Instructing Jurors on Reasonable Doubt: It’s All Relative
By: Michael D. Cicchini *
The Constitution protects us from criminal conviction unless the government can prove guilt beyond a reasonable doubt. However, this high burden is only as formidable as the words used to describe it to the jury. And many courts describe it in ways that lower, and sometimes even shift, the burden of proof.
This Article identifies four common jury-instruction flaws—the important-affairs-of-life analogy, the alternative-hypothesis test, the unreasonable-doubts warning, and the search-for-the-truth mandate—and then explains, both logically and empirically, how each one violates our due process rights.
After discussing the reasonable-doubt standard and common jury-instruction flaws in Parts I and II, Part III discusses my attempt to win a very modest reform of Wisconsin’s jury instruction—a disastrous piece of work that incorporates all four of these burden-lowering defects. However, because my reform effort achieved only limited success, this Article advocates for a more aggressive approach: rewriting the burden of proof jury instruction from scratch.
This new jury instruction, presented in Part IV, is rooted both in logic and empirical evidence. Specifically, it avoids the four defects discussed in this Article. More generally, it focuses the jury’s attention on the level of proof the government must present, rather than on the kind of doubt the defense must create. This ensures that the burden remains with the government and is not shifted to the defendant.
Finally, and most importantly, to avoid the problems associated with nearly every attempt to define “proof beyond a reasonable doubt,” the proposed instruction describes the burden on a relative basis by comparing it to lower burdens of proof. Because these lower burdens—especially the “more likely than not” standard—are far more intuitive, they offer the best framework for explaining the high level of proof the government must satisfy to win a criminal conviction.
I. The Reasonable Doubt Standard
Since the late 1800s, the United States Supreme Court has implicitly recognized the “beyond a reasonable doubt” standard as the required burden of proof in criminal cases. Then, in 1970, the Court explicitly held that “the Due Process Clause protects the accused against conviction except upon proof beyond a reasonable doubt.”
The Court has stated that, for a jury to convict a defendant under this high burden of proof, it must “reach a subjective state of near certitude of the guilt of the accused[.]” Requiring this high level of certainty protects the accused “from dubious and unjust convictions, with resulting forfeitures of life, liberty and property.” And, more broadly, this high standard “is indispensable to command the respect and confidence of the community in applications of the criminal law.”
However, as often happens in applications of the criminal law, lower courts have failed to embrace the Court’s lofty ideals. Instead, they have chipped away at the burden of proof by describing “reasonable doubt” to jurors in ways that lower the government’s burden below the constitutionally-guaranteed standard.
II. What’s in a Name?
Some burdens of proof are easy to conceptualize. For example, to find that a plaintiff proved something by the preponderance of evidence, “you must be persuaded that it is more probably true than not true.” This burden of proof even lends itself to a helpful numeric definition: it requires “that more then 50% of the evidence points to something.”
But what, exactly, is proof beyond a reasonable doubt? Its meaning is not nearly as intuitive, and “research has shown that there remains sizeable variability in interpretations of the standard when it is left undefined[.]” For example, consider the state of Wyoming, which does not define the standard. In one post-verdict study of real-life jurors, nearly one-third mistakenly believed that, as long as the state put on some evidence, “it becomes the defendant’s responsibility to persuade the jury of his innocence.”
Most states, however, do attempt to define the standard for their juries. But states often do such a poor job that jury instructions on the clear and convincing evidence standard (a lower burden of proof) offer defendants more protection than instructions on the proof beyond a reasonable doubt standard—even though the reverse should be true. Such flawed reasonable-doubt instructions vary dramatically by jurisdiction. However, they commonly include one or more specific defects that effectively diminish the constitutionally-mandated burden of proof. The most common defects are discussed below.
A. The Important Affairs of Life
One common way that courts explain reasonable doubt to jurors is by analogy to decision-making in the important affairs of their own lives. The states of Minnesota and Pennsylvania have used this approach, as have the courts of the Third and Fifth Circuits, among other state and federal jurisdictions. For example, Pennsylvania juries are instructed that “[a] reasonable doubt is a doubt that would cause a reasonably careful and sensible person to hesitate before acting upon a matter of importance in his or her own affairs.”
When read literally, this language appears to be pro-defendant. But in the jury-deliberation room, things quickly fall apart. “[T]ypical jurors . . . always hesitate before acting in the most important aspects of life.” Therefore, “the judge could not possibly mean that jurors should never convict because they always hesitate . . . Thus, a reasonable juror might infer that the judge must mean that one should not convict if, after deciding that the defendant is guilty, one still hesitates.”
In other words, the analogy is a bad fit. Jurors simply do not use the reasonable-doubt standard in their personal decision-making—not even when “acting upon a matter of importance.” According to the Federal Judicial Center, “decisions we make in the most important affairs of our lives—choosing a spouse, a job, a place to live, and the like—generally involve a heavy element of uncertainty and risk-taking. They are wholly unlike the decisions jurors ought to make in criminal cases.”
