Reasonable doubt pdf

Monday, May 27, 2019 admin Comments(0)

PDF | Reasonable doubt (RD) is the standard of proof used in criminal trials. It is, however, a fuzzy concept in most jurors' minds, and several. The Origins of Reasonable Doubt: Theological Roots of the Criminal Trial Beyond Reasonable Doubt and Probable Cause: Historical Perspectives on the. Beyond reasonable doubt (BRD) is the standard of proof used to convict defendants charged with crimes in the English criminal justice system . PDF Document.

Language: English, Spanish, Indonesian
Country: Germany
Genre: Biography
Pages: 103
Published (Last): 06.06.2016
ISBN: 840-2-34275-750-3
ePub File Size: 17.72 MB
PDF File Size: 9.12 MB
Distribution: Free* [*Regsitration Required]
Downloads: 26968
Uploaded by: JOLEEN

Beyond Reasonable Doubt. Standard of Proof and Evaluation of Evidence in. Criminal Cases. Christian Diesen. The public debate concerning evaluation of. T H E U NI V E R SI T Y OF T E X A S S C H OOL OF L A W Public Law and Legal Theory Research Paper Series Number Is Reasonable Doubt Reasonable. Reasonable Doubt and the Presumption of Innocence: The Case of the Bayesian Juror* Piers Rawling ABSTRACT. There is a substantial literature on the.

The aim of presenting the evidence is to corroborate the prosecution s theory of the crime. We used Wallsten and Budescu's theory of linguistic probabilities and their membership function MF method to capture peoples' quantitative interpretations of BRD. There is no such burden resting on the defendant or a juror in a criminal case. We have already seen powerful precedent for this gloss since BARD itself had evolved out of the philosophical literature on mo ral certainty. Appe llate cou rts hav e rejecte d both these c onstru als of BA RD co untles s time s. Glass, F. The task of the jury is to evaluate the prosecution s theory or story, deciding wheth er it points un ambig uously to the conc lusion tha t the defen dant is gu ilty.

Section II addresses the two Bayesianism; I shall not distinguish here between dif- legal notions key to my worries about Bayesian jurors: Section III explores what a Bayesian is to times, and each time it has come up heads and this is make of these notions, and how the legal system your only evidence concerning the relevant behavior of requires her to reason in their light.

If I am right, there the coin. What is the probability of tails on the next emerges a conflict between current legal prescriptions toss?

Section IV explores the that tails is more likely than heads here; whereas good import of this conflict and how it might be resolved.

(PDF) Is Reasonable Doubt Reasonable? | Larry Laudan -

For the sake of simplicity, suppose that you believe that either the coin is fair or it has two heads. Printed in the Netherlands.

Note that: And exist. Note evidence for the assignment of prior probabilities, yet also Lindley The one, simple idea: Yet what we want is the probability of in the [0, 1] interval. According to the Bayesian, then, you will have Furthermore, Bayesians are often not unduly worried assigned an initial value to T, prior to witnessing any by the possibility of differences in priors, because of tosses: The probability you assign to T after wit- gent prior opinions but reasonably open minds will be nessing the ten tosses is your posterior probability of forced into arbitrarily close agreement about future T.

What is the relation of prior to posterior? Although, as Edwards et al. And in this circumstance, acknowledge , p. Let E be a piece of evidence already known but anomalous by the lights of old theory. The is initially Note that probabilities of one are never shaken by the For actual historical cases, unlike the coin-flipping case, there is no single counterfactual degree of belief in the evidence ready to evidence: Glymour , p.

This upshot can be extremely close to one, and do not differ by orders of avoided if we accept a counterfactual solution here — magnitude. We is another difficulty lurking here. As to this latter part, the measure of is the ratio. This can be broken into two dicta: In this section I address the legal interpreta- indictment, and the arraignment, and to reach their conclusion tion of these dicta. This caution is indeed particularly needed presumptio juris, demonstrates that it is evidence in favor of the in criminal cases.

