the first study, participants read a vignette in which they imagined themselves as spectators at a murder trial at their local courthouse. The study manipulated two variables. First, at the conclusion of the trial, the prosecutor attempted to admit surprise “trapdoor” evidence that prejudiced the defendant. The judge then ruled the proposed evidence either admissible or inadmissible. Second, the study varied the form of the evidence the prosecutor sought to admit. The evidence, which was substantively similar in each experimental condition, took a form consistent with several major areas of evidence law: impeachment, prior bad acts, hearsay, and privilege. The experiment then examined whether the inclusion of trapdoor evidence increased or decreased participants’ perceptions of the trial’s accuracy, the trial’s fairness, and participants’ willingness to legitimize the verdict.

If the use of trapdoor evidence increases the accuracy of trials or at least a shared belief that trials are more accurate, several results in Study 1 would reflect support for that hypothesis. First, perceptions of decisional accuracy should increase when the judge admits the trapdoor evidence compared to when the evidence is ruled inadmissible. Second, to the extent that increases in perceptions of the court’s accuracy are tied to our beliefs regarding the court’s legitimacy, we would expect a similar increase in participants’ willingness to legitimize legal tribunals that admit trapdoor evidence. However, if, as psychological research suggests, people do not view trapdoor evidence as increasing the likelihood of accurate verdicts, and instead see the use of such evidence as violating norms of procedural fair play, then the use of trapdoor evidence should decrease the public’s willingness to legitimize the legal tribunal. The following Part reports the methodology and results of Study 1.

1.     Participants

We124 recruited 244 participants for this online study through the Amazon Mechanical Turk (“MTurk”) recruitment service.125 Once recruited, participants received a link to the study, which was hosted on the Qualtrics online survey platform. Participants were paid $2.00 for their participation, and they were told that the study was designed to measure their attitudes about a hypothetical legal case. Nearly all of the participants completed the study within 20 minutes.

Participants represented a cross-section of jury-eligible citizens. The average participant was 36.27 years old (with a standard deviation of 10.26).126 The sample was split roughly evenly by gender, with women composing 41.80% of the sample.127 The sample generally reflected the racial diversity of the United States population as well, with 28.70% of the sample identifying as non-white.128 Roughly 58.60% of participants had completed at least a college degree, and the median participant income was between $40,000 and $49,999.129 The political affiliation of participants varied across the political spectrum, although the majority of participants identified as moderate (31.30%) to liberal (30.50%).130 Table 1 provides descriptive statistics for the participants involved in this study.

Table 1: Participant Demographics (Study 1) 

103 3 Sevier Table 1

2.     Procedure and Measures

After giving their informed consent, participants read materials that asked them to imagine themselves in the gallery while attending a trial at their local courthouse. They were told to imagine the judge on a raised platform in the distance directly in front of them, and to imagine the jurors seated in a box-like structure to the participants’ right. Participants were told that they would be witnessing a criminal trial. They were asked to imagine the prosecutor seated toward their right, near the jury, and to imagine the defense counsel and the defendant seated at a table to their left. Once the judge called the trial to order, the attorneys gave their opening statements to the jury.

Participants learned from the attorneys’ opening statements that they were witnessing a second-degree murder trial. The prosecutor stated that the evidence would show that the defendant murdered the victim in the early morning hours in the parking lot of an upscale mall. The prosecutor suggested that the evidence would show that the victim died during a botched cocaine sale. The defense’s statement focused on the circumstantial nature of the evidence and asserted that the prosecutor would produce no compelling evidence to support the claim that the murder occurred in the context of a cocaine transaction.

The prosecutor then presented the case against the defendant. The prosecutor presented three witnesses. The prosecutor first called the police officer who responded to the scene. The officer identified the victim and testified that the victim had been shot before 7:00 a.m. The officer testified that he observed at the scene an unregistered 45-caliber handgun that appeared to have been recently fired. He also observed a hat bearing the logo of the local sports team, which did not appear to be owned by the victim, as well as a small bag of cocaine in the victim’s jacket pocket. The mall’s security footage did not provide a clear image of the perpetrator, he testified, but the footage showed the perpetrator speeding away from the scene in a silver or gray sedan. The officer concluded his testimony by stating that he arrested the defendant for the crime later that day, after a swift investigation.

The prosecutor next called a forensic expert to the witness stand. The expert first testified that the bullets in the chamber of the handgun that the officer found at the scene were consistent with the bullet found in the victim’s abdomen. The expert next testified to the results of scientific tests that his lab conducted. He testified that the defendant’s hands had tested positive for the presence of gunpowder residue when he was arrested. The expert stated that the test has a negligible error rate and that the test is commonly used in criminal investigations.

Finally, the prosecutor called the defendant’s co-worker to the witness stand. The co-worker described the defendant as a secretive person who enjoyed hunting and shooting guns, which he owned in abundance. He also testified that the defendant is a diehard fanatic of the local sports team, and that the defendant owns memorabilia and apparel that bears the local team’s logo. On cross-examination, however, he could not be sure that the hat found at the crime scene belonged to the defendant. Finally, he testified that the defendant drives a silver Acura sedan.

The prosecutor rested her case, and the defense attorney called the defendant. The defendant admitted that he owns a considerable amount of sports memorabilia, but he denied that he owned the hat that was admitted into evidence. He also admitted that he is an avid hunter who owns many types of guns, although he claimed that all of his guns are lawfully registered. He testified further that he had been on a hunting trip on the day of the murder, and that he had shot a gun during that trip. On cross-examination, he stated that he went on the trip alone, and that he had no record of it beyond his own testimony. He also confirmed that he drives a silver Acura sedan.

Participants were told that the trial adjourned for the day and that closing arguments would begin the next morning. Participants were told however, that the prosecutor made a surprise request to the judge that morning. The prosecutor moved to proffer additional testimony to the jury based on recently discovered information about the defendant. The nature of the motion constituted the study’s experimental manipulation.

The prosecutor’s request consisted of an “evidentiary trapdoor.” Within that context, we presented two different manipulations to participants simultaneously. First, we manipulated the form of the evidence. Second, we manipulated whether the judge admitted the evidence, regardless of its form. Participants were randomly presented with a request for new evidence in one of the following four forms: (1) impeachment of the defendant’s character for truthfulness; (2) hearsay testimony that implicated the defendant;
(3) propensity evidence that implicated the defendant; and (4) admissions made by the defendant in a potentially privileged conversation.131  Regardless of its form, the evidence was substantively the same insofar as it suggested that the defendant was selling cocaine.

A quarter of the participants learned that the prosecutor wished to call the defendant to the stand to impeach him. Outside the presence of the jury, the prosecutor argued to the judge that she had just discovered that the defendant had been arrested twelve years ago for lying to law enforcement in an unrelated cocaine investigation, although there was no record of any subsequent legal proceeding stemming from the arrest. The prosecutor wished to question the defendant about the arrest.

A quarter of the participants learned, instead, that the prosecutor wished to call the defendant’s brother-in-law to the witness stand. The prosecutor told the judge that the defendant’s brother-in-law would testify to a conversation he overheard between the defendant and the defendant’s sister. Specifically, the brother-in-law overheard the defendant’s sister excitedly exclaim, “What do you mean you’ve been selling drugs since you lost your job?  What are you thinking?!”

Another quarter of the participants instead learned that the prosecutor wished to question the defendant regarding a recent arrest for cocaine possession with intent to distribute. The arrest apparently did not lead to further legal proceedings. Nonetheless, the prosecutor argued that the past arrest for possession with intent to distribute cocaine provides critical evidence that the defendant intended, on this occasion, to sell cocaine to the victim.   

Finally, a quarter of the participants learned that the prosecutor wished to call the defendant’s drug counselor to the witness stand. The drug counselor would testify that the defendant met with him weekly at his office. At one of these meetings, the defendant told him that he had lost his job recently, and to make ends meet, he had begun selling cocaine to buyers in the surrounding neighborhood “for some quick cash.” 

We then manipulated whether the trapdoor evidence was admitted successfully by varying whether the judge granted or denied the prosecutor’s motion. In each experimental condition, participants read a transcript of a short oral argument in front of the presiding judge, in which the prosecutor and defense attorney presented the arguments for and against admitting the additional evidence.132 

After the defense and prosecutor presented their arguments, the judge issued her ruling. For each participant, the judge either ruled that the evidence was admissible or the judge excluded the evidence. If the judge admitted the evidence, the judge acknowledged the “gray area” within the evidentiary rule that was implicated by the prosecutor’s motion, but noted that the rule did not explicitly call for the exclusion of the evidence. If the judge excluded the evidence, the judge also acknowledged the “gray area,” but noted that, in his view, admitting the evidence would violate the spirit of the rule and its underlying policies.

Once the judge made his ruling, all participants read the parties’ closing arguments and read the instructions that were presented to the jury. Participants then answered several questions regarding everything they had observed. The questions covered three general topics: (1) their impressions of the decisional accuracy, procedural fairness, and legitimacy of the trial;
(2) their impressions of the evidence, the case against the defendant, and the verdict; and (3) several demographic and personality-related topics. After participants answered these questions, we thanked them for their participation, debriefed them regarding the experimental hypotheses, and concluded the study.

3.     Results

This Part proceeds in three sections. First, it reports preliminary analyses regarding the construction of the measures that we used to evaluate participants’ responses. Second, it reports the main results of the study:  how the type of evidentiary “trapdoor” evidence—and the judge’s decision to admit or exclude it—affected participants’ perceptions of the accuracy, procedural fairness, and legitimacy of the trial. Third, it reports a mediation analysis that explores how perceptions of the “trapdoor” evidence, the accuracy of the trial, and the trial’s procedural fairness affected participants’ willingness to legitimize the legal tribunal.

i.     Preliminary Matters

We measured participants’ impressions of several different aspects of the trial to which they were exposed. First, we measured their impressions of the accuracy and procedural fairness of the trial.133 We measured these concepts by posing several questions to participants with 7-point Likert scales,134 which we converted into composite variables.135

To ensure that these questions truly measured the psychological constructs of decisional accuracy and procedural fairness, we conducted a principal-component analysis with an oblique rotation.136 The analysis confirmed that the four questions that purported to measure decisional accuracy together explained 62.12% of the variance in participants’ responses. The four questions that purported to measure procedural fairness separately explained another 20.19% of the variance in participant’s responses, which suggests that (1) the eight items measured two distinct concepts (“factors”); and (2) they explained most of the variation in participants’ responses.137 The results of the principal-component analysis appear in the table below.

Table 2: Principal Component Analysis 103 3 Sevier Table 2

Because the principal-component analysis confirmed that the eight questions measured two different psychological constructs, we averaged participants’ responses to the decisional accuracy questions and, separately, their responses to the procedural fairness questions to create a “decisional accuracy” index scale and a “procedural fairness” index scale.  Both scales were highly reliable.138

We repeated this process with respect to five questions we posed to participants to measure their perceptions of the trial’s legitimacy,139 their view of the moral appropriateness of the judge’s decision to admit or exclude the surprise evidence, and their impressions of the strength of the evidence against the defendant. Each set of items measured different psychological constructs and each was highly reliable.140

Preliminary analyses revealed that participants believed that this was a close case legally: participants were almost evenly split regarding whether to convict the defendant.141 As expected, participants’ impressions of the strength of the case strongly predicted their hypothetical verdicts.142 Also as expected, their impressions of the strength of the case less strongly predicted their impressions of the accuracy, fairness, and legitimacy of the legal tribunal.143

ii.     Main Analysis

Next, we evaluated our hypotheses underlying Study 1. We hypothesized that—contrary to the expectations of evidence rule makers—the admission of trapdoor evidence would negatively impact participants’ views of the fairness of the trial and that it would not positively impact their views of the accuracy of the trial. We also expected that participants would find the trial to be more legitimate when the judge excluded the “trapdoor” evidence than when the judge admitted the evidence, and we expected this to be true across all types of evidentiary trapdoors.

To determine whether (1) the form of the evidentiary trapdoor evidence; and (2) the admissibility of the evidence affected participants’ views of the accuracy and fairness of the trial, we examined the effect of these factors on participants’ perceptions of the tribunal’s decisional accuracy and procedural fairness.144 The results confirmed our hypothesis. First, consistent with prior published work on perceptions of fairness and accuracy in the American adversarial system, the analysis revealed that participants found the trial to be significantly fairer than it was accurate.145 Also as expected, the results revealed the form of the trapdoor evidence had no effect on participants’ perceptions.146 Importantly—and as expected—we found a statistically significant effect of the admissibility of the evidence, such that participants’ ratings of accuracy and justice were higher when the judge excluded the trapdoor evidence compared to when he admitted the evidence.147

Most importantly, the analyses showed a significant relationship between the type of evaluation that participants made (i.e., the accuracy or fairness of the trial) and the judge’s decision to admit the trapdoor evidence.148 Specifically, the decision to admit or exclude the trapdoor evidence had no effect on participants’ perceptions of the accuracy of the trial.149 The admissibility decision, however, did affect participants’ perceptions of the fairness of the trial, such that participants deemed trials in which the judge excluded the prosecution’s trapdoor evidence fairer than trials in which the judge admitted the prosecution’s evidence.150

Finally, we examined participants’ willingness to legitimize the court that conducted the trial.151 As expected, the analysis revealed a significant effect of the judge’s admissibility decision on participants’ perceptions of the court’s legitimacy, and this effect was observed regardless of the type of trapdoor evidence that the prosecutor presented to the judge.152 Specifically, participants were less willing to legitimize legal tribunals that admitted the trapdoor evidence than they were to legitimize tribunals that excluded the evidence.153 The graphs below depict participants’ perceptions of the accuracy and fairness of the trials that they observed—along with their willingness to legitimize those trials—as a function of the type of trapdoor evidence presented and the court’s admissibility decision.