Of the above examples, “choosing a spouse” is probably the most important decision any juror will ever make. Yet even that example does not compare with the decision a juror must make in a criminal case. A California court explained it this way:
The marriage example is also misleading since the decision to marry is often based on a standard far less than reasonable doubt, as reflected in statistics indicating 33 to 60 percent of all marriages end in divorce. . . . “The judgment of a reasonable [person] in the ordinary affairs of life, however important, is influenced and controlled by the preponderance of evidence. . . . But in the decision of a criminal case involving life or liberty, something further is required.”
The published research also supports this view, particularly when the important-affairs-of-life instruction focuses the jury—as Pennsylvania’s instruction does—on what kind of doubt would cause a person to hesitate, rather than what kind of proof would induce a person to act.
The “doubt-hesitate” instruction lowered the standard considerably . . . [T]he findings do not support the judicial preference for the “doubt-hesitate” instruction. Rather, the . . . findings imply that by reducing the standard of proof below that intended by the law, the “doubt-hesitate” instruction is more likely to lead to false convictions . . .
In addition to this defective analogy, there are three other flaws that commonly appear in burden of proof jury instructions. And some of these flaws do more than merely lower the burden of proof—they actually shift it.
B. The Alternative Hypothesis
Some states, including Connecticut, have instructed jurors that, “[i]f two conclusions can reasonably be drawn from the evidence, one of innocence and one of guilt, you must adopt the one of innocence.” Other states, such as Wisconsin, word the test slightly differently, instructing the jury to decide whether the evidence can be reconciled upon “any reasonable hypothesis consistent with the defendant’s innocence[.]” When this type of alternative-hypothesis language is used in the burden of proof instruction, it poses two serious problems.
First, as the Second Circuit Court of Appeals has explained, when the defense does put on some evidence at trial, the instruction requires the jury to balance two competing theories. Once again, this “suggests that a preponderance of the evidence standard is relevant, when it is not.” That is, when the jury looks at the two competing theories, “if conviction of a crime fits the facts better than acquittal, it is extremely difficult to overcome the desire to match the facts with the better of the two models, even if the [state’s] case is not very strong.”
Second, more than merely lowering the burden, many alternative-hypothesis instructions actually shift it to the defendant. To demonstrate this, first consider an uncontroversial jury instruction on a lower burden of proof: “When I say that a particular party must prove something by ‘clear and convincing evidence,’ this is what I mean: When you have considered all of the evidence, you are convinced that it is highly probable that it is true.” Such language clearly articulates what the party with the burden must do. Conversely, alternative-hypothesis instructions focus the jury not on what the government must prove, but “on the defendant’s ability to produce alternatives to the government’s case, and thereby shift the burden of proof to the defendant.”
Aside from the obvious due process violation, such burden-shifting is also incompatible with the harsh realities of our criminal justice system. In many cases, the government controls and develops the physical evidence and even has exclusive access to key witnesses, all of which makes the development of alternative hypotheses very difficult. Worse yet, generating such a hypothesis is even more problematic for the factually innocent defendant. The reason is that, in many cases, an innocent person simply “knows nothing about the crime.”
In addition to these practical problems right out of the gate, there are more hurdles awaiting the defendant at trial. Even when a defendant has a strong hypothesis consistent with innocence, or has strong evidence that someone else committed the crime, courts often prevent this information from reaching the jury. Examples of truth-suppressing trial rules include witness-privacy statutes and, more alarmingly, judge-made barriers that exclude evidence of third-party guilt. Such trial rules often make it impossible for the defendant to satisfy the burden that the alternative-hypothesis instruction (illegally) imposes.
When defendants are unable to produce, or are not allowed to present, an alternative theory of the case, the natural implication of the alternative-hypothesis instruction is that the jury should convict. But, as the next section explains, even when the defendant does present evidence that raises reasonable doubts, there is yet another problem: many instructions tell jurors to ignore those doubts.
C. Unreasonable Doubts
In many criminal cases, there will be some doubt about the defendant’s guilt. The question—as framed by jury instructions that improperly shift the burden—is whether the defense has produced a reasonable doubt.
Prosecutors and some trial judges are obsessed with the possibility that a jury might acquit a guilty defendant. (In prosecutors’ minds, of course, every defendant is guilty, or the prosecutor would not have charged him or her in the first place.) Consistent with this irrational obsession, many jury instructions go to great lengths to bully jurors into convicting. At one time, for example, New York’s instruction cautioned each juror not to turn into a “weak-kneed, timid, jellyfish of a juror who is seeking to avoid . . . convict[ing] another human being[.]”
Today, many states’ instructions send the same message but with less hyperbolic flair. Wisconsin, for example, warns jurors that a doubt is not reasonable if it “arises merely from sympathy or from fear to return a verdict of guilt.” One problem with this instruction is that it neglects the flipside of the coin: it fails to tell the jury not to convict a defendant because of its sympathy for the accuser or its fear of finding the defendant not guilty—two emotional angles that prosecutors commonly but illegally exploit in closing argument.