This implication presumption of innocence is closely related to the arises because RDC is taken in a broad sense so as to requirement that conviction requires proof beyond a include the requirement that the burden of proof lies reasonable doubt. According to the majority of the with the prosecution. Under burden of proof beyond a reasonable doubt were logically separate this construal the Bayesian agrees with former legal and distinct.

The Court refers us to, amongst others, Wigmore , who tells us that: Straightforward calculation shows that guilt. How low? Lindley , p. Thus, even assuming that the prosecution is so persuasive as to push the [t]he presumption of innocence is itself an approximation: I shall address the arbitrariness of this figure cumstance of a Bayesian juror with that of a Bayesian below; p G cannot, of course, be zero: The detective faces the issue of whether or p.

How is the prior proba- if and only if her posterior probabilistic estimate of the bility that the suspect is guilty fixed for the detective? In United States v. These people are p. The upshot in United States v.

This is still, I think, too low it is definitely too and p GQ are both close to zero. Another is to reject randomly selected individual. Hence, in order to charge a suspect: This ratio is large, but achievable — particularly with Much of Tribe a is directed against Finkelstein DNA evidence.

Finkelstein and Fairley propose that proposition. Samantha Lundrigan. Beyond reasonable doubt BRD is the standard of proof used to convict defendants charged with crimes in the English criminal justice system. Otherwise, the decision maker will acquit the defendant.

It is generally agreed that BRD should be interpreted as a. The main goal of the present research was to better understand how members of the general public potential jurors quantitatively interpret BRD under different judicial instructions, case factors, attitudes, and demographic backgrounds. We used Wallsten and Budescu's theory of linguistic probabilities and their membership function MF method to capture peoples' quantitative interpretations of BRD.

Recently, Dhami showed that this method was a reliable and valid measure of BRD. In a series of studies, we examined the impact of specific court-level factors, case-level factors, and juror-level factors on interpretations of reasonable doubt. The studies involved individual mock jurors at the predeliberation stage of a trial summarised in written form.

This methodology is typical of jury research. Legal academics and judges have expressed that the undefined version of BRD "the defendant is presumed innocent unless the prosecution has proved guilt beyond a reasonable doubt" is difficult for jurors to understand. In England and Wales for instance the Legal Studies Board advocates the wording, "The defendant is presumed innocent unless the prosecution has proved guilt beyond a reasonable doubt.

Mock jurors gave their interpretation of each standard of proof and rated how helpful they found the instruction. However, for the undefined version we found significant differences in interpretation and understanding across demographic groups. Thus these two wordings of BRD are effective: They decrease demographic group differences in interpretation and in understanding while at the same time leading to the same evidentiary threshold.

A further set of studies investigated the influence of case characteristics on jurors' interpretations of beyond a reasonable doubt. All participants gave their interpretation of reasonable doubt as elicited by the MF method. Two further studies followed-up on the impact of case factors. In one cla ssic case dating from in the Utah Territory, we see precisely such hem hawing from the bench.

Th e judge in a trial for criminal polygam y artfully explained to the jury that a reaso nable doub t is not an unrea sonab le dou bt. Indee d. Arm ed w ith this powerful tautology, the jury convicted the defendant. Defendant s attorney appealed the case, arguing that the judge had failed to explain BARD satisfactorily.

The US Supr eme Cour t uph eld the convic tion, insis ting tha t the jud ge s langu age gives all the de finition o f reason able do ubt w hich a court c an be r equire d to give. The Current State of Play. Although BARD was very widely used in both state and federal courts from the middle of the nineteenth century until the middle of the twentieth, it was there chiefly by custom and common law precedent rather than by law. A famo us rulin g in 0, In re Wing ate, changed all that.

Winga te further made clea r that every key element of the prosecution s case must be established beyond a reasonable doubt before conviction was justified.

In the wake of BARD s increasing centrality in the criminal justice system, trial judges, appellate courts, and even legislators have been trying to make sense of what the doctrine means and, in more practical terms, how they should explain it if at all to puzzled jurors.