Figure 1: Perceptions of the Accuracy and Fairness of Trial Verdicts as a Function of Evidence Admissibility

103 3 Sevier Fig. 1

Figure 2: Perceived Legitimacy of Verdicts as a Function of Evidence Admissibility and the Type of Evidence Proffered 103 3 Sevier Fig. 2

In sum, the results of Study 1 support our experimental hypotheses. When the judge allowed the prosecutor to proffer trapdoor evidence—regardless of its form—participants viewed the trial as less fair and less legitimate than when the judge denied the prosecutor the opportunity to proffer the evidence. Further, the judge’s decision to admit or exclude the evidence had no effect on participants’ perceptions of the tribunal’s ability to ascertain the truth of the underlying conflict, even though increased accuracy is the purported rationale for admitting such evidence. Thus, if admitting trapdoor evidence actually increases the accuracy of a trial, participants did not believe that it does so.

iii.     Serial Mediation Analysis

Our final analysis examined the psychological processes that underlie the relationship between the judge’s admissibility decision with respect to the trapdoor evidence and participants’ perceptions of the tribunal’s legitimacy. We hypothesized that participants’ agreement or disagreement with the judge’s decision was associated with their perceptions of the fairness of the trial, which affected their perceptions of the trial’s legitimacy. We tested this hypothesis through a serial mediation analysis.154

The serial mediation analysis confirmed our hypotheses. The admission of the evidence was associated with greater assessments that the decision to admit the evidence was immoral.155 Also as predicted, participants’ views of the morality of the admission of the evidence were associated with their impressions of the fairness of the trial, such that participants who found the evidence to be less moral found the trial to be less fair overall.156 Finally, participants’ views of the fairness of the trial significantly predicted their willingness to legitimize the trial, such that participants who viewed the trial as less fair also viewed it as less legitimate.157 Additional statistical tests confirmed that the effect of the admissibility of the trapdoor evidence on participants’ willingness to legitimize the trial was fully mediated (that is, fully accounted for) by their views of the morality and fairness of the admissibility decision.158

A follow-up analysis strengthened these conclusions. The analysis demonstrated that the proposed mediation pathway explained the most variance of the different combinations of mediators and predictors in the model.159 Adding participants’ perceptions of the tribunal’s decisional accuracy into the model did not alter these results. Although perceptions of the tribunal’s accuracy did predict participants’ willingness to legitimize the trial, with perceived accuracy leading to more willingness to legitimize the tribunal, the effect was substantially weaker than the effect of participants’ perceptions of the trial’s fairness.160 An illustration of the mediation analysis, which includes beta coefficients from the linear regression analyses, appears below.161

Figure 3: Serial Mediation Analysis Examining The Role of Moral Judgment and Procedural Fairness in Explaining the Effect of Evidence Admissibility on Perceptions of Verdict Legitimacy

103 3 Sevier Fig. 3

iv.     Discussion

Study 1 yields several findings bearing on the relationship between a tribunal’s evidentiary admissibility decisions and the public’s willingness to legitimize the tribunal. These results do not support the primary justification offered by evidence rule makers for allowing parties to proffer trapdoor evidence. Although, consistent with our experimental hypotheses, participants recognized the equity and fairness concerns associated with the evidence, they did not recognize the purported benefits of the evidence for the tribunal’s truth-seeking function. They perceived no meaningful increase in the court’s ability to determine the truth by admitting the evidence and no meaningful decrease in the court’s truth-seeking ability when the court excluded the trapdoor evidence.

Study 1 confirmed and replicated several past research findings related to the psychological processes by which citizens legitimize legal tribunals. First, Study 1 confirmed that the public’s willingness to legitimize a legal tribunal is substantially a function of two competing, largely independent values: (1) decisional accuracy; and (2) procedural fairness. It also confirmed that participants tend to perceive American trials as fairer than they are accurate. Finally, it confirmed that, although the public’s perceptions of how well the courts determine the truth underlying a legal dispute is relevant to whether the public will legitimize its courts, the fairness of the legal process is a stronger influence on their willingness to do so.

Study 1 leaves several questions unanswered regarding the robustness of the reported effects, however. For example, is the cost to the tribunal’s legitimacy by admitting trapdoor evidence the same regardless of the party that attempts to admit the evidence, the nature of the proceeding, or the demographics of the members of the jury? Study 2 seeks answers to these questions to create a robust behavioral model regarding the effects of trapdoor evidence on the courts’ institutional legitimacy.

B.     Study 2: Legal Moderators

Our second study serves dual purposes. First, it attempts to replicate, with an additional independent sample, our findings from Study 1. Second, it extends these findings by focusing on potential contextual modifiers of the evidentiary trapdoor effect and exploring its robustness.

We explored the robustness of the effects in Study 1 by examining whether the evidentiary trapdoor effect is present in the context of both civil and criminal cases, and whether it exists regardless of the identity of the party offering the evidence. We manipulated both of these factors in Study 2, in which participants read a legal case similar to the case at issue in Study 1. This time, however, the case was sometimes a civil dispute and sometimes a criminal dispute. Additionally, sometimes the party proffering the evidence was the prosecutor (or plaintiff) and sometimes the party was the defendant.

Based on prior research, we identified two distinct hypotheses regarding the influence that the type of tribunal and party has on the evidentiary trapdoor effect as it relates to the public’s perceptions of the courts’ legitimacy. First, participants might view the admission of trapdoor evidence as particularly odious in a criminal case than in a “mere” civil case, where the stakes do not implicate the defendant’s liberty.162 Second, the identity of the party proffering the trapdoor evidence may also matter; to the extent that laypeople believe that defendants often win court cases on “legal technicalities,” they might be immune to the evidentiary trapdoor effect when it is used against a defendant, and might be hostile toward defendants who use it against prosecution (or plaintiff) witnesses.163

The psychology literature, however, offers an alternative set of hypotheses. The overall strength of the evidentiary trapdoor effect that we observed in Study 1—which was tied to the perceived morality and fairness of the admissibility procedure—suggests that participants would find trapdoor evidence objectionable regardless of the litigation setting. This also suggests that the party that proffers the evidence should have no bearing on participants’ views of the acceptability of the use of the evidence at trial.164 Our methods for testing these hypotheses, and the results we found, appear below.

1.     Participants

We recruited 239 participants for Study 2, MTurk.165 The logistics for recruiting our participants mirrored the procedure from Study 1. As in Study 1, participants in Study 2 were a representative cross-section of jury-eligible citizens. The average participant was 35.10 years old with a standard deviation of 9.99. The sample was split evenly by gender, with women composing 49.80% of the sample. The sample generally reflected the racial diversity of the United States population as well, with 24.60% of the sample identifying as non-white.166 Over 59.00% of participants had completed at least a college degree, and the median participant income was between $40,000 and $49,999. Participants’ political preferences were distributed across the political spectrum, and most participants described themselves as moderate (22.3%) or liberal (33.6%). Table 3 on page 39 provides detailed descriptive statistics for all participants involved in Study 2.

2.     Procedure and Measures

Study 2 followed many protocols used in Study 1. With minor deviations, participants read the same fact pattern as did participants in Study 1, involving a shooting in the parking lot of an upscale mall. This time, however, instead of randomly assigning participants to one of four types of evidentiary trapdoors, all participants read about trapdoor evidence in its ‘impeachment’ form from Study 1.167 We varied three other facets of the trial instead: the judge’s admissibility decision, the nature of the legal proceeding, and the identity of the party proffering the evidence. 

Table 3: Participant Demographics (Study 2)

 103 3 Sevier Table 3

We varied the nature of the legal proceeding by informing half of our participants that they were reading about a second-degree murder trial, and informing the other participants that they were reading about a civil wrongful death action filed by the victim’s next of kin. The witnesses and their testimony were identical in both experimental conditions and tracked the testimony from Study 1.

Within the civil and criminal case conditions, we also varied the identity of the party that proffered the trapdoor evidence. In the criminal case condition, if the prosecutor moved to proffer the trapdoor evidence, events played out as they did in the impeachment condition in Study 1. The prosecutor moved to recall the defendant to the stand to question him about his prior arrest for lying to investigators in an unrelated case. If the defense attorney moved to proffer the evidence, the evidence instead pertained to the same arrest for lying to investigators, but the arrest instead belonged to the defendant’s co-worker, who served as a witness for the prosecution. We used this same paradigm in the civil case conditions: either the plaintiff’s attorney or the defense attorney proffered the trapdoor evidence, which was relevant to the credibility of either the defendant or the defendant’s co-worker, depending on the experimental condition.

Participants then answered the same questions posed to participants in Study 1: they were asked questions designed to measure their impressions of (1) the accuracy, procedural fairness, and legitimacy of the trial; (2) the case against the defendant, along with their imagined verdicts if they had served on the jury; and (3) each piece of evidence against the defendant, including the trapdoor evidence. As in Study 1, we posed several demographic questions to participants before we concluded the experiment.

3.     Results

This Part proceeds in two sections. First, it examines the effect of the judge’s admissibility decision vis-à-vis trapdoor evidence, the effect of the type of case against the defendant, and the effect of the identity of the party proffering the evidence on participants’ perceptions of the accuracy, fairness, and legitimacy of the legal tribunal. Next, to the extent that participants’ perceptions of the court’s legitimacy are affected by the above factors, we examine the psychological variables that facilitate the effect. We report our results below.168

i.     Main Analysis

To test the experimental hypotheses that underlie Study 2, we examined the effect of the legal context, the identity of the proffering party, and the judge’s admissibility decision on people’s willingness to legitimize the legal tribunal in which the trapdoor evidence was introduced.169 The analysis revealed several results in line with our hypotheses. First, we observed no effect of the legal context on participants’ willingness to legitimize the proceedings; participants were willing to legitimize a criminal trial that potentially featured trapdoor evidence as much as they were willing to legitimize a civil trial that did so.170 We also observed no difference in participants’ willingness to legitimize the legal tribunal based on who proffered the evidence. As expected, we did however observe an effect of the judge’s admissibility decision regarding the trapdoor evidence. As in Study 1, participants were significantly less likely to legitimize legal tribunals that admitted the trapdoor evidence compared to tribunals that excluded that evidence.171 A graph of these general effects appears below.

Figure 4: Main Effect of Trial Context, Proffering Party, and Admissibility Decision on Perceived Legitimacy of Trial Verdicts

103 3 Sevier Fig. 4

Intriguingly, the effect of the judge’s admissibility decision on participants’ willingness to legitimize the tribunal was qualified by a statistically significant interaction with the identity of the actor who proffered the evidence.172 The extent to which participants delegitimized trials in which the judge admitted the trapdoor evidence differed depending on whether the actor proffering the evidence was the prosecutor (or plaintiff) or the defense attorney.

To explore the nature of this interaction, we examined the effect of the judge’s admissibility decision on participants’ willingness to legitimize the legal tribunal separately depending on whether the actor was the prosecutor (or plaintiff) or the defendant. The results confirmed our experimental hypotheses and appear in the graph below. When the judge refused to admit the trapdoor evidence, there was a marginally significant effect of the actor on participants’ perceptions of legitimacy, such that participants viewed the trial as more legitimate when the judge refused to admit the prosecutor’s (or plaintiff’s) evidence compared to when the judge refused to admit the defense attorney’s evidence.173 Conversely, when the judge ruled the trapdoor evidence admissible, we found a statistically significant effect of the proferring actor, such that participants viewed the trial as less legitimate when the judge allowed the prosecutor’s (or plaintiff’s) trapdoor evidence compared to trials in which the judge allowed the defense attorney’s evidence.174

Figure 5: Interactive Effect of Admissibility Decision and Party Identity on Perceptions of Verdict Legitimacy

103 3 Sevier Fig. 5

In sum, the results from Study 2 support our experimental hypotheses. First, we replicated the results from Study 1. Participants were more willing to legitimize a legal tribunal that excluded trapdoor evidence compared to a tribunal that admitted the evidence. This was true regardless of whether the subject matter involved a criminal murder trial—where the stakes are arguably higher—or involved a civil wrongful death dispute.175

Study 2 examined whether the identity of the party proffering the trapdoor evidence complicated the effect of the judge’s admissibility decision on people’s perceptions of the legitimacy of the tribunal. Study 2 found that the identity of the party proffering the trapdoor evidence did complicate people’s perceptions of the legitimacy of the trial, but in a counterintuitive way. The prosecutor’s use of trapdoor evidence—not the defense’s use of the evidence—was more salient to participants’ perceptions of the legitimacy of the trial.  Participants legitimized the trial marginally more often when the initiator of the proceedings (either the prosecutor or the plaintiff) was prohibited from admitting the evidence, and participants significantly delegitimized the trial when the lawsuit’s initiator was allowed to present the evidence to the jury. Conversely, participants legitimized the trial more often when the defense attorney was allowed to present the trapdoor evidence. The results suggest an asymmetry in how laypeople legitimize trials containing trapdoor evidence. If a party brings a case to trial against a defendant, laypeople are highly critical of that party’s gamesmanship of the evidentiary rules to prove her case. Indeed, laypeople are more critical of gamesmanship used against a defendant compared to its use by a party defending a lawsuit.

ii.     Mediated Moderation Analysis

We now extend these findings by exploring the psychological mechanism by which the judge’s admissibility decision and the identity of the party proffering the trapdoor evidence jointly explain participants’ willingness to legitimize the tribunal. As in Study 1, we hypothesize that participants’ perceptions of the fairness of the proceedings, itself affected by the perceived morality and fairness of the decision to admit the trapdoor evidence, will fully explain the joint effects of the admissibility decision and the party’s identity. To test this, we performed a more complex version of the mediation analysis that we performed in Study 1.176