Similarly, Florida’s instruction also enumerates several types of doubt that are not reasonable, warning that “[a] reasonable doubt is not a mere possible doubt, a speculative, imaginary or forced doubt.” The problem with this language is that “[d]oubting, after all, is a matter of speculation and imagination. It requires one to imagine alternative models consistent with the evidence.”
Aside from these serious problems, the larger point is that, by enumerating all of these supposedly unreasonable doubts that should not be used as the basis for an acquittal, the instruction is shifting the burden of proof to the defendant.
The weight of the instruction conveys a message to the jurors: The judge would not have presented so many ways in which the juror’s doubts can be used improperly if this were not the main problem to avoid. Such a message is likely to focus jurors on the strength of the defendant’s case as a criterion for acquittal rather than on whether the government has proven its case with near certitude.
And some states’ instructions go even further. For example, Connecticut warns jurors that they should not consider a doubt “simply for the sake of raising a doubt.” And Wisconsin is beyond the pale, specifically instructing jurors “not to search for doubt.” It is difficult to imagine two more blatantly unconstitutional instructions, given that “[t]he question for any jury is whether the burden of proof has been carried by the party who bears it. In a criminal case . . . [t]he jury cannot discern whether that has occurred without examining the evidence for reasonable doubt.”
But if, as some courts instruct, the jury is “not to search for doubt,” then for what should it search?
D. The Search for the Truth
After discussing reasonable doubt, individual courts in several jurisdictions—including Massachusetts, South Carolina, and the First and Third Circuits—have taken an odd twist. They conclude their burden of proof instructions by telling jurors to “evolve the truth, seek the truth, search for truth, or find the truth.” The problem with such language is this: the issue for the jury is not whether it believes something is true, but rather how confident it is in that belief.
The Fifth Circuit Court of Appeals states the problem this way: “‘seeking the truth’ suggests determining whose version of events is more likely true, the government’s or the defendant’s, and thereby intimates a preponderance of evidence standard.” That is, “if a jury feels the government’s version of events is slightly more likely than the defendant’s version to be true, it would follow that, in a search for the truth, the jury would be obligated to convict[.]”
Prosecutors have vehemently denied the idea that such search-for-the-truth language lowers the burden of proof, and have dismissed it as nothing more than defense lawyers’ wild speculation. Therefore, Lawrence T. White and I set out to empirically test the idea and provide the evidence that prosecutors claimed to want. Over the course of two controlled studies, we made several findings.
First, mock jurors who received the truth-based instruction voted to convict at a significantly higher rate than those who received the identical instruction but without the truth mandate. This statistically significant finding was replicated in the follow-up study. Second, in response to a post-verdict question in the follow-up study, mock jurors who received the truth-based instruction were nearly twice as likely to mistakenly believe it was legally proper to convict even if they had a reasonable doubt about guilt. And third, this mistaken belief matters: mock jurors who believed this, regardless of the jury instruction they received, convicted the defendant a rate two-and-one-half times that of jurors who correctly understood the burden of proof.
Most important, we found that mock jurors who received a truth-based reasonable-doubt instruction convicted at the statistically identical rate as those who received no reasonable-doubt instruction whatsoever. This finding proves what one ethically-challenged prosecutor already knew: truth-based jury instructions lower the burden of proof below the reasonable-doubt standard. Or, as the prosecutor once creatively argued to win a conviction, “You have a gut feeling he’s guilty, he’s guilty.”
III. The Politics of Legal Reform: A Case Study
I represent individuals charged with crimes in Wisconsin state courts. And Wisconsin’s pattern jury instruction on the burden of proof is, quite possibly, the worst in the country. After identifying “beyond a reasonable doubt” as the applicable standard, the drafters of the instruction wasted little time in decimating this standard by incorporating all four of the defects discussed in this Article.
The instruction’s definition of reasonable doubt begins with the important-affairs-of-life analogy; worse yet, it uses the doubt-hesitate version that has been empirically proven to lower the burden of proof. The instruction also asks the jury to decide whether there is a “reasonable hypothesis” consistent with the defendant’s innocence, thereby shifting the burden from the state to the defendant. Then, the instruction includes a litany of doubts that are not reasonable ones, thus deterring jurors from developing the alternative hypothesis the instruction demands, while sending the clear message that most doubts should not be used to acquit. And to cap things off in grand fashion, the instruction actually tells jurors “not to search for doubt” of any kind; instead, they are “to search for the truth.”
This is a pattern instruction created and perpetuated by a jury-instruction committee that was comprised, very recently, of nine trial-court judges, seven of whom are former prosecutors and two of whom are former government lawyers in other capacities. Given this staggeringly pro-government composition, I realized that meaningful reform was unlikely. Therefore, in June 2016, I lobbied the committee only to delete the worst part of its disastrous instruction: its closing mandate “not to search for doubt” but “to search for the truth.” I proposed that it should simply conclude as follows: “It is your duty to give the defendant the benefit of every reasonable doubt.”