From about until about , the most common way of handling it was for judges, following Shaw, to explain to jurors that proof beyond a reasonable doubt meant belief to a moral certainty. We have already seen powerful precedent for this gloss since BARD itself had evolved out of the philosophical literature on mo ral certainty. In the last thirty years, howe ver, the Suprem e Court has discouraged this usage, holding that the terminology is archaic, unhelpful at best, and m isleading, because, as Justice Blackm un put it in a fam ous dissent, there exists the real possibility that such lang uage wo uld lead jurors reason ably to believe that they could base their decision to convict upon moral standards or em otion in addition to or instea d of evidentiary standards.

The risk that jurors w ould und erstand moral certainty to authorize convictions based in part on value judgments regarding the defendant's behavior is particularly high in cases where the defendant is alleged to hav e com mitted a repu gnan t or bru tal crim e. Nebraska , Justice O Connor , writing for a unanimous Supreme Court, said emphatically that this Court does not condone the use of the antiquated moral certainty phrase.

In the same case, Justice Ginsberg opined that the phrase moral certainty, though not so misleading as to render the instructions [to juror s] unc onstitu tional, sh ould b e avoid ed as u nhelp ful in de fining re asona ble doubt. Taking their cues from Ginsberg and O Conno r, several state suprem e courts have recently overturned convictions in cases because the judges instructions referred to mor al certain ty.

This discon nect re veals itse lf not on ly in an u nders tanda ble retice nce to use th e archa ic term moral certainty but, more importantly, in an abandonment of the kinds of proof that m oral certainty represented. The key question , to which w e will return later, is w heth er BAR D, stripp ed of its em bedd ing in a p articula r epistem ic conte xt, is capable of standin g independ ently, that is, whether it is a coheren t notion in its own right.

With the most common traditional definition moral certainty-- excluded or at least sternly frowned on , judges and legal scholars have been casting about for other ways of explaining BARD to jurors. The various versions of BARD that I will reprise are not those proposed by some legal scholar or other; all but one have been extensively used by trial judges and endorsed and sometimes later disow ned b y app ellate co urts: Ever since the nineteenth century, one popular legal explanation of the idea of belief beyon d a reas onab le dou bt has involve d com paring it with im portan t decisio ns in the lives of ordinary citizens.

Juries are told that just as they do not undertake major life decisions unless they are sure of the beliefs undergirding them, so they shouldn t find the defendant guilty unless their confidence in his or her guilt is as great as the confidence they demand in their own important decision-making. The analogy here is that of a prudent person who undertakes an important action, vital to his or her affairs , only w hen v ery con fident a bout th e beliefs th at drive their actions.

We find a typical formulation of this idea in the pattern jury instruction recomm ended by th e Fifth U.

Beyond Reasonable Doubt

Circuit Court: Proof beyond reasonable doubt, therefore, is proof of such a conv incing c harac ter that you w ould b e willing to rely and act upon it without hesitation in the most important of your own affairs. Po nder a typical ra nge of a ctions th at mo st individ uals may count among the most important life decisions they make: Do you give up your c urren t job an d cha nge em ployer s or care ers on ly wh en you are alm ost certa in abou t the res ults of th e action?

Do y ou de cide to m arry on ly wh en you are certa in beyond a rea sonable dou bt that it will work out? If a doctor recomm ends surgery to a patient with a life-threatening disease, does the rational patient refuse to have the surger y unle ss he k now s beyo nd a re asona ble dou bt that it w ill cure w hat ails h im? You do not hesitate to call the police when you suspect you have heard the sound of someone moving around your living room in the middle of the night, even though you su rely ha ve sign ificant an d reaso nable doub ts abou t whe ther th e noise s really came from an intruder.

In our ordinary lives, as I just said, the existence of even sizable doubts often produces no hesitancy to act. The point is that many, perhaps most, of the important life decisions that each of us makes are taken under conditions of significant un certainty, where d oubt is not only ration al but ramp ant.

In these circum stances, we act even though there is considerable doubt. Ironically , that is precisely what the legal system wants jurors to avoid doing when deciding guilt in a criminal trial.

Yet the analogy on offer invites jurors to convict on the same precarious beliefs that typically guide their practical actions. The Suprem e Judicial Court of Massachusetts has forcefully argued against this an alogy fo r und erstan ding B ARD , believin g that d eterm ination s of crim inal gu ilt are typically much more important than decisions citizens ordinarily make: The degree of certainty required to convict is unique to the criminal law. We do not think that people customarily make private decisions according to this standard nor may it even be possible to do so.