Our hypothesis was confirmed. The judge’s admissibility decision regarding the trapdoor evidence influenced people’s willingness to legitimize the trial, but the magnitude of the influence depended on the identity of the party proffering the evidence. Participants were less likely to legitimize the trial when the prosecution’s trapdoor evidence was admissible compared to when the evidence was presented by the defense, and they were more likely to legitimize the trial when the court excluded the prosecution’s evidence compared to when the court excluded the defense’s evidence.177

The analysis confirmed the significant interactive effect that we found involving the judge’s decision and the party’s identity on participants’ perceptions of the overall fairness of the trial. Specifically, participants thought the trial was less fair when the court allowed the prosecution to present the trapdoor evidence and found the trial fairer when the court excluded the prosecution’s evidence.178 Further, participants’ perceptions of the fairness of the trial overall were significantly, strongly, and positively related to their willingness to legitimize the trial.179

Most importantly, when we added participants’ perceptions of the trial’s overall fairness to the model, which already included the judge’s admissibility decision, the identity of the actor, and participants’ willingness to legitimize the trial, the interactive effect of the judge’s decision and the party’s identity on participants’ perceptions of legitimacy became non-significant.180 Thus, as we predicted, participants’ perceptions of the fairness of the trial entirely explains the interactive effect found in Study 2.181 An illustration of the mediated moderation analysis, which includes the corresponding beta coefficients, appears on page 1203.182

Figure 6: Mediated Moderation Analysis Examining How Differing Perceptions of Procedural Fairness Explain the Interactive Effect of Evidence Admissibility and Party Identity on Perceptions of Verdict Legitimacy

103 3 Sevier Fig. 6

iii.     Discussion

Study 2 builds upon the results reported in Study 1 in several ways. In addition to replicating the negative effect of admitting trapdoor evidence on the public’s perception of the legal tribunal’s legitimacy, Study 2 also examined if several other legally relevant factors modify this effect. We found that the negative effect of admitting the trapdoor evidence was robust—the effect from Study 1 easily replicated in Study 2 and did so regardless of whether the trial was civil or criminal in nature.  Study 2 also examined whether the negative effect of the admission of trapdoor evidence interacts with the identity of the party proffering the evidence. We found such an interaction, but it was counterintuitive. Rather than punishing the defendant for using a potential “loophole” for admitting evidence at trial, participants were far more attentive to the behavior of the lawsuit’s initiator—the prosecutor in the criminal trial and the plaintiff in the civil dispute.  

This finding is consistent with participants’ perceptions of legitimacy being tied to their impressions of the trial’s fairness instead of their impressions of the trial’s accuracy. The other results reported in Study 2 also support this proposition.  Although the judge’s decision to admit the evidence and the identity of the party proffering the evidence jointly explained participants’ willingness to legitimize the trial—and their impressions of the trial’s fairness—these factors had no effect on participants’ perceptions of the trial’s accuracy. This is particularly interesting in light of the purported accuracy-focused rationale justifying the existence of evidentiary trapdoors. The mediated moderation analysis further confirmed these findings.  Although participants’ perceptions of the trial’s accuracy affected their perceptions of the tribunal’s legitimacy, decisional accuracy did not mediate the joint impact of the judge’s admissibility decision and the proffering party’s identity on participants’ willingness to legitimize the trial. Instead, participants’ perceptions of the fairness of the admissibility decision was a significant—and complete—mediator of that effect.

Study 2 answers several questions regarding the effects of situational variables on the public’s perceptions of the legitimacy of tribunals that allow trapdoor evidence.  Still, this behavioral model can be enhanced further. Study 3 examines the robustness of the trapdoor evidence’s effect, accounting not only for situational variables, but also individual jurors’ personality traits and attitudes toward the judiciary.

C.     Study 3: Psychological Moderators

The final study on which we report is a short, follow-up analysis regarding the effects we found in Studies 1 and 2. Readers may note that we studied the trapdoor effect in the sterile confines of the (virtual) psychology laboratory. They may wonder whether participants are truly less willing to legitimize legal tribunals that allow juries to consider “unfair” evidence that purports to increase the court’s decisional accuracy, considering the inevitable variations among participants with respect to their backgrounds, personality traits, and attitudes toward the courts. These psychological characteristics are distinct from the situational factors—including the type of evidence presented, the type of case in which the evidence is presented, and the identity of the party presenting the evidence—that were the focus of Studies 1 and 2. In this follow-up study, we examine the trapdoor effect that we found in Studies 1 and 2, but we explicitly control for several demographic variables, personality traits, and attitudes vis-à-vis our participants. Because the trapdoor evidence effect was so reliable and robust, we hypothesize that we will continue to observe the trapdoor evidence effect despite the differing backgrounds, personality traits, and court-related attitudes of our participants.

1.     Participants, Procedures, and Measures

We recruited no new participants for Study 3. Instead, we pooled together data that we gathered from those who participated in Study 1 and Study 2, for a total of 471 participants. Recall that we asked participants in Study 1 and Study 2 to answer several demographic questions, several questions about the case, and several questions about their own personality traits. In this Part, we expand on their responses to these questions.

Participants answered several demographic questions. They were asked for the year in which they were born, the gender with which they identify, and their race.  We asked them for their annual household income in increments of $10,000 from $0 to $100,000. We also asked them to select the highest level of education that they had achieved, from “some high school” to “Ph.D. and/or professional degree.” Finally, we twice asked them to describe their political orientation. First, we asked them to describe themselves on a five-point scale from “Very Conservative” to “Very Liberal,” and then we asked them to classify themselves either as “More Conservative” or “More Liberal.”

In addition to demographic questions, we asked participants three questions related to the case that they read. First, we asked them to tell us, on a 1–7 Likert scale, the likelihood that the defendant committed the crime. Second, we asked participants questions to determine their familiarity with the legal system. We asked them to tell us, on a 1–7 scale, how familiar they are with the legal system compared to the average person. We also asked questions about specific experiences that they may have had with courts, judges, and other courtroom personnel.

Finally, we asked participants several questions designed to measure different personality traits that prior research suggests are relevant to people’s perceptions of the legitimacy of legal tribunals: the extent to which participants are authoritarian, empathic, cynical, subscribe to a social dominance view of relationships, and believe that people generally get what they deserve.183 

2.     Results and Discussion

Study 3 examines the robustness of the evidentiary trapdoor effect in light of participants’ demographics, personality traits, and attitudes toward the courts. Specifically, we examine whether the evidentiary trapdoor effect exists even when we control for these additional variables. We hypothesized that the effect would remain statistically significant, even controlling for all of these considerations about our participants.

To test our hypothesis, we constructed three linear regression models from the data that we collected from participants in Study 1 and Study 2.184 Each model builds upon the prior model by controlling for different factors vis-à-vis our participants while examining the effect of the judge’s admissibility decision on participants’ willingness to legitimize the tribunal. The first model examines the effect of the admissibility decision (labeled “allow” in the model) controlling for demographic variables. The second model includes these variables and controls for participants’ personality characteristics. The final model includes the variables from Models 1 and 2 and controls for participants’ attitudes toward the courts generally and their specific attitudes toward the defendant’s culpability. All three models appear below.

Table 4: Evidentiary Trapdoor Effects on Perceptions of Legitimacy 103 3 Sevier Table 4

Table 4 supports our hypotheses. Even as we increased the control variables across the three models, the effect of the judge’s admission of trapdoor evidence on participants’ perceptions of the trial’s legitimacy remained strong and statistically significant. In sum, the evidentiary trapdoor effect is robust, and exists independent of variations with respect to the legal setting and with respect to the variance among the participants who evaluate the trial.185

V.     Implications, Objections, and Conclusions

The law of evidence is replete with trapdoors—broadly defined as the strategic use of counterintuitive, “accuracy-enhancing” applications of the Federal Rules—to which crafty litigators are allowed (and encouraged) to avail themselves. These trapdoors exist within the law of impeachment, the law governing character evidence, the rules covering the bar against hearsay and its exceptions, and the rules that govern privileged information at trial.

These trapdoors emanate primarily from two sources: (1) the text of the Federal Rules of Evidence, where the Rules are silent on a critical issue; and (2) the courts, for even when the Rules are clear, courts frequently and substantially deviate from the written text. With respect to trapdoor evidence, these deviations and ambiguities are justified largely based on rule makers’ non-empirical intuitions that trapdoor evidence increases the odds that the fact finder will reach a legally accurate verdict. Professional legal norms therefore encourage—or even require—advocates to avail themselves of these counterintuitive “accuracy-enhancing” applications of the evidentiary rules on behalf of their clients.

The three original experiments reported in this Article reveal that rule markers have failed to consider the significant cost of the use of evidentiary trapdoors: a decrease in the perceived procedural fairness of trials that allow advocates to present trapdoor evidence, and a concomitant decrease in the public’s willingness to confer legitimacy on these legal tribunals. Prior research strongly, however, suggests that there is a reciprocal relationship between the public’s perception of the truth-finding ability of American courts and their ability to provide fair procedures for adjudicating legal disputes.186 This prior research has also demonstrated that the public’s willingness to legitimize governmental actors depends—to a much greater degree—on the public’s perceptions of the procedural fairness provided by the governmental actor’s decision making process rather than on the actor’s substantive accuracy.187 It was, therefore, an open question whether a legal actor’s use of “accuracy-enhancing” trapdoor evidence—with the blessing of the Federal Rules of Evidence—actually leads to more legitimate trials. Three studies reported in this Article suggest that this it does not.

These studies produce myriad implications for the nature of the conversation regarding how policymakers interpret the Federal Rules of Evidence, the role of empirical science in policymaking more generally, and the consequences for attorneys’ strategic choices at trial. These implications, along with the limitations and future directions of this research, are discussed below.

A.     Research and Policy Implications

The psychological research on the popular legitimacy of American courts suggests that people are most willing to legitimize legal tribunals when they achieve two related but psychologically distinct objectives: (1) find the correct facts upon which to apply the law (“decisional accuracy”); and (2) arrive at those conclusions in a manner the parties perceive to be fair (“procedural justice”).188 The achievement of these objectives has important effects on the courts, as the consequences following a loss of a governmental institution’s popular legitimacy can be severe and substantial.189

To date, the academic conversation regarding attorneys’ use of the Federal Rules of Evidence more broadly—and their use of trapdoor evidence more specifically—has focused primarily on its impact on a legal tribunal’s decisional accuracy.190 While decisional accuracy is important, the conversation virtually ignores a second condition precedent to the court’s popular legitimacy: the impact of trapdoor evidence on public perceptions of the trial’s procedural fairness. This is so despite the fact that a wealth of empirical evidence demonstrates: (1) that trial procedures that enhance or are perceived to enhance decisional accuracy frequently result in diminished perceptions of the trial’s procedural justice; and (2) procedural justice is a much stronger predictor of whether people will legitimize the decisions of legal tribunals that present trapdoor evidence to fact finders.191

The experiments reported in this Article provide a necessary empirical counterweight to the prevailing assumption that evidentiary trapdoors will lead to increased perceptions of the ability of American trial courts to reach accurate verdicts and, therefore, an increase in the public’s willingness to legitimize them. As many scholars have noted, evidentiary policy has too often been made—however well-meaning and well-reasoned—based on policymakers’ hunches and best guesses about human behavior that do not always align with the findings of empirical scientific research.192 With respect to trapdoor evidence, the story is no different. The findings we report in this Article provide preliminary evidence suggesting that not only do people disbelieve that trapdoor evidence improves the accuracy of American trial courts, but trapdoor evidence may actively undermine a more fundamental goal: creating a dispute resolution system whose decisions the public respects and follows without coercion.

To that end, the academic conversation regarding the application of the Federal Rules of Evidence must be expanded substantially to include empirically-informed debates regarding the effects of evidentiary rules on people’s perceptions of the fair treatment of litigants. Because the use of trapdoor evidence has been viewed (incorrectly) as improving decisional accuracy, the conversation regarding whether any individual advocate should avail herself of such evidence has instead revolved around the attorneys’ personal level of comfort in doing so, or the attorneys’ moral imperative to do so.193 However, if the admission of trapdoor evidence produces substantial burdens on the legitimacy of the legal system, then this conversation deserves renewed discussion among policymakers.

To the extent that these findings require us to rethink traditional attitudes toward the application of trapdoor evidence, several areas of the Federal Rules of Evidence can be re-examined and restructured. Perhaps the best example is the ban on hearsay evidence under Article VIII of the rules. The rule barring hearsay evidence rests almost exclusively on the empirically suspect assumption that hearsay decreases the tribunal’s decisional accuracy because jurors overvalue its probative weight.194 From that starting assumption, the Rules create a dizzying array of nearly thirty exceptions for hearsay statements that are believed to be reliable and will increase the tribunal’s decisional accuracy. Scientific research, however, strongly suggests that the use of hearsay evidence pursuant to a hearsay exception does not increase the odds that the fact finder will make a more accurate decision and, perversely, the use of hearsay leads to greater perceptions that the trial is unfair.195 Instead, rule makers should re-conceptualize the rationale for the bar on hearsay—by premising the bar on a recognition that using second-hand evidence to find a defendant legally liable for an offense violates our notions of fair process. This, in turn, would lead to a reduction of the number of exceptions to the hearsay rule that are based on unsupported intuitions about their effect on decisional accuracy.196

The Article VI impeachment rules could also be restructured to better align with the psychological legitimacy and procedural justice research. One example stems from FRE 609, which governs the admissibility of a defendant’s prior convictions.197 Subject to several complex caveats, the rule allows the prosecutor to use most of a testifying criminal defendant’s prior convictions to impeach the defendant’s credibility, even if the prior conviction does not stem from an act of dishonesty.198 This places a criminal defendant in a catch-22: if she does not testify, she is unable to present her case to the fact finder as effectively; but if she does testify, the prosecutor can use the conviction on cross-examination to discredit her. Researchers recently conducted a study examining several real-world cases involving the application of FRE 609 and found (1) no relationship between the presence of a prior conviction and the tribunal’s odds of reaching a factually correct verdict; and (2) prior convictions made jurors more likely to convict a defendant when the other evidence was otherwise too weak to support a conviction, therefore potentially leading to erroneous convictions.199 Under a procedural justice rationale—rather than a decisional accuracy rationale—rule makers could restructure FRE 609 either to disallow prior convictions entirely, or to allow the presiding judge greater flexibility in determining the admissibility of the evidence in light of its threat to the public’s perceptions of procedural justice.