This seemed like a reasonable—even uncontroversial—request. After all, the burden of proof is beyond a reasonable doubt. And I even cited the recently-published empirical evidence that proved the obvious: telling jurors “to search for the truth” lowered the burden below the constitutionally-guaranteed standard. Finally, under my extremely modest proposal for reform, the committee would still retain all of its other burden-lowering language: the important-affairs-of-life analogy, the alternative hypothesis test, and its litany of supposedly unreasonable doubts.
In addition to lobbying the jury-instruction committee, many members of the criminal defense bar were citing the published studies and asking individual trial judges to modify the instruction on a case-by-case basis. Fortunately, trial judges are not bound by the committee’s pattern instructions and, in fact, may not use them when they fail to accurately convey the legal concept they purport to describe.
Not surprisingly, prosecutorial reaction throughout Wisconsin has been frantic, and they have marshaled at least twenty different arguments to oppose defense lawyers’ requests to modify the pattern instruction. Their arguments have ranged from the mathematically false—for example, claiming that the studies’ findings are not statistically significant—to the merely ignorant. The more entertaining of these include attacks on the “defense attorney journals”—the University of Richmond Law Review and the Columbia Law Review Online—in which the studies were published.
Prosecutors have also trotted out their older, tried-and-true arguments, including misrepresenting the holding of a 1923 case and attempting to use Medieval Latin to justify our state’s defective, modern-day jury instruction. In short, the prosecutorial response has been a collection of misinformed and disingenuous arguments to preserve the burden-lowering language on which they rely to win convictions.
While all of that was expected, what was surprising was the impenetrable black box in which the jury-instruction committee operated. Impenetrable, that is, to anyone who is not a prosecutor. Since September 2016, prosecutors have been enthusiastically reporting that the committee decided not to modify the instruction. Then, nine months later on June 29, 2017, I received an email from the reporter of the committee, informing me that the committee had, in fact, decided against modification. The reporter was apparently unaware that prosecutors had been spreading the news of this decision since September 2016; he claimed the committee had discussed the matter in October, and did not make its decision until December, of 2016.
Fortunately, the defense bar has enjoyed some success on an individual, case-by-case basis. Nineteen Wisconsin trial judges—including one that serves on the jury-instruction committee—have modified the pattern burden of proof instruction. Most judges have not gone as far as I have proposed, but instead opted for even more modest changes. And while some of those changes are next to nothing, anything is an improvement over the committee’s pattern instruction.
The lesson in all of this, from a legal-reform perspective, is that when you ask for a little, you will get just that—or even less. Perhaps the more politically astute approach would have been to immediately identify all four of the defects in the committee’s instruction (as this Article has done), enlist the efforts of an individual-rights group with some political muscle, and then demand a brand new instruction—one that accurately describes the state’s high burden of proof.
I will save the committee and the courts the effort of drafting such a jury instruction. Below I will introduce a clear, simple, and accurate instruction that is free of burden-lowering and burden-shifting defects. And I challenge all judges and prosecutors to identify any way in which it fails to accurately convey the constitutionally-mandated burden of proof.
IV. Building a Better Instruction: It’s All Relative
The first part of the reasonable-doubt instruction must explain the presumption of innocence and describe it in a meaningful way. For this, I draw from the instructions of Wisconsin and Hawaii.
Defendants are not required to prove their innocence. The law presumes every person charged with a crime to be innocent. This presumption requires a finding of not guilty unless, in your deliberations, you find it is overcome by evidence which satisfies you beyond a reasonable doubt that the defendant is guilty. The presumption of innocence is not a mere slogan but an essential part of the law that is binding upon you.
The next part must identify the applicable burden of proof. Here, I again rely heavily on the relevant portion of Wisconsin’s instruction.
The burden of establishing every fact necessary to constitute guilt is upon the State. Before you can return a verdict of guilty, the State must prove beyond a reasonable doubt that the defendant is guilty.
The third and final part must explain the concept of proof beyond a reasonable doubt. In doing so, it must avoid the four defects described in this Article. Further, in a general sense, it must avoid shifting the burden to the defendant. That is, instead of discussing what kind of doubt is reasonable, it must focus on what constitutes proof beyond a reasonable doubt. “By emphasizing the government’s task—and not the defendant’s—it goes a long way toward ensuring that the burden of proof in criminal cases remains where it belongs: on the government.”
Previous attempts to define “proof beyond reasonable doubt” have been, at best, a “grand conglomeration of garbled verbiage and verbal garbage.” But neither can the phrase be left undefined. As discussed earlier, empirical studies demonstrate that jurors interpret the phrase in wildly different, and often unconstitutional, ways. Given this, a New Hampshire court’s concern rings true: “this court feels strongly that a jury must be given some assistance in understanding the concept.”