Individu als may often h ave the luxury of und oing p rivate m istakes; a verdict o f guilty is freq uently irrevocable. Eventu ally mindful of such problems, the Supreme Court has of late been conceding that the analogy just discussed can be misleading and has proposed an alternative.

Specifically, appellate courts including the Supreme Court have suggested that judge s use a v ariant o f this ana logy, no t to exp lain w hat is a b elief beyond a reaso nable doubt but rather to illustrate what a rational doubt would be.

A rational doubt is the sort of doubt about your beliefs that would cause you to hesitate to act on those beliefs. By contrast, a less than rational doubt an d thus the sort jurors sh ould ignore w ould not cau se you to hesitate.

The courts are saying that the only sort of doubt that a juror should allow to prevent him from finding the accused gu ilty is the sort of doubt that would cause him to hesitate to act in his own personal, but important, affairs.

Here is the language pro posed in the m odel jury instruction en dorsed by the Second Circu it: A reasonable doubt is a doubt based on reason and common sense the kind of doub t that would make a rea sonable person hesitate to act.

Proof beyond a reasonable doubt must, therefore, be proof of such a convincing ch aracter that a reasona ble person w ould not he sitate to rely and act upon it in the most important of his own affairs. Ask ourselves: Man y peop le, wh en con fronted by m ajor life de cisions, fid get an d fret eve n wh en it is wholly clear and beyond doubt what course of action they should take.

A battered wife, who knows perfectly well that her husband will continue beating her when mon ey is sho rt in the hous ehold , will very often h esitate a bout e xtricatin g herse lf from the situation. Does that imply, as the Second Circuit would have it, that she has rea sonab le dou bts abo ut her husb and s violen t streak?

Such inertia, inh erent in many situations, leads to hesitancy and inaction, even when the parties involved have no real doubt that their current situation is contrary to their own best interes ts. Clearly , ordina ry peo ple often hesitate to act ev en lack ing reas onab le doubts. A committee of distinguished federal judges, reporting to the Judicial Conference of the United States, has criticized all such glosses on BARD that assimilate it to ordinary situations of personal choice and decision-making: In the de cisions p eople make in th e most im portant of their ow n affairs, resolution of conflicts about past events does not usually play a major role.

Indeed, decisions we make in the most important affairs of our lives -- choosing a spouse, a job, a place to live, and th e like -- generally involve a very heavy element of uncertainty and risk-taking. They are wholly unlike the decisions jurors ought to make in criminal cases. It is meant to exemplify the old saw that it is better that ten guilty men go free than that one innocent man is condemned. Yet, if a judge tells jurors that they can co nvict if th ey are a s sure o f the de fenda nt s guilt as th ey are a bout th e beliefs that guide their important practical decisions, this is an open invitation to convict on quite low levels of belief in guilt.

Either o f these g losses th us un derm ines th e ration ale for ins isting on guilt beyond a reasonable doubt. Worse, both dilute the reasonable doubt standard to the point of trivializing it since we often act, and it is often rational to act, even when the beliefs driving an important action think of the suspicion of the prowler in the hou se are little mor e than bare po ssibilities.

Ra tional a ctions a bout im portan t life matters are not generally grounded in nearly certain beliefs. Guilty verdicts, by contra st, are m eant to be base d on b eliefs tha t are virtu ally certa in. BAR D as an abiding co nviction o f guilt.

Another common w ay for judges to describe what it is to believe in guilt beyond a reasonable doubt is to suggest that jurors have such a belief w hen th ey hav e an ab iding co nviction that th e defen dant is guilty.

Here, for instance, is a model jury instruction adopted by California courts: Reasonable d oubt is defined as follow s: It is not a mere possible do ubt; because everything relating to human affairs, and depending on moral evidence, is open to some possible doubt or imaginary doubt.