These findings have implications for other areas of the law as well. One such example is binding arbitration—a less formal dispute resolution procedure that is a cheaper alternative to a bench or jury trial and explicitly does not follow a binding evidentiary code in the absence of a specific agreement between the parties.200 Arbitration therefore allows participants even greater freedom to present evidence to the arbitrator than at trial, but it also increases the likelihood that substantively or procedurally unfair evidence reaches the fact finder.201 In light of the growth of the binding arbitration market in recent years, future research should examine whether laypeople’s perceptions of the legitimacy and procedural fairness of the arbitration process is affected by the use of trapdoor evidence.202 

B.     Conclusion

Elven Joe Swisher was eventually incarcerated, but not for his brazen perjury in the trial of David Hinkson. The United States convicted Swisher of swindling the government out “of nearly $100,000 in veterans’ benefits,” and he served “[one] year and [one] day in federal prison.”203 The damage to Hinkson’s trial, however, was already done. Guided by a novel, counterintuitive interpretation of the Federal Rules of Evidence—albeit a well-intentioned attempt at maximizing the jury’s ability to reach an accurate verdict—the information regarding Swisher’s phony Korean War claims never reached the jury. The trial court excluded the evidence, Hinkson was convicted of the charged offenses, and the jury was none the wiser.

Empirically uninformed applications of the Federal Rules of Evidence form the basis of the evidentiary trapdoor, a pernicious feature of the federal evidentiary code that arises when courts apply evidentiary rules in ways that flatly deviate from their plain language or expand the reach of those rules in unintended ways. The resulting overemphasis on decisional accuracy has two potentially devastating and unintended consequences: the use of evidentiary trapdoors lowers the public’s perceptions of the fairness of American trials and, more disturbingly, causes the public to become less willing to legitimize the legal system. The academic conversation surrounding the use of trapdoor evidence is therefore incomplete and unsatisfying until it accounts for these important unintended effects. A more robust attention to these psychological phenomena will have beneficial effects not only for American trial courts, but also for the public citizens who rely upon—and legitimize—the legal system.


  1. [1]. Trap, The Free Dictionary, (last visited Jan. 20, 2018). Dictionaries sometimes cross-list the term “trapdoor” with the related term “back door,” which they define as “a secret, furtive, or illicit method, manner, or means.” Back door,, (last visited Oct. 22, 2017).

  2. [2]. United States v. Hinkson, 585 F.3d 1247, 1268 (9th Cir. 2009) (en banc) (Fletcher, J., dissenting). “A Purple Heart is an award given to” United States military personnel “who are wounded in combat.” Id. at 1270; see The Military Order of the Purple Heart, http://www. (last visited Jan. 20, 2018).

  3. [3]. See Hinkson, 585 F.3d at 1269–70 (noting that the United States argued in its opening statement to the jury that Swisher was a Korean War combat veteran, and arguably suggested at trial that Swisher’s experience in the Korean War formed the basis of Hinkson’s sincere desire to hire Swisher to murder federal officials).

  4. [4]. Swisher produced a DD Form 214 (“Certificate of Release or Discharge from Active Duty”). Id. at 1254, 1275. A DD-214 is a document issued by the United States Department of Defense when a military service member retires, separates, or is discharged from active duty in any branch of the United States Armed Forces. See The Road to Acquire Your DD214, http://www. (last visited Jan. 20, 2018).

  5. [5]. Hinkson, 585 F.3d at 1276 (recounting how Swisher fabricated his records).

  6. [6]. The defense attorney’s strategy to impeach Swisher’s trial testimony regarding his military service constitutes a classic example of the “impeachment by contradiction” doctrine, pursuant to Federal Rule of Evidence (“FRE”) 607. Id. at 1282. The Federal Rules place no formal restrictions on the manner in which the defense conducts the impeachment. See Fed. R. Evid. 607; see also United States v. Castillo, 181 F.3d 1129, 1132–33 (9th Cir. 1999) (noting that “the concept of impeachment by contradiction [under FRE 607] permits courts to admit extrinsic evidence that specific testimony is false” on the theory that “the witness should not be permitted to engage in perjury, mislead the trier of fact, and then shield himself from impeachment by asserting the collateral-fact doctrine”) (quoting 2A Charles A. Wright & Victor J. Gold, Federal Practice and Procedure, § 6119, at 116–17 (1993) (citations omitted)). The court majority, however, construed the defense attorney’s strategy as trying to demonstrate the witness’s general character for untruthfulness, pursuant to FRE 608(b). Hinkson, 585 F.3d at 1282. Rule 608(b), however, forbids the use of “extrinsic” documents to prove a witness’s untruthful character, because the courts deem a witness’s character as per se collateral. See id.; see also United States v. Castillo, 181 F.3d 1129, 1132–33 (9th Cir. 1999) (discussing the distinction between impeachment evidence offered pursuant to FRE 607 and FRE 608(b)).

  7. [7]. Hinkson, 585 F.3d at 1277, 1281. The trial court did, however, allow defense counsel to question Swisher regarding the alleged falsehoods with respect to Swisher’s military service. Id. at 1271. Swisher denied the allegations. Id. From the jury’s perspective: (1) defense counsel questioned Swisher regarding his military service; (2) in response, Swisher dramatically produced a document purporting to show his military service; and (3) Swisher denied defense counsel’s claims (which were unsubstantiated from the jury’s perspective) that Swisher had fabricated the details of his service. See id. at 1271–72. In sum, what should have resulted in a successful impeachment of the witness appeared to the jury, instead, as a failed impeachment attempt.

  8. [8]. United States v. Musquiz, 45 F.3d 927, 929 (5th Cir. 1995). Although Musquiz was not a police officer, he was tried jointly, and accused of conspiring with, former police officer Robert Martinez Gatewood. Id. at 928–29.

  9. [9]. Id. at 929.

  10. [10]. Id. at 930. Additional facts of this case appear in the brief for the United States. See, e.g., Brief for the United States at 15, United States v. Musquiz, 45 F.3d 927, 929 (5th Cir. 1995) (stating that the defendant was not Mirandized at the scene of his arrest). See generally Miranda v. Arizona, 384 U.S. 436 (1966) (holding that statements that defendants make in response to police interrogation are admissible at trial only if the prosecution can show that the government informed the defendant of her right to consult with an attorney before and during questioning and the right against self-incrimination before questioning, among other rights).

  11. [11]. See Musquiz, 45 F.3d at 930. The reason for the delay is not entirely clear from the record.

  12. [12]. Id. at 929–30.

  13. [13]. Id. at 930; see also U.S. Const. amend. V (stating, in relevant part, that “nor shall [a defendant] be compelled in any criminal case to be a witness against himself, nor be deprived of life, liberty, or property, without due process of law”).

  14. [14]. The appeals court reasoned that pre- and post-Miranda silence provide different incentives, such that the Supreme Court has treated them differently (by affording Constitutional protection to the latter, but not the former). Musquiz, 45 F.3d at 930–31; see also Brecht v. Abrahamson, 507 U.S. 619, 627–29 (1993) (explicitly addressing the issue). The Federal Rules of Evidence therefore are permitted to classify a suspect’s silence as an assertion that can be inconsistent with “non-silence” testimony at trial. Cf., e.g., United States v. Truman, 688 F.3d 129, 142 (2d Cir. 2012) (finding that in-court silence is inconsistent with prior statements); United States v. Iglesias, 535 F.3d 150, 159 (3d Cir. 2008) (same). See also Fed. R. Evid. 613(b) and advisory committee’s note (discussing the procedure for impeaching a witness with a prior inconsistent statement).

  15. [15]. Musquiz, 45 F.3d at 933.

  16. [16]. Additionally, certain rules, such as the rules governing privileged information, also rely on development under the common law. See Fed. R. Evid. 501 (explicitly leaving the nature of privileged information to the common law). Many rules also have various, complex subparts. See, e.g., id. R. 803 (listing more than 20 different exceptions to the hearsay rule).

  17. [17]. The distinguishing feature of the evidentiary trapdoor is its transitive property. It is not the mere use of a counterintuitive evidentiary strategy by itself that constitutes the trapdoor. Rather, it is the purpose for which courts permit the novel application of the evidentiary rule: to increase the fact finder’s ability to get at the truth underlying a dispute (the legal tribunal’s “decisional accuracy”).

  18. [18]. See, e.g., Monroe H. Freedman & Abbe Smith, Understanding Lawyers’ Ethics (4th ed. 2010) (advocating for a traditional, client-centered model of the role of counsel in the adversary system); W. Bradley Wendel, Lawyers and Fidelity to Law (2010) (arguing that especially in cases that would lead to perverse outcomes, attorneys act unethically when their tactics allow their clients to achieve ends that they are not legally permitted to achieve); Charles Fried, The Lawyer as Friend: The Moral Foundations of the Lawyer-Client Relation, 85 Yale L.J. 1060 (1976) (discussing how traditional legal ethics rules may require an attorney to make decisions for her client that she would not, as a matter of morality, make for herself); Stephen L. Pepper, The Lawyer’s Amoral Ethical Role: A Defense, a Problem, and Some Possibilities, 1986 Am. B. Found. Res. J. 613, 613 (presenting a moral justification for the self-described “amoral ethical role” of the attorney); Andrew M. Perlman, A Behavioral Theory of Legal Ethics, 90 Ind. L.J. 1639, 1639 n.1 (2015) (citing David A. Binder et al., Lawyers as Counselors: A Client-Centered Approach (3d ed. 2011) (citations omitted)) (“Some legal ethicists argue that the lawyer should comply with the client’s instruction and pursue every permissible tactic, that is, go right up to the line but not cross it.”); see also generally David Luban, Lawyers and Justice: An Ethical Study (1988) (discussing the unique moral dilemmas facing attorneys in the adversary system and discussing ways to resolve those dilemmas); William H. Simon, The Practice of Justice: A Theory of Lawyers’ Ethics (2009) (discussing, among other topics, the extent to which an attorney should exploit legal loopholes in ways that enable clients to gain unintended advantages).

  19. [19]. See, e.g., Alex Stein, Foundations of Evidence Law 1 (2005) (in the context of critiquing the traditional justifications for evidence law, articulating three of its objectives, and listing “accuracy in fact-finding” as the first objective); Lindsey C. Boney IV, Note, Forum Shopping Through the Federal Rules of Evidence, 60 Ala. L. Rev. 151, 170 (2008) (explaining a line of attack on the Hanna Rule in civil procedure involving an argument that “the Federal Rules of Evidence were enacted to address accuracy”).

  20. [20]. See infra Part III.A.

  21. [21]. See infra Parts III–V.

  22. [22]. Stein, supra note 19, at 2–12.

  23. [23]. See generally George L. Priest, The Common Law Process and the Selection of Efficient Rules, 6 J. Legal Stud. 65 (1977) (discussing the effects of evidentiary and procedural rulemaking in the pre-Federal Rules of Evidence era).

  24. [24]. See, e.g., Edmund M. Morgan & John MacArthur Maguire, Looking Backward and Forward at Evidence, 50 Harv. L. Rev. 909, 924 (1937) (“The common law has amply tried its little-drops-of-water-little-grains-of-sand and has amply failed.”). Several legal scholars wrote harsh, contemporaneous critiques of the common law development of evidentiary rules. See, e.g., Mason Ladd, A Modern Code of Evidence, 27 Iowa L. Rev. 213, 218 (1942) (“A review of the history of evidence, with its spotted and often accidental growth, is persuasive proof of the need of introspective study of the law of evidence with a view to far-reaching improvement.”); Morgan & Maguire, supra (“[A] picture of the hearsay rule with its exceptions would resemble an old-fashioned crazy quilt made of patches cut from a group of paintings by cubists, futurists and surrealists.”). For a modern overview, see Josh Camson, History of the Federal Rules of Evidence, A.B.A.: Litig. News, https://apps.americanbar.
    org/litigation/litigationnews/trial_skills/061710-trial-evidence-federal-rules-of-evidence-history.html (last visited Jan. 20, 2018).

  25. [25]. See Stephen A. Saltzburg, The Federal Rules of Evidence and the Quality of Practice in Federal Courts, 27 Clev. St. L. Rev. 173, 193–94 (1978).

  26. [26]. Id. at 181–83. Although the Supreme Court circulated drafts of a unified Federal Rules of Evidence nearly every year starting in 1969, Congress would not codify the Federal Rules until 1975. Id. Largely because of the political fallout from the Watergate scandal during the Nixon administration, Congress made substantial revisions beforehand, particularly to address lawmakers’ concerns regarding the rules governing privilege. See, e.g., Josh Camson, History of the Federal Rules of Evidence, American Bar Ass’n,
    litigationnews/trial_skills/061710-trial-evidence-federal-rules-of-evidence-history.html (last visited Jan. 20, 2018).

  27. [27]. See Robert H. Aronson, The Federal Rules of Evidence: A Model for Improved Evidentiary Decisionmaking in Washington, 54 Wash. L. Rev. 31, 42–43 (1978) (noting that juries are traditionally composed of non-legal experts and discussing the federal common law’s attitude toward jurors). See generally Stein, supra note 19 (discussing decisional accuracy as a primary goal of the newly codified Federal Rules of Evidence).