The key to avoiding “verbal garbage,” while at the same time accurately describing the burden of proof for jurors, is to define it on a relative basis. It is important to start with a concept that juries intuitively understand: proof by the preponderance of evidence, or the “probably true” standard. The next step is to explain, on a relative basis, the higher burden of proof of clear and convincing evidence, or the “highly probable” standard. The final step is to explain that proof beyond a reasonable doubt is, relatively speaking, even higher still.
Several states, including Hawaii, New York, Vermont, and Virginia already define the burden of proof on a somewhat relative basis. For example, Vermont states that “[i]f you have a reasonable doubt, you must find [the] [d]efendant not guilty even if you think that the charge is probably true.” Oregon and Arizona go further by comparing the burden not only to the preponderance of evidence standard, but also to the clear and convincing standard. For example, Arizona states that, “In civil cases, it is only necessary to prove that a fact is more likely true than not or that its truth is highly probable. In criminal cases such as this, the State’s proof must be more powerful than that. It must be beyond a reasonable doubt.” I rely heavily on this language from Arizona and Vermont for the final part of the instruction.
Some civil cases use the preponderance of evidence standard. In those cases, it is only necessary to prove that something is probably true, or more likely true than not. But this is a criminal case, and the State’s proof must be more powerful than that.
Other civil cases use the clear and convincing evidence standard. In those cases, it is necessary to prove that the truth of something is highly probable. But this is a criminal case, and the State’s proof must also be more powerful than that.
In criminal cases such as this, you can convict the defendant only if the State’s proof satisfies you beyond a reasonable doubt that the defendant is guilty. If it does not, you must find the defendant not guilty even if you think that the charge is probably true, and even if you think it is highly probable that the charge is true.
This instruction clearly, simply, and accurately describes the burden of proof on a relative basis by anchoring it to the easily understood concept of “probably true,” and then building from that point. And it eliminates the burden-lowering and burden-shifting problems inherent in most garbled attempts to define proof beyond a reasonable doubt.
While prosecutors will not be able to identify any inaccuracy in this instruction, they will, of course, complain that it imposes too high of a burden. Predictably, they will argue that the jury will mistakenly believe that it must be satisfied beyond all doubt before it may convict.
Such prosecutorial hand-wringing is not persuasive. To begin with, the legal standard is nearly that high; the Supreme Court has described being convinced beyond a reasonable doubt as being in a “subjective state of near certitude.” Additionally, the jury instruction already uses the words reasonable doubt, which necessarily excludes unreasonable doubts—a point that prosecutors are free to hammer home in their closing arguments.
But if prosecutors do not like this clear, simple, and accurate instruction, they are always free to request North Carolina’s instruction instead. It concludes as follows: “Proof beyond a reasonable doubt is proof that fully satisfies or entirely convinces you of the defendant’s guilt.”
On second thought, given the lesson I learned about the dangers of asking for too little, perhaps I should ask our state’s jury-instruction committee and individual trial judges to adopt North Carolina’s “fully satisfies or entirely convinces you” standard. Such language would ensure that all defendants receive the due process the Constitution guarantees. And because every prosecutor has an ethical duty to act as “a minister of justice”—a duty that “carries with it specific obligations to see that the defendant is accorded procedural justice”—surely they would not object.
The Supreme Court has long recognized the importance of a high burden of proof in criminal cases. Requiring the government to prove guilt beyond a reasonable doubt simultaneously protects individuals from unjust convictions and promotes confidence in the criminal justice system. Therefore, the way judges define the burden of proof for juries is of the utmost importance.
A legally proper jury instruction on the state’s burden of proof must accomplish three things. First, it must avoid the specific jury-instruction defects discussed in this Article. Second, it must focus the jurors’ attention on what the state must prove to win a conviction, rather than on what the defendant must do to win an acquittal. This ensures that the burden of proof is not unconstitutionally diminished or, worse yet, shifted to the defendant.
And third, a legally proper jury instruction must define the concept of proof beyond a reasonable doubt on a relative basis. Because the lower burdens of proof—in particular, the “probably true” or preponderance of evidence standard—are far more intuitive, this relative approach provides jurors with much needed points of reference. In so doing, it ensures that jurors properly understand the high level of proof the government must produce in order to convict the defendant.
* Criminal Defense Lawyer, Cicchini Law Office, LLC, Kenosha, Wisconsin. J.D., summa cum laude, Marquette University Law School (1999); C.P.A., University of Illinois Board of Examiners (1997); M.B.A., Marquette University Graduate School (1994); B.S., University of Wisconsin—Parkside (1990).
 See, e.g., Davis v. United States, 160 U.S. 469, 488 (1895), https://supreme.justia.com/cases/federal/us/160/469/case.html [https://perma.cc/6D2E-H8ZP]; Miles v. United States, 103 U.S. 304, 312 (1881), https://supreme.justia.com/cases/federal/us/103/304/case.html [https://perma.cc/64WA-LVKL].
 In re Winship, 397 U.S. 358, 364 (1970), https://supreme.justia.com/cases/federal/us/397/358/case.html [https://perma.cc/GHG6-HBR2].