It is that state of the case which after the entire comparison and consideration of all the evidence, leaves the minds of the jurors in that condition that they cannot say they feel an abiding conviction, to a moral certainty, of the truth of the charge. Take n literally , an abid ing con viction is one th at one will hav e for a lon g time as opposed to a tran sitory or fleeting b elief.

Still, wh o can s ay of an y inferen ce-bas ed belie f that he has jus t form ed tha t he w ill not eventually develop second thoughts about it? Retrospec tively, of course, you can say of someone s belief, including your own, that it was an abiding conviction. The issue is h ow, prospectiv ely, a juror ca n dete rmin e that a new ly form ed belie f of hers w ill be one of those. Since, on pain of incoherence, this cannot be what is meant by an abiding conviction, we have to look elsewhere for its meaning than on its surface.

The legal literature on this question tends to suggest that, in asking jurors to decide whe ther th eir belief in guilt is abid ing, the courts really are asking whe ther th ey ho ld the belief with firmness and unwaveringly. That presumably would be an abiding convic tion. On this construal, you can figure ou t here a nd no w w heth er a fresh belief w ill be an a biding one by asking whe ther it is firm ly held.

Still, does th is feature of a belief in gu ilt show that it is a b elief beyo nd a re asona ble dou bt?

Doubt pdf reasonable

Th e firmn ess of a belief, that is, the depth of one s conviction in it, does nothing to settle whether the belief is rational or founded on the evidence. For instance, I may have an abiding belief that my recently departed Fido is now in dog heaven. I may even go to my grave believing it.

Nevertheless, the en durance of this belief or m y current tenacity in asser ting it ha s little to do with w heth er it is ration ally we ll found ed. If it is true, as surely it is, that we don t want jurors to be deterred from convicting someone because of irrational doubts those jurors may entertain about the defendant s guilt, it must be equally true that we don t want to encourage jurors to rush to convict someone based on irrational suspicions of that person s culpability.

Reasonable Doubt

There can be no acceptable exegesis of BARD that fails to make clear that acquittals cannot be grounded on ill-considered foundations any m ore than convictions can. This gloss fails on those grounds.

Doubt pdf reasonable

Saying to a juror that he ought not convict unless he has strong and abiding convictions about the guilt of the accused is unacceptable because it confuses a necessary condition with a sufficient one. We do indeed want jurors to be persuaded of the guilt of the accused before they convict; but such persuasion is of no probative value unless it derives from a careful reflection on, and wise inferences from, the evidence.

True, the m odel California instruction qu oted above en joins the juror to consider all the evidence; but considering all the evidence and weighing it in an appropriate way are very distinct things. Unfortunately, most jury instructions say nothing about the kind of evidence necessary for conviction, focusing instead almost entirely on the subjective state of the juror a point that we will return to in much detail fart her alo ng.

BARD, w e must rem ember, w as designed, am ong other thin gs, to protect a defendan t from the rush to judgme nt that we asso ciate with tribal or vigilante justice. It was meant to codify the meaning of the presumption of the innocence of the accused.

Instructing juries that they should not convict unless firmly convinced that the defendant is guilty does nothing to insure that their beliefs about guilt and innoc ence w ill be base d on a careful a ssessm ent of th e evide nce of th e case. If it is impo rtant to stress th at dou bts, if they lead to a cquitta l, mus t be ratio nal do ubts, it is similarly crucial to insist that beliefs in guilt, if they lead to conviction, must be rationally well-founded.

You may think that the problem could be remedied here by combining the language of abiding convictions with an insistence that jurors must attend carefully to the evidence in coming to their belief about guilt. That is, the judge might say that the evidence must make the defendant s guilt overwhelmingly likely, or some words to that effect as well as referring to an abiding conviction.

Yet if he says th e form er, then the talk abou t an ab iding co nviction is gratuito us. If gen uinely reasonable doubt vanishes when the evidence points powerfully to the guilt of the accused, then the juror need only decide whether the evidence presented is of that character.

It is irrelevant whether, in addition, he has a firm or abiding conviction abou t it. It only further muddies the waters that some courts have argued that the language of firm conviction of guilt represents a seriously weakening of the BARD standard and that it makes the prosecutorial burden much lighter than it had been when moral certainty was required. This construal of BARD appears to be alm ost tau tologo us an d thu s harm less, if unin form ative. On th is view, d oubts for wh ich no reason can be given fa il to be rea sonab le and th us can not leg itimate ly block convic tion.