  28. [28]. See generally John Henry Wigmore et al., Evidence in Trials at Common Law (1961) (explaining age and speech requirements under the common law).

  29. [29]. See, e.g., Frederick Schauer, On the Supposed Jury-Dependence of Evidence Law, 155 U. Pa. L. Rev. 165, 178 n.53, 179 n.60 (2006) (noting “th[e] weakening [of the common law authentication rules] now enshrined in the Federal Rules of Evidence,” including the requirement that a subscribing witness testify in court to authenticate all admissible writings).

  30. [30]. See, e.g., Joseph H. Levie, Hearsay and Conspiracy: A Reexamination of the Co-Conspirators’ Exception to the Hearsay Rule, 52 Mich. L. Rev. 1159, 1159–61 (1954) (discussing the “weakening [of] the hearsay rule,” and describing that weakening as “a quiet steady process long under way and hardly commented on”); Morgan & Maguire, supra note 24, at 910 (observing that “the hearsay rule has been judicially liberalized by the creation of new exceptions,” and noting that “each new exception is a fertile mother of appellate litigation”).

  31. [31]. The common law featured the “voucher rule” from England, which forbade a party from impeaching its own witnesses unless four special circumstances were met. See, e.g., Chambers v. Mississippi, 410 U.S. 284, 296–98 (1973) (rejecting Mississippi’s “voucher rule,” in an era predating the Federal Rules of Evidence, when it violated the defendant’s constitutional right to cross-examination under the Sixth Amendment’s Confrontation Clause). The Federal Rules of Evidence explicitly removes the voucher rule. See Fed. R. Evid. 607 and advisory committee’s note.

  32. [32]. See Aronson, supra note 27, at 59–62 (discussing the “[a]ccommodation of [a]dvances in [s]cientific or [s]ocio-[p]sychological [k]nowledge” in the Federal Rules of Evidence).

  33. [33]. But see John H. Langbein, Understanding the Short History of Plea Bargaining, 13 L. & Soc’y Rev. 261, 263–64 (1979) (describing the rise of plea bargaining and attributing it, in part, to developments in the law of evidence that created “vast complexity in[] modern criminal trials,” diminishing the workability of the trial process).

  34. [34]. See generally Jon R. Waltz, Judicial Discretion in the Admission of Evidence Under the Federal Rules of Evidence, 79 Nw. U. L. Rev. 1097 (1984) (noting that judicial attitudes toward increasingly sophisticated jurors have softened such that the Federal Rules relax particularly onerous requirements that are deemed less necessary in modern times to reach accurate verdicts).

  35. [35]. See, e.g., Edward J. Imwinkelried, The Use of Evidence of an Accused’s Uncharged Misconduct to Prove Mens Rea: The Doctrines Which Threaten to Engulf the Character Evidence Prohibition, 51 Ohio St. L.J. 575, 577–78 (1990) (discussing loopholes created by the codified scheme); see also David P. Leonard, The Use of Character to Prove Conduct: Rationality and Catharsis in the Law of Evidence, 58 U. Colo. L. Rev. 1, 21–25 (1986) (discussing the character rule’s increased complexities).

  36. [36]. See, e.g., Edward J. Imwinkelried, The Silence Speaks Volumes: A Brief Reflection on the Question of Whether It Is Necessary or Even Desirable to Fill the Seeming Gaps in Article VI of the Federal Rules of Evidence, Governing the Admissibility of Evidence Logically Relevant to the Witness’s Credibility, 1998 U. Ill. L. Rev. 1013, 1018–22 (discussing gaps in the Federal Rules regarding the impeachment and examination of witnesses).

  37. [37]. See generally Edward R. Becker & Aviva Orenstein, The Federal Rules of Evidence After Sixteen Years—The Effect of “Plain Meaning” Jurisprudence, the Need for an Advisory Committee on the Rules of Evidence, and Suggestions for Selective Revision of the Rules, 60 Geo. Wash. L. Rev. 857 (1992) (discussing circuit splits, gaps, and inconsistencies involving FRE 103, 201, 301, 407, 410, 606, 703, 801, 803, and 804).

  38. [38]. This is so despite guidance from the United States Supreme Court that the Federal Rules of Evidence should be construed by using “the ‘traditional tools of statutory construction.’” Beech Aircraft Corp. v. Rainey, 488 U.S. 153, 163 (1988) (quoting INS v. Cardoza-Fonseca, 480 U.S. 421, 446 (1987)); See generally Edward J. Imwinkelried, A Brief Defense of the Supreme Court’s Approach to the Interpretation of the Federal Rules of Evidence, 27 Ind. L. Rev. 267 (1993) [hereinafter Imwinkelried, A Brief Defense] (explaining the benefits and drawbacks of such an approach); Edward J. Imwinkelried, Whether the Federal Rules of Evidence Should Be Conceived as a Perpetual Index Code: Blindness Is Worse than Myopia, 40 Wm. & Mary L. Rev. 1595 (1999) (same).

  39. [39]. See Daniel J. Capra, Case Law Divergence from the Federal Rules of Evidence 1–2 (2000),

  40. [40]. Id. at 1. Another scholar framed the issue this way: “Sometimes [advocates] must look beyond the words of the Rules to understand evidentiary doctrine. We must do so when the Rules are not definitive or are ambiguous ... but sometimes even when the text is clear ... .” Randolph N. Jonakait, Text, Texts, or Ad Hoc Determinations: Interpretation of the Federal Rules of Evidence, 71 Ind. L.J. 551, 571 (1996).

  41. [41]. See Capra, supra note 39, at 3, 17.

  42. [42]. See id. at 1–2. For representative examples, see generally United States v. Winchenbach, 197 F.3d 548 (1st Cir. 1999) (holding that FRE 608(b)’s ban on extrinsic evidence to prove a witness’s credibility vis-à-vis her character for truthfulness excludes evidence for bias, impeachment by contradiction, and prior inconsistent statements despite the Rule’s language suggesting that the Rule reaches all forms of credibility challenges); Palmquist v. Selvik, 111 F.3d 1332 (7th Cir. 1997) (extending to civil defendants FRE 404(a)’s exception to the ban on introducing propensity evidence to prove character despite the language of the Rule allowing such testimony from the accused only); Schultz v. Butcher, 24 F.3d 626 (4th Cir. 1994) (limiting arguments regarding unfair prejudice and confusion of the evidence to jury trials only, despite language in FRE 403 that allows advocates to make such arguments at bench trials); United States v. Sutton, 41 F.3d 1257 (8th Cir. 1994) (reverting to the common law rule requiring foundation for impeaching a witness’s credibility through her prior inconsistent statement, even though FRE 613(b) does not require it); United States v. Schoenborn, 4 F.3d 1424 (7th Cir. 1993) (holding that an adopter of a statement who does not have personal knowledge of its contents can create an admissible past recollection recorded under FRE 803(5) even though the Rule’s language states that the witness must have created or adopted the record with personal knowledge); United States v. Tarantino, 846 F.2d 1384 (D.C. Cir. 1988) (reinterpreting the rule of completeness under FRE 106 to apply only to oral statements and not documentary evidence despite the Rule’s language that includes writings and other personally recorded accounts).

  43. [43]. See Fed. R. Evid. 407 (stating that, subject to certain exceptions, “[w]hen measures are taken that would have made an earlier injury or harm less likely to occur, evidence of the subsequent measures is not admissible to prove: negligence[,] culpable conduct[,] a defect in a product or its design[,] or a need for a warning or instruction”).

  44. [44]. See id. at advisory committee’s note (discussing the “broad” reach of FRE 407 and stating that the “ground for exclusion rests on a social policy of encouraging people to take” remedial steps, instead of limiting the Rule’s reach to defendants).

  45. [45]. See, e.g., Louisville & Nashville R.R. v. Williams, 370 F.2d 839, 843–45 (5th Cir. 1966) (holding that third-party repairs are not covered under the subsequent remedial measures doctrine); Steele v. Wiedemann Mach. Co., 280 F.2d 380, 382 (3d Cir. 1960) (same).

  46. [46]. See, e.g., O’Dell v. Hercules, Inc., 904 F.2d 1194, 1204 (8th Cir. 1980) (“An exception to Rule 407 is recognized for evidence of remedial action mandated by superior governmental authority or undertaken by a third party because the policy goal of encouraging remediation would not necessarily be furthered by exclusion of such evidence.”). These courts reason that the defendant was acting as an agent of the third party, and (implicitly) that the third party’s actions nonetheless provide valuable information regarding the state of the allegedly dangerous condition before the plaintiff’s alleged injury.

  47. [47]. See Louisville, 370 F.2d at 844.

  48. [48]. See id. at 844–45.

  49. [49]. See Fed. R. Evid. 704(b) (“In a criminal case, an expert witness must not state an opinion about whether the defendant did or did not have a mental state or condition that constitutes an element of the crime charged or of a defense.”).

  50. [50]. Id.

  51. [51]. See, e.g., United States v. Gastiaburo, 16 F.3d 582, 587–89 (4th Cir. 1994) (carving out an exception for testimony by law enforcement experts by (1) noting that “Rule 704(b) was enacted in the wake of the attempted assassination of President Reagan and the murder of John Lennon, and was an attempt to constrain psychiatric testimony on behalf of defendants asserting the insanity defense,” and (2) arguing that “[t]he application of the same rule in an entirely different context—a law enforcement officer’s expert opinion testimony on behalf of the government at the trial of an alleged drug dealer—is murky at best”).

  52. [52]. Id.

  53. [53]. See, e.g., United States v. Williams, 980 F.2d 1463, 1466 (D.C. Cir. 1992) (allowing into evidence a law enforcement expert’s responses to hypothetical questions that matched the facts of the case at issue).

  54. [54]. Cf. Dana R. Hassin, How Much is Too Much? Rule 704(b) Opinions on Personal Use vs. Intent to Distribute, 55 U. Miami L. Rev. 667, 671–73 (2001) (observing that FRE 704(b) is an accuracy-enhancing rule insofar as it “ensur[es] that the jury does not place undue weight on expert testimony” and arguing that FRE 704(b) should apply to police officer testimony as well, because accuracy is threatened “when law enforcement agents testify as expert witnesses because there is the potential that juries may unduly grant more weight to the agent’s testimony”). As of 2017, we have located no cases in which a court has argued, alternatively, that the justification for FRE 704(b) is somehow absent when the mental state testimony at issue originates from police officers instead of mental health experts.

  55. [55]. See Fed. R. Evid. 803(3) (“The following [is] not excluded by the rule against hearsay[:] ... A statement of the declarant’s then-existing state of mind (such as motive, intent, or plan) ... .”).

  56. [56]. Compare United States v. Jenkins, 579 F.2d 840, 843 (4th Cir. 1978) (holding that statements of future intent cannot prove the subsequent behavior of anyone other than the declarant), with United States v. Delvecchio, 816 F.2d 859, 863 (2d Cir. 1987) (holding that such statements are admissible to prove the actions of third parties if corroboration is present).

  57. [57]. See Fed. R. Evid. 410 (listing this general rule and its exceptions).

  58. [58]. The text of FRE 410(b) states only that the court may admit evidence of pleas, plea discussions, and related statements:

    (1) in any proceeding in which another statement made during the same plea or plea discussions has been introduced, if in fairness the statements ought to be considered together; or
    (2) in a criminal proceeding for perjury or false statement, if the defendant made the statement under oath, on the record, and with counsel present.

    Id. R. 410(b).

  59. [59]. See Trading Techs. Int’l, Inc. v. BCG Partners, Inc., No. 10 C 715 (consolidated), 2011 WL 3946581, at *1 (N.D. Ill. Sept. 2, 2011) (holding that FRE 408 “is not waivable—a party cannot introduce evidence of its own settlement offers”). See generally Fed. R. Evid. 408 (governing civil settlement discussions).

  60. [60]. See, e.g., United States v. Mezzanatto, 513 U.S. 196, 210 (1995) (“[A]bsent some affirmative indication that the agreement was entered into unknowingly or involuntarily, an agreement to waive the exclusionary provisions of the plea-statement Rules is valid and enforceable.”); see also United States v. Burch, 156 F.3d 1315, 1321 (D.C. Cir. 1998) (admitting such statements for the purpose of impeachment).

  61. [61]. See, e.g., Paul F. Kirgis, A Legisprudential Analysis of Evidence Codification: Why Most Rules of Evidence Should Not Be Codified—but Privilege Law Should Be, 38 Loy. L.A. L. Rev. 809, 816 (2004) (noting “the need for a uniform system of evidence law to replace the patchwork of federal and state laws then in effect”).

  62. [62]. See Imwinkelried, A Brief Defense, supra note 38, at 267–72 (describing the various approaches to statutory interpretation taken by the courts).

  63. [63]. For a thorough application of libertarian ethics to legal advocacy, see William H. Simon, Ethical Discretion in Lawyering, 101 Harv. L. Rev. 1083, 1083 (1988) for a rejection of categorical ethics rules and arguing that “lawyers should exercise judgment and discretion in deciding what clients to represent and how to represent them.”

  64. [64]. Id.

  65. [65]. See, e.g., Perlman, supra note 18, at 1639 n.1 (2015) (citing David A. Binder et al., Lawyers as Counselors: A Client-Centered Approach (3d ed. 2011) (citations omitted) (“Some legal ethicists argue that the lawyer should comply with the client’s instruction and pursue every permissible tactic, that is, go right up to the line but not cross it.”). Deontological ethics is defined as “normative theories regarding which choices are morally required, forbidden, or permitted.” Deontological Ethics, Stan. Encyclopedia of Phil. (Oct. 17, 2016), https://plato.

  66. [66]. For a review of the deontological view, see generally Freedman & Smith, supra note 18; Wendel, supra note 18.

