For C.S., who drew my attention to the similarity between Blackstone’s formulation and that of Maimonides.
God, in this past week’s Parshah:
מִדְּבַר-שֶׁקֶר, תִּרְחָק; וְנָקִי וְצַדִּיק אַל-תַּהֲרֹג, כִּי לֹא-אַצְדִּיק רָשָׁע.1
Rambam, in explanation of the above verse:
שהזהירנו שלא לחתוך הגדרים באומד הדעת החזק ואפילו היה קרוב אל האמת. כמו שיהיה אדם ירדפהו שונאו להרגו ולהנצל ממנו יכנס בבית אחד ויכנס הרודף ההוא אחריו ונכנס אנחנו אחריהם ונמצא הנרדף הרוג והוא מפרפר ושונאו שהיה רודפו עומד עליו והסכין בידו ושניהם מנטפין דם. הנה זה הרודף לא יהרגוהו הסנהדרין על צד חתוך הגדר אחר שאין שם עדים מעידים שראו ההריגה.
ובאה האזהרה בתורת האמת מהרוג זה והוא אמרו יתעלה ונקי וצדיק אל תהרוג כי לא אצדיק רשע …
ולא תרחיק זה ולא תפלא מזה הדין. כי הדברים האפשריים מהם קרובי האפשרות מאד ומהם רחוקי האפשרות ומהם אמצעיים בין זה לזה. ולאפשר רוחב גדול מאד. ואילו התירה התורה לחתוך דיני נפשות באפשר הקרוב מאד שאפשר שיהיה קרוב מן המחוייב המציאות כגון זה שהמשלנו היינו חותכים הגדר במה שהוא רחוק מזה מעט ובמה שהוא יותר רחוק גם כן עד שיחתכו הגדרים וימיתו האנשים פעמים במעט אומד לפי דמיון הדיין ומחשבתו. ולכן סגר יתעלה את הפתח הזה ואמר שלא ייחתך גדר העונש אלא כשיהיו העדים מעידים שהם ידעו בודאי שזה עשה המעשה ההוא באמת בלא ספק ובלא דמיון כלל.
וכאשר לא נחתוך הגדרים בדמיון החזק מאד הנה תכלית מה שיהיה שנפטור החוטא וכאשר חתכנו הגדרים בדמיון ובאומד הנה פעמים נהרוג נקי יום אחד. ולזכות אלף חוטאים יותר טוב ונכסף מהרוג זכאי אחד יום אחד.2
The great fifteenth-century English jurist, Sir John Fortescue, in his celebrated classic De laudibus legum Angliæ (written for the instruction of the young Edward of Westminster, Prince of Wales):
It becomes now absolutely necessary to inquire thoroughly how the Laws of England come at the truth in cases criminal; …
If any suspected person who stands accused for felony or treason committed in England, denies the crime of which he stands accused, before his Judges: the Sheriff of the County where the fact is committed, shall cause to come before the Judges twenty-four good and lawful men of the neighbourhood to the Vill where the fact was done, who are in no wise allied to the person accused, who have lands and revenues to the value of an hundred shillings; and they are to certify to the Judges upon the truth of the fact, wherewith the party is charged. Upon their appearance in Court, as they come to the book to be sworn, before they be sworn, the person accused may challenge them, …
Further, in favour of life, he may challenge five and thirty; such as he most feareth and suspecteth, who upon such challenge shall be struck out of the Panel, or such marks set over against their names, that (to use the term in law) they shall not pass upon him in trial; and this peremptorily, without assigning any cause for such challenge; and no exceptions are to be taken against such his challenge: who then in England can be put to death unjustly for any crime? since he is allowed so many pleas and privileges in favour of life: none but his neighbours, men of honest and good repute, against whom he can have no probable cause of exception, can find the person accused, guilty.
Indeed, one would much rather that twenty guilty persons should escape the punishment of death, than that one innocent person should be condemned, and suffer capitally.
Neither can there be any room for suspicion, that in such a course and method of proceeding, a guilty person can escape the punishment due to his crimes; such a man’s life and conversation would be restraint and terror sufficient to those who should have any inclination to acquit him; in a prosecution, carried on in this manner, there is nothing cruel, nothing inhuman; an innocent person cannot suffer in life or limb; he has no reason to dread the prejudices or calumny of his enemies, he will not, cannot, be put to the rack, to gratify their will and pleasure. In such a Constitution, under such laws, every man may live safely and securely. Judge then, good Sir! which law is rather to be chosen, putting yourself in the private capacity of a subject.3
The influential Puritan Minister Increase Mather, in his Cases of Conscience concerning evil SPIRITS Personating Men, Witchcrafts, infallible Proofs of Guilt in such as are accused with that Crime. All Considered according to the Scriptures, History, Experience, and the Judgment of may Learned men:
It were better that Ten Suspected Witches should escape, than that one Innocent Person should be condemned. That is an old saying and true, Prestat ream nocentum absolvi, quam ex prohibitis Indiciis & illegetima probatione condemnari It is better that a Guilty Person should be ABSOLVED, than that he should without sufficient ground of Conviction be condemned. I had rather judge a witch to be an honest women, than judge an honest women as a witch. The Word of God directs men not to proceed to the Execution of the most capital offenders, until such time as upon searching diligently the matter is found to be a truth, and the thing certain. Deu. 13. 14. 15.4
The definitive treatment of this topic is unquestionably Prof. Alexander Volokh’s incredibly erudite, encyclopedic, brilliant and utterly hilarious paper n Guilty Men. Excerpts (but the entire thing is a must-read):
“Better that ten guilty persons escape than that one innocent suffer,” says English jurist William Blackstone. The ratio 10:1 has become known as the “Blackstone ratio.” Lawyers “are indoctrinated” with it “early in law school.” “Schoolboys are taught” it. In the fantasies of legal academics, jurors think about Blackstone routinely.