As o bviou s as it m ay seem , this explication of reasonable doubt has been frequently frowned upon, and sometimes explicitly repudiated by, appellate courts. In Burnett v. Nebraska, for example, the ben ch instructed the ju ry that a reasonable doubt was a doubt that you could explain to your fellow jurors and for which some reason could be given in light of the evidence adduced in the case. The judge instructed the jury, in part: This seemingly innocuous langu age w as faulte d on a ppea l; the ap pellate c ourt h eld tha t this de finition w as in error and the guilty verdict was reversed.

Nebraska, we read: It is error to charge a jury that [reasonable doubt] is a doubt for the ha ving of w hich th e juror c an give a reaso n deriv ed from the test imon y. Oklahoma, an instruction conta ining th e phra se 'a dou bt that y ou can give a rea son for ' is wron g. S ome of the th inking behin d the te nden cy to see this reading of BARD as incorrect comes from an appellate court s review of another case, Morg an v.

Oh io, in which the judge had used the doubt-for-which-you-have-a- reason language. The appellate court asked: What kind of a reason is mean t? Would a poor reason an swer, or must the reason be a strong one?

W ho is to judge? Th e definition fails to enlighten, and further explanation would seem to be needed to relieve the test of indefiniteness. The expression is, also, calculated to mislead. To whom is the reason to be given?

The juror himself? The charge does not sa y so. An d juror s are no t requ ired to a ssign to others reason s in support of their verd ict.

A reasonable doubt may exist in th e min d of a jur or with out h is being able to fo rmu late a rea son. The US Second Circuit Court, in particular, mentions a concern that this instruction might intimidate a juror by suggesting that he may be called u pon to explain his dou bts, alth ough it surely d oes no t requ ire him to justify them.

If a juror fee ls doubtful abou t guilt but cannot iden tify or formulate the reaso n for that doub t, then how can she possibly decide whether the doubt in question is rational or irrational? A second, to my mind equally specious, line of thinking about why reasonable doubts needn t be doubts for which you have a reason is linked to the notion that the accused in a criminal trial is not obliged to mount a defense.

Requiring a juror to have a reason for acquittal, it is said, is like requiring the defendant to present a case. Here is the argument of the appellate court in Iowa v. Who shall determine whether [a juror is] able to give a reason, and what kind of a reason will suffice?

To whom shall it be given? One juror may declare he does not believe the defendant guilty. Under this instruction, another may demand his reason for so thinking. Indeed, each juror may in turn be held by his fellows to give his reasons for acquitting, though the better rule would seem to require these for convic ting.

Th e burd en of fu rnishin g reaso ns for n ot findin g guilt established is thus cast on the defendan t, whereas it is on the state to make out a case excluding all reasonable doubt. Besides, jurors are not bound to give reasons to others for the conclusion reached. There, the appellate court found incorrect an instruction that jurors must have some reason for such doubts as they entertain about the defendant s guilt.

To the contrary, the appeals judges insisted: Such an inst ruction as the o ne w e are co nsider ing can , we th ink, on ly lead to confusion, and to the detriment of the defendant.

You might also like: REASON 5 MANUAL PDF

A juror may say he does not believe the defendant is guilty of the crime with which he is ch arged. Anoth er juror a nsw ers tha t if you h ave a re asona ble doubt of the defendant's guilt give a reason for your doubt. And, under the instruction given in this cause, the defendant should be found guilty unless every juror is able to give an affirmative reason why he has a reasonable doubt of the defendant's guilt.

It puts upon the defen dant th e burd en of fu rnishin g to eve ry juror a reason why he is not satisfied of his guilt, with the certainty which the law requires, before there can be a conviction. There is no such burden resting on the defendant or a juror in a criminal case. W e can su rely gran t that a ju ror, after h earing all the ev idence in a case, may be able to say that she still has lingering doubts about the guilt of the accused.