  67. [67]. See infra Part III.A.

  68. [68]. See generally John R. Schermerhorn et al., Organizational Behavior (12th ed. 2012) (discussing “interactional legitimacy” between social actors); Carl Schmitt, Legality and Legitimacy (Jeffrey Seitzer ed. & trans., 2004) (discussing governmental legitimacy).

  69. [69]. John Locke: Critical Assessments 524 (Richard Ashcraft ed., 1991); see also John Rawls, A Theory of Justice 135 (Otfried Höffe ed., Joost den Haan trans., 2013) (discussing a similar theory of legitimacy).

  70. [70]. See Joseph Raz, The Morality of Freedom 80 (1986); see also Tom R. Tyler, Why People Obey the Law 29 (2006) (explaining the conditions under which citizens will voluntarily confer legitimacy onto legal institutions and the laws they enact).

  71. [71]. Max Weber, Politics as a Vocation, in From Max Weber: Essays in Sociology 77, 79 (H. H. Gerth & C. Wright Mills eds. & trans., 1991). This concept is sometimes referred to as “civil legitimacy.” See, e.g., Rodney D. Peterson, Political Economy and American Capitalism 47 (2012) (“[C]ivil legitimacy refers to a system of government based on agreement between equally autonomous constituents who have combined to cooperate toward some common good.” (citations omitted)).

  72. [72]. See John Rawls, Political Liberalism 121 (1993) (suggesting that political institutions that lack legitimacy exercise their power unjustifiably and will not be obeyed).

  73. [73]. See, e.g., John Thibaut & Laurens Walker, Procedural Justice: A Psychological Analysis 7 (1975) (theorizing that people view governmental actions that are instrumental to the individual’s attainment of social goods as legitimate).

  74. [74]. See generally Florian Weigand, Investigating the Role of Legitimacy in the Political Order of Conflict-Torn Spaces (London Sch. of Econ., Working Paper No. SiT/WP/04/15, 2015), http://www. (suggesting both force and legitimacy are sources underpinning obedience to social control).

  75. [75]. See Tom R. Tyler, The Psychology of Legitimacy: A Relational Perspective on Voluntary Deference to Authorities, 1 Personality & Soc. Psychol. Rev. 323, 325 (1997) (comparing resource-based and relation-based models of legitimacy); see also Jeffrey Fagan, Legitimacy and Criminal Justice, 6 Ohio St. J. Crim. L. 123, 138 (2008) (calling for a restructuring of the criminal justice system due to popular dissatisfaction in communities where poverty and crime intersect, in an attempt to restore legitimacy to the system in these areas).

  76. [76]. See Tom R. Tyler, The Psychology of Procedural Justice: A Test of the Group-Value Model, 57 J. Personality & Soc. Psychol. 830, 836–38 (1989) (conducting experiments and finding that the neutrality of the decisionmaking process, trust in the decision maker, and the information conveyed to an individual regarding her social standing influence perceptions of governmental legitimacy). Other researchers have replicated these effects. See, e.g., Heather J. Smith et al., The Self-Relevant Implications of the Group-Value Model: Group Membership, Self-Worth, and Treatment Quality, 34 J. Experimental Soc. Psychol. 470, 489–90 (1998) (“People care about treatment quality because fair and reasonable treatment by authorities communicates to them that they are respected and valued.”); Fátima H. Sousa & Jorge Vala, Relational Justice in Organizations: The Group-Value Model and Support for Change, 15 Soc. Just. Res. 99, 117–18 (2002) (finding respondents stressed “consistency” and “[the ability] to obtain a result according to the solution found for others in the same circumstances” as valuable aspects of justice).

  77. [77]. Tom R. Tyler & Justin Sevier, How Do the Courts Create Popular Legitimacy?: The Role of Establishing the Truth, Punishing Justly, and/or Acting Through Just Procedures, 77 Alb. L. Rev. 1095, 1097 (2014).

  78. [78]. See Tyler, The Psychology of Procedural Justice, supra note 76, at 830.

  79. [79]. Id.

  80. [80]. See generally Rebecca Hollander-Blumoff, The Psychology of Procedural Justice in the Federal Courts, 63 Hastings L.J. 127 (2011) (discussing these principles in detail).

  81. [81]. See generally Or Bassok, The Sociological-Legitimacy Difficulty, 26 J.L. & Pol. 239 (2011) (focusing on the decreased legitimacy the Supreme Court may face if the public begins to perceive its decisions as political in nature); Or Bassok, The Supreme Court’s New Source of Legitimacy, 16 U. Pa. J. Const. L. 154 (2013) (claiming that advances in scientific public polling afford the Supreme Court a new source of legitimacy in an era in which it is more difficult for the Court to derive legitimacy based on its perceived expertise in the law).

  82. [82]. See John Thibaut & Laurens Walker, A Theory of Procedure, 66 Calif. L. Rev. 541, 541 (1978) (proposing “a general framework for analyzing and classifying all conflict resolution procedures” and recognizing a “fundamental dichotomy between the potential dispute resolution objectives of ‘truth’ and ‘justice’”).

  83. [83]. See Justin Sevier, Redesigning the Science Court, 73 Md. L. Rev. 770, 795–96 (2014). The Cartesian plane in the figure above/below first appeared in this article. Id. at 795.

  84. [84]. Specifically, one study found that groups of laypeople who read an abbreviated version of a criminal trial perceived that the American adversarial system for resolving the dispute produced significantly greater procedural fairness than factually accurate results. Conversely, participants who read about the same trial, but who believed that the procedures resembled the inquisitorial style of civil law countries (in which the judge acts as a central investigator and the parties do not present the evidence) perceived that the trial produced greater accuracy but at the cost of significantly less procedural fairness. Justin Sevier, The Truth-Justice Tradeoff: Perceptions of Decisional Accuracy and Procedural Justice in Adversarial and Inquisitorial Legal Systems, 20 Psychol., Pub. Pol’y, & L. 212, 220–21 (2014). The tug-of-war relationship between the public’s perceptions of the decisional accuracy of the courts and the fairness of the procedures that they provide has been replicated in other studies as well. See, e.g., Justin Sevier, Redesigning the Science Court, 73 Md. L. Rev. 770, 808–10 (2014) (replicating the tradeoff between perceptions of accuracy and fairness in different dispute adjudication paradigms).

  85. [85]. Tyler & Sevier, supra note 77, at 1097–98.

  86. [86]. Id. at 1098.

  87. [87]. Tyler, supra note 70, at 100–10.

  88. [88]. Id.; see also Sevier, supra note 84, at 220–21 (reporting similar findings).

  89. [89]. Tyler & Sevier, supra note 77, at 1111.

  90. [90]. See id. at 1029.

  91. [91]. Id. at 1097.

  92. [92]. Id. (“To some degree court legitimacy is linked to the attainment of truth and the enactment of just punishments and using fair procedures is important because it is viewed as leading to these goals.”).

  93. [93]. See, e.g., D. Michael Risinger, Searching for Truth in the American Law of Evidence and Proof, 47 Ga. L. Rev. 801, 803 (2013) (“[T]ruth and justice are the official ends of the trial system, from first investigation to last post-trial proceeding.”). For example, the Article VI rules governing impeachment evidence allow attorneys to test the accuracy of a witness’s testimony by challenging the witness’s testimonial capacities for truthfulness, perception, narration, and memory. Fed. R. Evid. 601–15. But the Rules also limit the manner in which attorneys can test the witness’s accuracy with respect to collateral issues, explicitly because of fairness concerns. Id. Similarly, the Article VIII rules governing hearsay ban the use of such evidence generally, and allow exceptions for certain types of hearsay, on the theory that both policies will improve the odds that the fact finder reaches an accurate verdict on the facts. Id. RR. 801–07. But Article VIII also allows other types of hearsay including Rule 804’s “unavailability exceptions”—which do not necessarily enhance the likelihood of accurate verdicts—explicitly on a theory of equity and fairness. Id. Rules governing the use of character evidence and privileged information similarly attempt to balance the need to improve the odds of accurate verdicts while increasing the public’s perceptions of the trial’s procedural fairness. Id. RR. 404–06, 501–02.

  94. [94]. Justin Sevier, Vicarious Windfalls, 102 Iowa L. Rev. 651, 665–66 (2017) (footnotes omitted).

  95. [95]. See, e.g., F.J. Ponseti et al., Fair Play, Cheating and Gamesmanship in Young Basketball Teams, 5 J. Physical Educ. & Health 29, 29 (2016) (“These behaviours can be defined as the intention to intimidate or attack the opponent through physical contact, or the intention to deceive and put the opponent at a disadvantage through the use of gamesmanship, including faking injury, wasting time, or trying to unnerve the opponent.”). The authors explain that “[a]ll these behaviours result in negative consequences for the opponent and reflect an absence or diminution of fair play.” Id. The term “gamesmanship” in the psychological literature (and in this Article) is meant to be descriptive instead of normative. The research seeks to understand what gaming behavior is, when it is used, and how it is perceived by others without assigning either moral opprobrium or absolution to the term.

  96. [96]. See, e.g., Mary Hassandra, Marios Goudas et al., A Fair Play Intervention Program in School Olympic Education, 22 European J. of Psychol. Ed. 99, 103 (2007) (describing cheating and gamesmanship as antisocial behaviors, but noting that “‘[g]amesmanship’, unlike cheating, does not involve violating the rules of the game in the hope of avoiding detection, but it is the use of legal, however, morally dubious, designed tactics to unsettle opponents.”).

  97. [97]. See infra note 117 and accompanying text.

  98. [98]. See generally Maria Konnikova, Inside the Cheater’s Mind, New Yorker (Oct. 31, 2013), (discussing the major theories of cheating and noting the shift between early theories of cheating as linked to deficits in moral development and later theories focusing on social situational factors that foster cheating).

  99. [99]. See Christoph Kogler et al., Testing the “Slippery Slope Framework” Among Self-Employed Taxpayers, 16 Econ. Governance 125, 126–27 (2015).

  100. [100]. Miguel Roig & Amanda Marks, Attitudes Toward Cheating Before and After the Implementation of a Modified Honor Code: A Case Study, 16 Ethics & Behav. 163, 164 (2006) (“[S]tudents who perceive that their peers are reluctant to cheat are themselves less likely to engage in such behaviors, and therefore, a culture of integrity is established.”).

  101. [101]. See Ponseti et al., supra note 95, at 29 (discussing cheating and fair play in youth basketball).

  102. [102]. Kelly Campbell, Why People Cheat, Psychol. Today (Mar. 26, 2014), https://www.

  103. [103]. See Shaul Shalvi et al., Self-Serving Justifications: Doing Wrong and Feeling Moral, 24 Current Directions Psychol. Sci. 125, 126–29 (2015) (expounding on the antecedents and consequences of cheating); see also Natalie Angier, The Urge to Punish Cheats: It Isn’t Merely Vengeance, N.Y. Times (Jan. 22, 2002), (“The [c]heat is [a] transgressor of fair play, the violator of accepted norms, the sneak who smiles with Chiclet teeth while ladling from the community till.”).

  104. [104]. For a real-world illustration of this principle, see Konnikova, supra note 98 (recounting the story of twenty Long Island teens who, because of their wealth, were able to purchase the services of an individual who took standardized academic tests on their behalf and procured outstanding test scores).

  105. [105]. See Roy F. Baumeister & Brad J. Bushman, Social Psychology and Human Nature 518–19 (2014) (discussing cheating in the context of altruistic punishment).

  106. [106]. Robert Boyd et al., Coordinated Punishment of Defectors Sustains Cooperation and Can Proliferate When Rare, 328 Sci. 617, 617 (2010).

  107. [107]. See id.

  108. [108]. Eric Michael Johnson, Punishing Cheaters Promotes the Evolution of Cooperation, Sci. Am.: The Primate Diaries (Aug. 16, 2012),

  109. [109]. See id.; see also Boyd et al., supra note 106, at 619.

  110. [110]. Joris Lammers et al., Power Increases Hypocrisy: Moralizing in Reasoning, Immorality in Behavior, 21 Psychol. Sci. 737, 742 (2010).

  111. [111]. In one such study, researchers told participants that they would engage in a “learning task” with a partner, in which they were required to punish their partner for her mistakes by administering a blast of white noise into the partner’s headset. Jeffrey C. Savitsky & James Babl, Cheating, Intention, and Punishment from an Equity Theory Perspective, 10 J. Res. Personality 128, 129 (1976). In the study, half of the partners, “who w[ere] actually confederate[s] of the experimenter,” cheated on the learning task in a way that was apparent to the participant. Id. Additionally, the experimenters manipulated the purpose behind the partner’s cheating (and therefore the degree to which the participants would empathize with the confederate): either to benefit the participant, the partner, or a charity that the partner supported. Id. The results were consistent with general principles of moral attitudes toward rule breaking; participants administered greater white noise (and therefore greater discomfort) for the mistakes of cheating partners compared to non-cheating partners. Id. at 132. But the noise blasts also varied depending on the objective of the partner’s cheating: the blasts were significantly less severe when the cheating benefitted a benevolent charity and were most severe when they benefitted the partner who cheated. See id. at 133, 135.

  112. [112]. See, e.g., Tamera Burton Murdock et al., Effects of Classroom Context on University Students’ Judgments About Cheating: Mediating and Moderating Processes, 10 Soc. Psychol. Educ. 141, 162–63 (2007) (examining scenarios in which cheating is seen to be more or less acceptable in an academic setting).

  113. [113]. See Savitsky & Babl, supra note 111, at 134.

  114. [114]. See id. at 135 (finding that subjects of an experiment punished cheaters more harshly when the subject stood to lose from the cheating. In contrast, the experiment found “[i]t is possible that subjects ignored the presence of cheating if they themselves expected to gain by its presence”).