 Jackson v. Virginia, 443 U.S. 307, 315 (1979) (emphasis added), https://supreme.justia.com/cases/federal/us/443/307/case.html [https://perma.cc/M9YN-3Q79].
 In re Winship, 397 U.S. at 362 (quoting Brinegar v. United States, 338 U.S. 160, 174 (1949)).
 Id. at 364.
 Fed. Civil Jury Instructions of the Seventh Circuit, No. 1.27 (2009), http://www.ca7.uscourts.gov/pattern-jury-instructions/7th_cir_civil_instructions.pdf [https://perma.cc/V3XP-3JCU].
 Cornell Law School: Legal Information Institute, Preponderance of the Evidence, at https://www.law.cornell.edu/wex/preponderance_of_the_evidence [https://perma.cc/RM3C-A2XC].
 Mandeep K. Dhami, et al., Instructions on Reasonable Doubt: Defining the Standard of Proof and the Juror’s Task, 21 Psychol. Pub. Pol’y & L. 169, 176 (2015), http://eprints.mdx.ac.uk/16067/ [https://perma.cc/8RCH-48KY].
 Lawrence Solan, Refocusing the Burden of Proof in Criminal Cases: Some Doubt About Reasonable Doubt, 78 Tex. L. Rev. 105, 119–120 (1999) (discussing Bradley Saxton, How Well Do Jurors Understand Jury Instructions? A Field Test Using Real Juries and Real Trials in Wyoming, 33 Land & Water L. Rev. 59 (1998)), http://brooklynworks.brooklaw.edu/cgi/viewcontent.cgi?article=1293&context=faculty [https://perma.cc/5YL7-NFBF].
 See id. at 105.
 Id. at 114.
 Pa. Crim. Jury Instructions No. 7.01 (2016).
 Solan, supra note 9, at 143.
 Id. (emphasis added).
 Fed. Jud. Ctr. Pattern Crim. Jury Instructions No. 21 (1987); see also United States v. Jaramillo-Suarez, 950 F.2d 1378, 1386 (9th Cir. 1991) (criticizing the important-affairs-of-life analogy), http://law.justia.com/cases/federal/appellate-courts/F2/942/1412/282241/ [https://perma.cc/3BQ5-93QT].
 People v. Nguyen, 46 Cal. Rptr. 2d 840, 845 (1996) (emphasis added) (quoting People v. Brannon, 47 Cal. 96, 97 (1873)), http://law.justia.com/cases/california/court-of-appeal/4th/40/28.html [https://perma.cc/D3NG-BYF5].
 Dhami, supra note 8, at 175.
 State v. Griffin, 749 A.2d 1192, 1197 (Conn. 2000). Connecticut has since changed to an awkward, but preferable, instruction: “Proof beyond a reasonable doubt is proof that precludes every reasonable hypothesis except guilt and is inconsistent with any other rational conclusion.” Conn. Crim. Jury Instructions No. 2.2-3 (2007), https://jud.ct.gov/JI/Criminal/Criminal.pdf [https://perma.cc/DC37-MNCZ].
 Wis. Crim. Jury Instructions No. 140 (2016).
 New York, for example, uses this instruction only in the context of evaluating cases that are entirely circumstantial. See N.Y. Crim. Jury Instructions 2d, Circumstantial Evidence, http://www.nycourts.gov/judges/cji/1-General/CJI2d.Circumstantial_Evidence.pdf [https://perma.cc/UX2B-S7GE].
 United States v. Kahn, 821 F.2d 90, 93 (2d Cir. 1987), http://law.justia.com/cases/federal/appellate-courts/F2/821/90/255553/ [https://perma.cc/CCR7-PPAM].
 Solan, supra note 9, at 108–09.
 Fed. Civil Jury Instructions of the Seventh Circuit, No. 1.28 (2005).
 Solan, supra note 9, at 105 (emphasis added).
 See Keith A. Findley, Adversarial Inquisitions: Rethinking the Search for the Truth, 56 N.Y. L. Sch. L. Rev. 912, 912–14 (2011), http://www.lawschool.cornell.edu/international/study_abroad/paris_summer/admitted-students/Comparative-Law-and-Social-Science/loader.cfm?csModule=security/getfile&PageID=70696 [https://perma.cc/Q4UQ-6FF4].
 Solan, supra note 9, at 108.
 See, e.g., State v. Carter, 782 N.W.2d 695 (Wis. 2010) (statute prevents defendant from presenting an alternative source of child accuser’s “detailed sexual knowledge,” despite prosecutor’s repeated arguments that such knowledge proved defendant’s guilt), http://law.justia.com/cases/wisconsin/supreme-court/2010/50358.html [https://perma.cc/9MU6-JKSX]; People v. Arenda, 330 N.W.2d 814 (Mich. 1982) (same). Cf. State v. Colburn, 366 P.3d 258 (Mont. 2016) (accuser’s prior sexual experience may be relevant to show an alternative source of sexual knowledge), http://law.justia.com/cases/michigan/supreme-court/1982/65578-4.html [https://perma.cc/84EY-A2PD].