W hat I think that juror can not possibly do is to certify that thos e doubts are reasonab le doubts unless she can formulate them to herself and scrutinize them, preferably with the aid of fellow jurors. If a juror cannot put her finger on any elem ent in th e prose cution s case th at seem s less tha n fully co nvinc ing, the n it is impossible, even for her, to figure out whether her hesitation to accept a verdict of guilty is rational or not.

Indeed, that was the motive years ago for replacing the older criterion acquit if you have any doubt about guilt with BARD. If a juror s doubt cannot be expressed, we cannot parse it along the axis that separates the rational from the irrational. Nor can the juror. The inevitable corollary of the unwillingness of the courts to distinguish for jurors between reasonable and unreasonable doubts is that BARD has ceased to be a criterion that allows a juror to convict even if she has certain doubts.

Instead, it has become a rule whose import is: If you have any doubts, vote to acquit. The rationality test for doubts has simply dropped out of the picture. The profound insight that initially motivated BARD --that conviction was appropriate even in the face of certain sorts of doubts-- has been abandoned in favor of an ep istem ically na ive view that co nviction require s the to tal abse nce of d oubt.

As for the analogy in this passage between reason-giving by a juror and defense-constructing by the accused, the parallel is a red herring. Requiring a juror to have some reason for doubting the guilt of the defendant in no way requires that reason to have come from testimony offered by the defendant or arguments offered by his counsel. The juror may disbelieve some prosecution witness or doubt some expert witness offered by the state or have any of a thousand other good reasons for doubt.

Requiring the juror to be able to express such doubts to fellow jurors, or at least to herself, is perfectly compatible with the presumption of innocence and the placin g of the fu ll burde n of pro of on th e prose cution.

It is tempting to wonder aloud about what possible model of jury deliberation the courts who take this view of non-reason-giving have in mind. Imagine a case, U. Jo nes. Ha lf an ho ur into jury de liberation , a straw ballot ind icates n ine in favor of convicting and three voting to acquit Jones.

By way of blocking the deadlock, several in the guilty camp review the chief elements of the case, as they see it, for the assembly. The three hold-outs remain silent. Well, says one of the guilty voters, are you persuaded.

Holdouts A, B, and C chorus: Why not? Well, I ca n t put m y finger o n it exac tly. I m jus t not fu lly persuaded. I don t have to give you a reason. I don t even have to have a reason. At this p oint, the y ma y decid e to rep ort to th e judge that th e jury is sp lit and th at pros pects fo r furthe r com mu nication look lim ited.

Th is imag inary sc enario makes a mockery of the notion of jury deliberation. One might have hoped that judges would take steps to see that it will not occur. And of course, they do. Here is a mode l jury instruction from the First Circuit: Each of you mu st decide the case for yourself, but you should do so only after considering all the evidence, discussing it fully with the other jurors, and listening to the view s of the other jurors. Do n ot be afraid to change your opinion if you think you are wrong.

But do not come to a decision simply because other jurors think it is right If it looks at some point as if you may have difficulty in reaching a unanimous verdict, and if the greater number of you are agreed on a verdict, the jurors in both th e ma jority an d the m inority sh ould re exam ine the ir positions to see whether they have given careful consideration and sufficient weight to the evidence that has favorably impressed the jurors who disagree with them.

Happily, not all US courts have been so quick as those quoted above to say that you can genuinely have a reasonable doubt about guilt even if unable to give any re asons for that d oubt.

T he Su prem e Cou rt of Wis consin has h eld spe cifically that: An instruction in a criminal case that a reasonable doubt is a doubt for which a reason can be given based on the evidence in the case, is correct, adding for good measure: A doubt cannot be reasona ble unless a reason therefor exists, and, if such reason exists, it can be given.

According to this approach, much more popular with legal scholars and the public than among judges, the right way to understand what is involved in believing in the guilt of the accused beyond a reasonable doubt is to say th at such a belief m ust be h ighly pr obable. One o bviou s inspira tion for th is approach is found in the civil law standard of proof.