  115. [115]. See Martin J. Lee et al., Development of the Attitudes to Moral Decision-Making in Youth Sport Questionnaire (AMDYSQ), 8 Psychol. Sport & Exercise 369, 372 (2007) (“[T]here appear to be other conceptually similar categories of behaviour that occur in sport by which players violate the implicit nature of the contract to compete fairly and seek to gain a ‘dishonourable’ advantage.”).

  116. [116]. See, e.g., Erich Kirchler, The Economic Psychology of Tax Behaviour 64–71 (2007) (discussing social and societal norms in the context of tax loopholes); Michael C. Dorf, FindLaw Forum: Americans Believe Lawyers to Be Necessary but Dishonest, Survey Finds, CNN (Apr. 17, 2002, 12:11 PM), (discussing, as an example, public perceptions of the exclusionary rule, in “which courts refuse to admit evidence [that was] obtained in violation of the Fourth Amendment”).

  117. [117]. See, e.g., Luigi Guiso et al., The Determinants of Attitudes Towards Strategic Default on Mortgages 3 (Eur. U. Inst. Florence, Paper No. 2010/31, 2013) (noting that, among other things, individual attitudes regarding fairness and morality affect their views of the acceptability of strategically defaulting on a mortgage); Mary Hassandra et al., Attitudes Towards Fair Play in Physical Education: The Role of Intrinsic Motivation and Gender, XIth Eur. Congress Sport Psychol. (2002),
    gender (last visited Oct. 30, 2017) (finding gender differences and differences in intrinsic motivation with respect to attitudes toward gamesmanship).

  118. [118]. Pere Palou et al., Acceptance of Gamesmanship and Cheating in Young Competitive Athletes in Relation to the Motivational Climate Generated by Parents and Coaches, 117 Perceptual & Motor Skills 290, 299–300 (2013).

  119. [119]. See Ponseti et al., supra note 95, at 29, 31–32.

  120. [120]. See Palou, supra note 118, at 298–99.

  121. [121]. See supra note 105 and accompanying text.

  122. [122]. Id.

  123. [123]. See generally Tyler, supra note 70 (discussing tenets of procedural justice and linking procedural justice explicitly to psychological perceptions of fair process).

  124. [124]. The arguments and claims in this Article are mine. I use the word “we” to acknowledge the work of the research assistants and others who assisted the author in designing the study and interpreting the results.

  125. [125]. MTurk is an inexpensive platform for collecting high-quality data from a representative sample of the population. See, e.g., Adam J. Berinsky et al., Evaluating Online Labor Markets for Experimental Research:’s Mechanical Turk, 20 Pol. Analysis 351, 366 (2012) (replicating prior experimental work with mTurk samples and concluding that the mTurk platform produces quality data); Michael Buhrmester et al., Amazon’s Mechanical Turk: A New Source of Inexpensive, Yet High-Quality, Data?, 6 Persp. Psychol. Sci. 3, 5 (2011) (describing the data collected from mTurk participants as comparable in quality to data collected via traditional methods); Winter Mason & Siddharth Suri, Conducting Behavioral Research on Amazon’s Mechanical Turk, 44 Behav. Res. Methods 1, 2–4 (2012) (describing mTurk as a useful tool for collecting data for behavioral research).

  126. [126]. See infra Table 1.

  127. [127]. See infra Table 1.

  128. [128]. See infra Table 1.

  129. [129]. See infra Table 1.

  130. [130]. See infra Table 1.

  131. [131]. The impeachment condition implicated FRE 609, which regulates the conditions under which a witness’s prior conviction can be entered into evidence. Fed. R. Evid. 609. The hearsay condition implicated FRE 803, which governs the admissibility of certain types of hearsay statements when the declarant is either available or unavailable as defined by FRE 804. Id.
    RR. 803–04. The propensity condition implicated FRE 404(b), which governs the purposes, apart from propensity, whereby information that initially appears to be character evidence is admissible. Id. R. 404(b). The privilege condition implicated the common law privilege against the admissibility of statements made to a mental health professional, such as a psychotherapist. See, e.g., Jaffee v. Redmond, 518 U.S. 1, 8–15 (1996) (recognizing the privilege and discussing its contours).

  132. [132]. In the impeachment condition, the defense attorney argued that a conviction for lying to a law enforcement officer would be presumed inadmissible under FRE 609, because the conviction would have occurred more than ten years before the current trial. Fed. R. Evid. 609(b). The defense attorney argued that allowing the prosecutor to enter into evidence the arrest—which did not result in an adjudication on the merits—should therefore also be presumed inadmissible. The prosecutor countered that FRE 609 does not apply (because there was no conviction), and instead the judge should look to FRE 608, which does not explicitly disallow questions about arrests. See id. RR. 608–09. In the hearsay condition, the defense attorney argued that allowing the prosecutor to present the excited utterance made by the defendant’s sister would prevent him from cross-examining the sister to determine whether she was confused, mistaken, or joking when she made the statement (if she truly made the statement at all). The defense attorney also noted that the defendant’s sister was seated in the courtroom gallery and that the prosecution should not be permitted to avoid calling her as a witness by using her husband to admit her statements into evidence. The prosecutor argued that FRE 803 plainly allows non-testimonial hearsay evidence that qualifies as an excited utterance regardless of the availability of the witness who uttered the statement. See id. R. 803(2). In the propensity evidence condition, the defense attorney argued that the reason the prosecutor wished to question the defendant about his prior arrest for cocaine distribution was to show that the defendant’s relevant character trait is that he is a lawbreaking drug dealer, and so he was dealing drugs on this occasion (which led to the murder). The defense attorney argued that this is exactly the propensity inference that FRE 404 forbids fact finders from making, because defendants should be convicted or acquitted based on their (proven) actions instead of their character. See id. R. 404(a)(1). The prosecutor countered that she was not using the evidence to prove the defendant’s propensity for dealing drugs or to prove that he is the “type” of person who would commit the crime of which he is accused. Rather, the prosecutor argued, the evidence is proffered pursuant to FRE 404(b), to prove something other than the defendant’s general propensity to break the law: specifically, it shows his intent to deal drugs here, which provides the context within which the murder occurred. See id. R. 404(b)(2). Finally, in the privilege condition, the defense attorney argued that the prosecutor was attempting to elicit information from the counselor that the courts have deemed can be withheld pursuant to privacy principles and pursuant to a desire to encourage honesty and disclosure in certain types of personal relationships without fear that the communications will become public record in open court. The defense attorney further argued that the defendant’s alleged statements to his counselor regarding any alleged drug use constitute privileged information, akin to the privilege for psychotherapists, married couples, priests, and attorneys. See id. R. 501 (declaring that “the common law ... governs a claim of privilege” in criminal trials); see also Jaffee, 518 U.S. at 1 (creating a “psychotherapist-patient” privilege under the Federal Rules of Evidence). The prosecutor argued that it is debatable whether the counselor is truly a “psychotherapist” for purposes of the evidentiary privilege and, in any event, unlike the attorney-client privilege, the privilege for psychotherapists is not absolute; rather, it is a qualified privilege that can be overcome by the adversary’s compelling need for the information. See Ann Bowen Poulin, The Psychotherapist-Patient Privilege in the Aftermath of Jaffee v. Redmond: Where Do We Go from Here?, 76 Wash. U. L.Q. 1341, 1362–73 (1998) (discussing the scope of the psychotherapist-patient privilege in the aftermath of the Jaffee decision). The prosecutor then argued that there was a compelling need in this situation, because the defendant’s alleged admissions of drug dealing constitute the crucial missing link in the prosecutor’s case.

  133. [133]. We also measured their impressions of the evidence against the defendant, the strength of the case against the defendant, their willingness to convict the defendant (had they served on the jury), and their verdict in the case (had they served on the jury).

  134. [134]. A Likert Scale is a psychometric scale that is routinely used in questionnaires and is analyzed as an ordinal variable (frequently a range from 1 to 7). See Robert M. Lawless et al., Empirical Methods in Law 172 (Vicki Been et al. eds., 2010).

  135. [135]. To measure how accurate and how fair the court’s decision was likely to be, we asked participants eight questions based on previously validated scales in the psychology literature. See, e.g., Jojanneke van der Toorn et al., More than Fair: Outcome Dependence, System Justification, and the Perceived Legitimacy of Authority Figures, 47 J. Experimental Soc. Psychol. 127, 130, 133 (2011) (explaining that the study used 9-point Likert Scales). These scales have been replicated and validated. See, e.g., Jason A. Colquitt & Jessica B. Rodell, Justice, Trust, and Trustworthiness: A Longitudinal Analysis Integrating Three Theoretical Perspectives, 54 Acad. Mgmt. J. 1183, 1191–93 (2011) (reporting research findings using, among other measures, validated procedural justice scales).

  136. [136]. See, e.g., Harold Hotelling, Analysis of a Complex of Statistical Variables into Principal Components, 24 J. Educ. Psychol. 417, 421 (1933) (explaining the tenets of a principal-component analysis); see also Marjorie A. Pett et al., Making Sense of Factor Analysis 149–61 (2006) (discussing the oblique factor rotation).

  137. [137]. Factors in a principal component analysis are meaningful if their statistical “eigenvalues” are greater than 1.0. Norman Cliff, The Eigenvalues-Greater-than-One Rule and the Reliability of Components, 103 Psychol. Bull. 276, 276 (1988). The two factors whose eigenvalues exceeded 1.0 jointly explained 82.31% of the variance in participants’ responses.

  138. [138]. The reliability of a psychometric scale is measured by a Cronbach’s alpha statistic ranging from 0.00 (lowest reliability) to 1.0 (highest reliability). See Jose M. Cortina, What Is Coefficient Alpha? An Examination of Theory and Applications, 78 J. Applied Psychol. 98, 98 (1993). The statistic tells us how well the items, when averaged together, measure the underlying psychological concept. Id. The Cronbach’s alpha value for the accuracy scale was 0.95, and the alpha value for the fairness scale was 0.89.

  139. [139]. The legitimacy items originally appeared in van der Toorn et al., supra note 135, at 127. Participants responded on a Likert scale from 1 (“not legitimate,” “very unwilling,” “unwilling,” “little confidence,” and “little trust”) to 7 (“legitimate,” “very willing,” “unjust,” “little confidence,” and “little trust”).

  140. [140]. Most importantly, the Cronbach’s alpha value for the legitimacy scale was 0.96. The morality scale had a Cronbach’s value of 0.96. Regarding the prosecution’s evidence, preliminary statistical tests indicated that participants found the hat evidence to be less persuasive than the other evidence, and they found the cocaine conviction to be more persuasive.

  141. [141]. Forty-four percent of participants voted to convict, whereas fifty-six percent voted to acquit.

  142. [142]. The bivariate correlation (0.77, p < .001) was strong and significant. Bivariate correlations range from -1.0 (a perfect negative relationship) to +1.0 (a perfect positive relationship). A bivariate correlation of 0.0 indicates no relationship. See, e.g., Earl R. Babbie, The Practice of Social Research 436–40 (12th ed.) (2009).

  143. [143]. The bivariate correlations were 0.51 (p < .001), 0.24 (p < .001), and 0.33 (p < .001), respectively. The statistics literature suggests these correlations range from low to high. See Jacob Cohen, A Power Primer, 112 Psychol. Bull. 155, 156 (1992) (reporting that the lower boundaries for small, medium, and large correlation coefficients are 0.10, 0.30, and 0.50, respectively).

  144. [144]. We performed a 4 (evidentiary trapdoor type: impeachment vs. hearsay vs. propensity vs. privilege) x 2 (trapdoor admissibility: admitted vs. excluded) x 2 (evaluation: accuracy scale vs. fairness scale) “mixed model” analysis of variance on participants’ responses to our questionnaires. As I have noted elsewhere:

    An analysis of variance (“ANOVA”) provides a statistical test of whether the means of several groups are equal. ANOVA results are represented by an F-statistic, and the sizes of the effects are represented by 103 3 Sevier Science Letter2p. Means are denoted by the letter “M” and standard deviations are denoted by the letters “SD.” Differences are denoted as “statistically significant” in this Article if the statistical tests indicate that the likelihood that the difference observed would occur by chance is 5% or less (as indicated by the p-value as p <0.05). A difference is “marginally significant” if the likelihood of seeing such a difference by chance is greater than 5% but less than 10%.

    Sevier supra note 94, at 677 n.102 (2017) (citations omitted).

  145. [145]. MAccuracy = 4.44 (SD = 1.43), MFairness = 4.90 (SD = 1.50); F(1, 236) = 29.34, p < .001, hp = .11.

  146. [146]. F(3, 236) = 1.25, p = .291, hp = .02.

  147. [147]. MAdnit = 4.42 (SD = 0.11), MExclude = 4.92 (SD = 0.11); F(1, 236) = 9.63, p = .002, hp = .04.

  148. [148]. F(1, 236) = 38.66, p < .001, hp = 14.

  149. [149]. MAdmit = 4.46 (SD = 1.40), MExclude = 4.42 (SD = 1.47); F(1, 236) = 0.04, p = .740, hp = .00.

  150. [150]. MAdmit = 4.39 (SD = 1.63), MExclude = 5.42 (SD = 1.13); F(1, 236) = 32.76, p < .001, hp = .12.

  151. [151]. To test our experimental hypotheses, we conducted a 4 (trapdoor type) x 2 (admissibility decision) “between-subjects” analysis of variance on participants’ perceptions of the legitimacy of the trial.

  152. [152]. Effect of admissibility decision: F(1, 236) = 11.99, p = .001, hp = .05; non-significant effect of the form of the evidence: F(3, 236) = 0.74, p = .530, h2p = .01.

  153. [153]. MAdmit = 4.50 (SD = 1.73), MExclude = 5.20 (SD = 1.44).