 See David Schwartz & Chelsey B. Metcalf, Disfavored Treatment of Third-Party Guilt Evidence, 2016 Wis. L. Rev. 337, 378 (2016) (“The argument that evidence of third-party guilt is excludable because it is a waste of time is breathtaking in its disregard for a criminal defendant’s due process rights.”), http://wisconsinlawreview.org/wp-content/uploads/2016/06/4-Schwartz-Metcalf-Final.pdf [https://perma.cc/6TJQ-XDR7].
 People v. Feldman, 71 N.E.2d 433, 443 (N.Y. 1947) (disapproving of the jellyfish analogy).
 Wis. Crim. Jury Instructions No. 140 (2016).
 E-mail from Lawrence T. White, Professor of Psychology, Beloit College (June 13, 2017, 09:03 a.m. CST) (on file with author).
 See, e.g., Rhodes v. State, 547 So. 2d 1201, 1206 (Fla. 1989) (prosecutor urged jury to “show [the defendant] the same mercy shown to the victim on the day of her death.”), http://law.justia.com/cases/florida/supreme-court/1989/67842-0.html [https://perma.cc/K2TF-PP9Y]; Commonwealth of Northern Mariana Islands v. Mendiola, 976 F.2d 1475, 486–87 (9th Cir. 1992) (prosecutor argued to jury, “that gun is still out there. If you say not guilty, [the defendant] walks right out the door, right behind you.”), http://law.justia.com/cases/federal/appellate-courts/F2/976/475/47197/
 Fla. Crim. Jury Instructions No. 3.7 (1997), http://www.floridasupremecourt.org/jury_instructions/instructions-ch3.shtml [https://perma.cc/X437-GL5D].
 Solan, supra note 9, at 143.
 Id. at 144 (emphasis added).
 Conn. Crim. Jury Instructions No. 2.2-3 (2008).
 Wis. Crim. Jury Instructions No. 140 (2016).
 State v. Berube, 286 P.3d 402, 411 (Wash. Ct. App. 2012) (emphasis added), http://law.justia.com/cases/washington/court-of-appeals-division-i/2012/63579-4.html
 Michael D. Cicchini & Lawrence T. White, Truth or Doubt? An Empirical Test of Criminal Jury Instructions, 50 Rich. L. Rev. 1139, 1143 (2016) (internal quotations and citations omitted), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2747718 [https://perma.cc/25Y3-ZPRT].
 Id. at 1144 (“truth refers to a judgment about whether something happened; doubt refers to the level of certainty in that judgment.”).
 United States v. Gonzalez-Balderas, 11 F.3d 1218, 1223 (5th Cir. 1994) (emphasis added), http://law.justia.com/cases/federal/appellate-courts/F3/11/1218/488621/ [ https://perma.cc/L7UR-P8KK].
 Cicchini & White, supra note 39, at 1145.
 Cicchini & White, supra note 39, at 1154–56.
 Michael D. Cicchini & Lawrence T. White, Testing the Impact of Criminal Jury Instructions on Verdicts: A Conceptual Replication, 117 Colum. L. Rev. Online 22, 30–31 (2017), http://columbialawreview.org/content/testing-the-impact-of-criminal-jury-instructions-on-verdicts-a-conceptual-replication/ [https://perma.cc/8ZZX-QNAK].
 Id. at 32.
 Cicchini & White, supra note 39, at 1157.
 Randolph v. State, No. 36080, 8 (Nevada, 2001) (emphasis added), http://law.justia.com/cases/nevada/supreme-court/2001/36080-1.html [https://perma.cc/8G8S-WXZM]. Such astute legal reasoning catapulted the prosecutor to the bench, where he has found more ethics-related trouble in his capacity as judge. See Mauricio R. Hernandez, Vegas Judge Had Long History of Prosecutorial Misconduct, The Irreverent Lawyer (May 31, 2017), at https://lawmrh.wordpress.com/2017/05/31/vegas-judge-had-long-history-of-prosecutorial-misconduct/ [https://perma.cc/W8AE-J45G].
 Wis. Crim. Jury Instructions No. 140 (2016).
 This jury-instruction committee actually has eleven members. However, the committee operates in a black box and, further, is often in a state of some flux. My representation as to its make-up refers to its likely composition on January 2, 2017. See Michael D. Cicchini, The Battle over the Burden of Proof: A Report from the Trenches, 79 U. Pitt. L. Rev. __, Part III.E. (2017), at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2916389.
 See Cicchini & White, Truth or Doubt?, supra note 39; Cicchini & White, Testing the Impact, supra note 44.
 See, e.g., State v. Neumann, 32 N.W.2d 560, 584 (Wis. 2013) (when relying on pattern instructions, “A circuit court must, however, exercise its discretion in order to fully and fairly inform the jury of the rules of law applicable to the case and to assist the jury in making a reasonable analysis of the evidence.”).