There, courts agree that you should find for one party rather than the other if the preponderance of the evidence favors the former. That is normally taken to mean that the trier of fact should favor plaintiff over defendant if the case of the plaintiff is more likely than not, that is, if it has a probability greater than 0. The criminal standard, of course, is meant to be much higher than the civil standard, so the relevant probabilities would have to be much higher for conviction 0.

Proponents of this construal of BARD argue that judges should instruct juries that they must acquit, unless their confidence or degree of belief in the guilt of the accused is close to certainty in numerical terms, close to 1.

It w ill prove worth our time to investigate with some care their basis for doing so. Proba bilities, of co urse, ru n alon g a scale from 0. It is c ustom ary if not strictly correct to identify a probability of 1. Since as BARD reco gnizes we can not be com pletely certain about an ything to do w ith human affairs, this entails that jurors can never be fully certain of the guilt of an accused party.

This naturally invites the suggestion that the standard for conviction could be defined by specifying a threshold probability, a numerical degree of confid ence, th at belief in guilt m ust rea ch befo re con viction in both se nses o f this term can be justified. Although lower court judges from time to time explain BARD to jurors in just such terms, virtually no appellate court in the land will endorse such a gloss of reason able doubt.

On the contrary, they h old to the view th at any such qua ntification must be stou tly resisted. Reversing a conviction in a trial w here the jud ge had defined BARD in prob abilistic lan guage , an app eals cou rt in Massach usetts insisted: The idea of reason able doubt is not su sceptible to quan tification; it is in heren tly qua litative.

The id ea of rea sonab le dou bt is inhe rently q ualitativ e. Any attem pt to qu antify it may impe rmissibly lower the prosecution's burden of proof, and is likely to confuse rather than c larify. This is no argument against using the langu age of probabilities per se but sim ply against failing to specify that a very h igh pro bability w ould b e requ ired to co nvict.

Th e sam e shor t-sighte d ana lysis afflicts a dissenting opinion of Justices Blackm un and S outer on the h igh court when they argued th at: The w ord 'pro bability' br ings to m ind term s such as 'chan ce,' 'possibility,' 'l ikelihood' and 'plausibility' -- none of wh ich appear to suggest the high level of certainty which is required to be convinced of a defendan t's guilt 'beyond a reasonable d oubt. Becau se these argum ents are so shabby, I think that the real worry among jurists about specifying probabilities must be sought elsewhere.

One source of their w orry is th is: W hile ack now ledgin g in the abstrac t that n o me thod of proo f is infallible and thus admitting in principle that mistakes will occur from time to time, the judiciary has an entrenched resistance to any explicit admission that the system has this inbuilt tolerance for wrongful convictions. Some elements of this problem com e out cle arly in Ju stice Bre nnan s rem arks in a case w e have already discus sed, In re Wingate.

To avo id this problem, he says, the ordinary man in the street must believe that he cannot be sent to jail unless his guilt has been p roved with utm ost certainty. These are remarkab le utterances to be m ade in an age th at has suppo sedly come to terms with the fallibility of hum an jud gme nt.

But leave th at curio us an achro nism to one side. What should be clear is that identifying BARD with any level of probability less than unity would m ark an explicit admission that the system officially condones a certain fraction of wrongful convictions.

That, in turn, would supposedly threaten the ordinary person s faith in the justice system. Better, it seems, to avoid any talk of probability as the standard for conviction than to acknowledge publicly that the system expressly expects incorrect judgments of guilt. Daniel Shaviro has argued that the criminal justice system goes to heroic lengths to reduce the appearance of the possibility of error, even while ackn owledging th at such a possibility is omn ipresent.

He suggests that this may reflect the self-serving interest of lawyers [and, we may add, judges] in promoting the appearance of justice under the legal system rather than the true state of affairs. The courts want to eat their cake and have it too. They acknowledge that juries make m istakes from time to time.

They even say to jurors that no certain ty can be had in hum an affairs. On the othe r hand, they ap parently don t want jurors to bring to their deliberative process the epistemic savvy that the judges themselves have.

They want jurors to be fully convinced even when they, as judges, know that even a fully convinced jury can make m istakes and when they know that no one has the right to claim to be fully convinced about the kinds of complex decisions that juries must usually make.