  154. [154]. Mediation analysis determines “when a predictor affects a dependent variable indirectly through at least one intervening variable, or mediator.” Kristopher J. Preacher & Andrew F. Hayes, Asymptotic and Resampling Strategies for Assessing and Comparing Indirect Effects in Multiple Mediator Models, 40 Behav. Res. Methods 879, 879 (2008). We performed this mediation using a linear regression analysis. This analysis reports unstandardized coefficients, “B,” and standard errors, “SE.” It determines whether the coefficients are statistically significant via a “student’s t” statistic. A linear regression examines the independent effects of independent variables on a continuous dependent variable. See Lawless et al., supra note 134, at 29, 300–31.

  155. [155]. B = 5.37, SE = 0.16, t = 33.16, p < .001.

  156. [156]. B = 0.63, SE = 0.03, t = 20.13, p < .001.

  157. [157]. B = 0.75, SE = 0.06, t = 13.01, p < .001.

  158. [158]. Total effect of admissibility decision on perceptions of legitimacy: B = -0.71, SE = 0.20,
    t = -3.47, p < .001. Total direct effect of admissibility decision on perceptions of legitimacy:
    B = 0.16, SE = 0.10, t = 1.65, p = .10 (marginally significant). This represents a 77.46% reduction of the direct effect of admissibility on perceptions of legitimacy that is explained by the mediating variables.

  159. [159]. Standardized indirect effect: B = -0.21, SE = 0.04, CI [-0.29, -0.14].

  160. [160]. Standardized indirect effect of accuracy pathway: B = 0.05, SE = 0.02, CI [0.02, 0.09].

  161. [161]. Asterisks in the mediation analysis indicate statistically significant associations.

  162. [162]. See generally Fagan, supra note 75 (discussing general tenets of legitimacy in criminal trials and perceptions of legitimacy in criminal trials).

  163. [163]. See Dorf, supra note 116 (discussing the public’s views of legal loopholes and their views of how attorneys—particularly defense attorneys—facilitate their use); see also Paul C. Giannelli, Daubert and Criminal Prosecutions, Crim. Just., Fall 2011, at 61, 62 (noting that empirical evidence suggests that criminal prosecutors have higher success rates of admitting expert scientific testimony than do criminal defendants and civil plaintiffs).

  164. [164]. There may be, however, a third outcome. Although participants will legitimize trials decided with trapdoor evidence less than they do other trials, this effect might be heightened when the prosecutor or plaintiff—the person who brought the action to court in the first place—uses trapdoor evidence, compared to the party who has been brought into court unwillingly.

  165. [165]. As in Study 1, participants were a representative sample from throughout the United States. See supra Part IV.A.1.

  166. [166]. See, e.g., QuickFacts: United States, U.S. Census Bureau,
    facts/fact/table/US/PST045216 (last visited Jan. 20, 2018) (listing the percentage of persons identifying as “White alone” as 23.10 as of July 1, 2016).

  167. [167]. We made this decision because Study 1 revealed that the evidentiary trapdoor effect was statistically significant regardless of whether the evidence took the form of impeachment, propensity, hearsay, or privilege.

  168. [168]. At the outset, we replicated several of our preliminary results from Study 1. We again created scales for our perceptions of accuracy and perceptions of fairness measures, which were statistically different from each other, and which were highly reliable measures of each psychological construct. We measured participants’ willingness to legitimize the tribunal in the same manner in which we measured it in Study 1, and we asked participants the same questions regarding the strength of the evidence and their verdicts in each case. As in Study 1, each of these measures was statistically reliable. Participants’ responses to these statistically reliable items were averaged to form several scales that we used in the remainder of our analyses in Study 2.

  169. [169]. We conducted a 2 (admissibility decision: admitted vs. excluded) x 2 (legal context: civil vs. criminal) x 2 (proffering actor: prosecutor/plaintiff vs. defendant) between-subjects analysis of variance on participants’ willingness to legitimize the legal tribunal.

  170. [170]. MCivil = 4.72 (SD = 1.42), MCriminal = 4.73 (SD = 1.54); F(1, 231) = 0.00, p = .966, hp = .00.

  171. [171]. MAdmit = 4.41 (SD = 1.52), MExclude = 5.04 (SD = 1.37); F(1, 231) = 11.75, p = .001, hp = .05.

  172. [172]. F(1, 231) = 7.81, p = .006, hp = .03.

  173. [173]. MProsecutor = 5.24 (SD = 1.33), MDefendant = 4.85 (SD = 1.39); F(1, 120) = 2.55, p = .113, hp = .02.

  174. [174]. MProsecutor = 4.08 (SD = 1.67), MDefendant = 4.73 (SD = 1.30); F(1, 115) = 5.46, p = .021, hp = .05.

  175. [175]. See infra Part III.A.

  176. [176]. The mediated moderation proceeds in a series of regression analyses, which will show that the judge’s admissibility decision affects people’s perceptions of legitimacy, but that it does so differently depending on the actor who has proffered the evidence. It will then show that the judge’s decision also affects people’s perceptions of the fairness of the trial overall, but differently depending on whether the proffering party is the initiator or the defendant in the lawsuit. It will then show that participants’ perceptions of the trial’s overall fairness predict their willingness to legitimize the legal tribunal.

  177. [177]. Simple effect of admissibility decision on perceptions of legitimacy when the evidence was excluded: B = -0.40, SE = 0.25, t = -1.60, p < .113; simple effect of admissibility decision on perceptions of legitimacy when the evidence was admitted: B = 0.65, SE = 0.28, t = 2.34, p = .021.

  178. [178]. Interactive effect: B = 1.16, SE = 0.35, t = 3.34, p = .001; simple effect of admissibility decision on perceptions of fairness when the evidence was admitted: B = 0.64, SE = 0.26, t = 2.47, p = .015; simple effect of admissibility decision on perceptions of fairness when the evidence was excluded: B = -0.52, SE = 0.23, t =- 2.25, p = .026.

  179. [179]. B = 0.92, SE = 0.04, t = 25.90, p < .001.

  180. [180]. B = -0.02, SE = 0.19, t = =0.12, p = .900.

  181. [181]. We repeated this analysis by replacing decisional accuracy for procedural fairness as the mediator in the analysis. We found the same pattern of results here as we did when procedural fairness served as the mediator: participants saw the trial as less accurate—not more accurate—when the judge allowed the prosecutor’s (or plaintiff’s) trapdoor evidence, which caused their ratings of the trial’s legitimacy to significantly decrease. Notably however—and consistent with our previous research—participants’ perception of the trial’s decisional accuracy was a substantially weaker mediator of participants’ willingness to legitimize the trial than was their perception of the trial’s overall fairness.

  182. [182]. Asterisks in the mediated moderation model indicate statistically significant associations.

  183. [183]. For a review of traits that change a person’s perceived legitimacy of authority, see generally van der Toorn et al., supra note 135.

  184. [184]. A linear regression is a statistical test that estimates the independent effects of several predictor variables on a continuous dependent variable. See Lawless et al., supra note 134, at 29, 300–31.

  185. [185]. The analyses revealed that participants’ income, education, political orientation, theories of social dominance and attitudes toward the courts affected their perceptions of the legitimacy of the legal tribunal. Specifically, people who are more highly educated, who do not subscribe to a social dominance theory of human relationships, and who self-identify as more familiar with the legal system were less likely to view American courts as legitimate. People who are more liberal, wealthier, authoritarian, system-justifying, and who were more inclined to find the defendant in this study guilty were all more likely to view the courts as legitimate. Although these effects are fascinating and deserve further study (particularly the effect of one’s knowledge of the legal system), we focus solely on the effects of trapdoor evidence in this study when these other variables are held constant.

  186. [186]. See generally Shari Seidman Diamond, Truth, Justice, and the Jury, 26 Harv. J.L. & Pub. Pol’y 143 (2003) (examining this issue in the context of research on the American jury); Kenneth S. Klein, Truth and Legitimacy (in Courts), 48 Loy. U. Chi. L.J. 1 (2016) (discussing two studies comparing the inquisitorial and adversarial systems in this context); Sevier, supra note 88 (discussing two studies comparing the inquisitorial and adversarial systems in this context).

  187. [187]. See generally Thibaut & Walker, supra note 73 (articulating this theory for the first time and providing empirical support for the propositions).

  188. [188]. See generally Tyler & Sevier, supra note 77 (evaluating the influence of public views on inaccuracies in verdicts and substantive justice on public perceptions of the legitimacy of the courts).

  189. [189]. See, e.g., Bruce Gilley, The Right to Rule: How States Win and Lose Legitimacy 140–206 (2009) (describing in detail the benefits of popular legitimacy on the functioning of state and federal governments and discussing the dangers of the loss of that legitimacy).

  190. [190]. See, e.g., Barron v. Ford Motor Co. of Can. Ltd., 965 F.2d 195, 199 (7th Cir. 1992) (“A pure rule of evidence, like a pure rule of procedure, is concerned solely with accuracy and economy in litigation and should therefore be tailored to the capacities and circumstances of the particular judicial system ... .”); Stein, supra note 19, at 9 (articulating three objectives to evidence law, and listing “accuracy in fact-finding” as the first objective); Boney, supra note 19, at 170 (explaining a line of attack on the Hanna Rule in Civil Procedure involving an argument that “the Federal Rules of Evidence were enacted to address accuracy”).

  191. [191]. See, e.g., Sevier, supra note 88; Tyler & Sevier, supra note 77.

  192. [192]. See Justin Sevier, The Unintended Consequences of Local Rules, 21 Cornell J.L. & Pub. Pol’y 291, 343 (2011) (footnote omitted) (“Much of what we know about legal rules stems not from an empirical examination of how these rules operate, but from intuitive hunches. Sometimes, however, hunches are wrong.”); see, e.g., Michael J. Saks & Barbara A. Spellman, The Psychological Foundations of Evidence Law 7–8 (2016) (noting that, in pursuing conflicting goals including fairness, accuracy, and efficiency, “rulemakers are acting not only as amateur psychologists, but also as amateur logicians, statisticians, and scientists of various kinds”).

  193. [193]. See Perlman, supra note 18, at 1639 n.1.

  194. [194]. See Fed. R. Evid. art. VIII, Introductory Note: The Hearsay Problem (focusing on cross-examination, which is absent from many types of hearsay evidence, as a critical feature involved in assuring that juries reach accurate verdicts).

  195. [195]. See, e.g., Justin Sevier, Popularizing Hearsay, 104 Geo. L.J. 643 (2016) (examining the results of statistical studies in support of this assertion).

  196. [196]. See, e.g., Justin Sevier, On Hearsay Dragon-Slaying, 67 Fla. L. Rev. F. 269, 279 (2016) (footnote omitted) (“[T]here is clear value in premising the hearsay doctrine on the normative philosophical concept of confrontation instead of premising it on slippery, empirically suspect notions of reliability.”); id. at 278 (footnote omitted) (suggesting that “rule makers ... get out of the reliability business—and the empirical baggage that comes with it—in fashioning the rationale for the hearsay bar and its exceptions”).

  197. [197]. Fed. R. Evid. 609 (labeled “Impeachment by Evidence of a Criminal Conviction” and creating a complex, byzantine scheme by which some convictions, but not others, are admissible in civil and criminal trials to impeach a witness).

  198. [198]. Id. Subject to certain complex exceptions for old convictions, juvenile offenses and pardons, the Rule generally allows as impeachment evidence all convictions stemming from crimes of dishonesty. See id. R. 609(a)(2). Felony convictions of any kind are generally admissible against civil and criminal witnesses, as well as defendants, based on tests that balance the probative value of the conviction against its unfair prejudice. Cf. id. R. 403 (articulating the balancing of interests for non-conviction evidence). The balancing test is most likely to allow prior convictions into evidence when the target is a mere witness, and the balancing test is most restrictive when the target is a (testifying) criminal defendant. Compare id. R. 609(a)(1)(A), with id. R. 609(a)(1)(B) (under the former, a non-defendant witness’s conviction is admissible unless its prejudicial effect substantially outweighs its probative value; under the latter, a defendant witness’s conviction is admissible unless its prejudicial effect outweighs or is equal to its probative value).

  199. [199]. See generally Theodore Eisenberg & Valerie P. Hans, Taking a Stand on Taking the Stand: The Effect of a Prior Criminal Record on the Decision to Testify and on Trial Outcomes, 94 Cornell L. Rev. 1353 (2009) (examining over “300 criminal trials in four large counties” around the United States to study the relationship “between the existence of a prior criminal record” and the defendant’s decision to testify at trial). More troubling, an analysis of cases involving DNA exonerations demonstrated that many defendants placed in this evidentiary catch-22 declined to testify, and therefore deprived the fact finder of information that may have resulted in their acquittal. Id. at 1389.

  200. [200]. See, e.g., Rebecca Hollander-Blumoff & Tom R. Tyler, Procedural Justice and the Rule of Law: Fostering Legitimacy in Alternative Dispute Resolution, 2011 J. Disp. Resol. 1, 12–18 (examining the intersection of procedural justice in alternative dispute resolution systems, including arbitration, mediation, and negotiation).

  201. [201]. Id. at 14.

  202. [202]. Id.

  203. [203]. Idaho Man Gets Prison for Stealing Vet Benefits, Oregonian (Jan. 6, 2009, 8:04 PM), http://www.


Charles Ehrhardt Assistant Professor, Florida State University College of Law.

I thank Shawn Bayern, Avlana Eisenberg, Joni Hersch, Jake Linford, Mark Seidenfeld, Mark Spottswood, Tom R. Tyler, Kip Viscusi, the Florida State University College of Law Faculty, and the Yale University Department of Psychology for their guidance and for helpful oral and written comments regarding the contents of this Article.