 Statistical significance is not a matter of opinion; it is a mathematical calculation. See Cicchini & White, Truth or Doubt?, supra note 39, at 1154–55 (discussing the statistic called the p-value).
 While “defense attorney journals,” such as The Champion, do exist, the two journals in which the studies were published are general interest law reviews.
 Manna v. State, 192 N.W. 160, 166 (1923) (holding that the trial judge’s instruction “to ascertain the truth” was acceptable when instructing the jury on how to resolve factual disputes, not when explaining the state’s burden of proof).
 Prosecutors often argue that “verdict,” from the Medieval Latin, means “to speak the truth.” But Medieval jury trials were quite different from modern trials; back then, “jurors themselves were considered the witnesses . . . they reported facts to the judges. They were self-informing; they came to court more to speak than to listen.” Daniel Klerman, Was the Jury Ever Self-Informing?, 77 S. Cal. L. Rev. 123, 123 (2016), http://mylaw2.usc.edu/assets/docs/directory/1000184.pdf [https://perma.cc/SHL2-DD5Q].
 See, e.g., Letter from Brandon Wigley, Assistant District Attorney, Milwaukee County, Wisconsin (Sept. 22, 2016) (“It is the State’s understanding that this issue is before the a [sic] state-wide [sic] jury instruction committee . . . and that they [sic] have elected not to change the language of jury instruction 140 . . .”) (emphasis added) (on file with author). While the letter is self-contradictory on this point, and incorrect on other points, several other prosecutors were verbally sharing the news as early as September 2016.
 E-mail from David Schultz, Reporter, Wisconsin Criminal Jury Instructions Committee (June 29, 2017, 11:19 a.m. CST), at http://www.cicchinilawoffice.com/uploads/Letter_from_JI_Committee.pdf [https://perma.cc/8HG9-PTKK].
 Michael D. Cicchini, Wis. Crim. Jury Instruction 140 Resource Page, at http://www.cicchinilawoffice.com/Wis_JI_140.html [https://perma.cc/8FY3-DW8H].
 Other members of the defense bar have reported to me that one common change is to modify the instruction to conclude as follows: “You are to search for the truth while giving the defendant the benefit of every reasonable doubt.” This is certainly preferable to instructing the jury “not to search for doubt.”
 See Wis. Crim. Jury Instructions No. 140 (2016). This proposal uses “a crime” instead of the pattern instruction’s more cumbersome “the commission of an offense.”
 Haw. Crim. Jury Instructions No. 3.02 (2014), http://www.courts.state.hi.us/docs/docs4/crimjuryinstruct.pdf [https://perma.cc/7XL8-DBXT].
 See Wis. Crim. Jury Instructions No. 140 (2016). The second sentence has been modified to be structurally consistent with the first sentence which keeps the burden of proof on the state.
 Solan, supra note 9, at 106.
 State v. Aubert, 421 A.2d 124, 127 (N.H. 1980) (internal citations omitted), http://law.justia.com/cases/new-hampshire/supreme-court/1980/80-005-0.html [https://perma.cc/ZN5Q-SJLE].
 Vt. Crim. Jury Instructions No. 04-101 (2005) (emphasis added).
 Ariz. Crim. Jury Instructions No. 5(b)(1) (2016) (emphasis added), http://www.azbar.org/media/1179884/rajicriminal-4thed2016-final.pdf [https://perma.cc/SN55-PABY]; see also Elisabeth Stoffelmayr & Shari S. Diamond, The Conflict Between Precision and Flexibility in Explaining “Beyond a Reasonable Doubt”, 6 Psychol., Pub. Pol’y, & L. 769, 776–77 (2000) (citing the Federal Judicial Center’s model instruction and arguing that contrasting the criminal burden of proof with the two lesser burdens of proof will be helpful to jurors with prior civil jury service and will also increase comprehension of the criminal-court standard), http://www.law.northwestern.edu/faculty/fulltime/diamond/papers/conflictBetweenPrecisionAndFlexibility.pdf [https://perma.cc/KH77-EFFK].
 It is important not to “anchor” the burden of proof to something absurdly low, such “mere suspicion of guilt,” as this would completely defeat the purpose of a relative definition. Penn. Crim. Jury Instructions No. 7.01 (2016).
 Jackson v. Virginia, 443 U.S. 307, 315 (1979), https://supreme.justia.com/cases/federal/us/443/307/case.html [https://perma.cc/HW89-KTJS].
 N.C. Crim. Jury Instructions No. 101.10 (2008) (emphasis added), https://www.sog.unc.edu/sites/www.sog.unc.edu/files/pji-master/criminal/100.10.pdf [https://perma.cc/D8V3-7BYB].
 Model Rules of Prof’l Conduct, R. 3.8, Cmt. 1, Special Responsibilities of a Prosecutor (Am. Bar. Ass’n 2011), https://www.americanbar.org/groups/professional_responsibility/publications/model_rules_of_professional_conduct/rule_3_8_special_responsibilities_of_a_prosecutor.html [https://perma.cc/2M24-QMUW].