But why ten? Other eminent legal authorities through the ages have put their weight behind other numbers. “One” has appeared on Geraldo. “It’s better for four guilty men to go free than one innocent man to be imprisoned,” says basketball coach George Raveling. But “it’s better to turn five guilty men loose than it is to convict one innocent man,” according to ex-Mississippi executioner and roadside fruit stand operator Thomas Berry Bruce, who ought to know. “It is better to let nine guilty men free than to convict one innocent man,” counters lawyer Bruce Rosen from Madison, Wisconsin. Justice Benjamin Cardozo certainly believed in five for execution, and allegedly favored ten for imprisonment, which is a bit counterintuitive. Benjamin Franklin thought “that it is better [one hundred] guilty Persons should escape than that one innocent Person should suffer.” Mario Puzo’s Don Clericuzio heard about letting a hundred guilty men go free and, “struck almost dumb by the beauty of the concept . . . became an ardent patriot.” Denver radio talk show host Mike Rosen claims to have heard it argued “in the abstract” that it’s better that 1000 guilty men go free than one innocent man be imprisoned, and comments, “Well, we get our wish.” …
n guilty men, then. The travels and metamorphoses of n through all lands and eras are the stuff that epic miniseries are made of. n is the father of criminal law. This is its story. …
According to some researchers, though, the maxim is considerably older. At least three commentators — one Hebrew prophet, one Founding Father and one appellate judge — have dated it back to the beginning of time. Moses’ precept, from the book of Exodus, was supposedly handed down from Someone who was around “in the beginning.” Benjamin Franklin claims that the maxim, with n = 100 and for suffering, “has been long and generally approved; never, that I know, controverted.” According to Ninth Circuit Judge Alex Kozinski, the “popular notion” that n = 10 (for conviction) is just something “we have always said.” A then-future U.S. president, John Adams, was more modest and merely dated the saying (with n = “many” and for suffering) back to the beginning of laws, saying that “there never was a system of laws in the world, in which this rule did not prevail.” …
In 1471, English chief justice John Fortescue suggested n = 20 for execution: “Indeed I would rather wish twenty evil doers to escape death through pity, than one man to be unjustly condemned.” It was apparently widely believed in English courts during the Middle Ages and the Renaissance that it was better to let many guilty men escape than to convict one innocent person, and a form of Fortescue’s maxim was cited in a 1607 case from the Star Chamber court. In the seventeenth century, Matthew Hale used n = 5 for execution, “for it is better five guilty persons should escape unpunished, than one innocent person should die.” Hale admitted that this doctrine had certain inconveniences; in particular, that it is hard to get good direct evidence of witchcraft, so that many undoubtedly guilty persons escape. As Increase Mather put it in 1692, during the Salem witch trials, “it were better that ten suspected Witches should escape, than that one innocent Person should be Condemned.”
Edward Seymour, in 1696, favored n = 10 for suffering. Seymour reportedly declared, “I am of the same opinion with the Roman, who, in the case of Catiline, declared, he had rather ten guilty persons should escape, than one innocent should suffer”; though Lieutenant General Mordant is said to have replied, “The worthy member who spoke last seems to have forgot, that the Roman who made that declaration was suspected of being a conspirator himself.”
Then, in the 1760s, came Blackstone. Blackstone, it seems, wrote his Commentaries with a bottle of wine by his side, and his doubling of Hale’s n = 5 may have been a case of “seeing double.” The maxim had become part of common law by 1802. By 1823, Blackstone’s doctrine had become a “maxim of English law” and was cited in judicial opinions, though Thomas Starkie used n = “ninety-nine (that is, an indefinite number)” as “the maxim of the law” in his book on evidence only the following year. (For an indefinite number, though, Starkie’s ninety-nine seems quite definite.) John Stuart Mill also allegedly endorsed the maxim in an address to Parliament in 1868.
Of late, British courts have taken both the position that n = 1 and that n = 10, and allegorically refer to the dilemma as “trying to steer between the Scylla of releasing to the world unpunished an obviously guilty man and the Charybdis of upholding the conviction of a possibly innocent one.” Some British laymen have been more generous, though. London Metropolitan Police Commissioner Sir Peter Imbert has expressed a belief that n = 100, while ex-police superintendent Ian McKenzie, once a police officer and later a doctor of psychology with the Fort Worth, Texas, police department, told BBC television that n = 5000, leaving some Britons to ask how the ratio can have risen 500 times or more since earlier, more innocent times. “There has been some inflation” since Hale.
The maxim has apparently made its way throughout the former British empire, to Canada, Australia, and Hong Kong. Hong Kong, which has now been returned to China, actually got off to a halting start; in 1857, during the “incident of the poisoned bread,” attorney general Thomas Chisholm Anstley said in open court, “Better hang the wrong men than confess that British sagacity and activity have failed to discover the real criminals.”5 More recently, politician Martin Lee, of the United Democratic party, held that n = 99, though politician Elsie Tu disagreed: “What I want is justice for that one innocent man, but not a free ride for the [ninety-nine] guilty ones.” …
Whoever the innocent man is, he can also naturally be an innocent woman. The literature, in a less self-conscious time, usually said “man,” but today usually says “person” or “defendant.” In the early modern period, many commentators wrote of guilty and innocent “perfons,” warning that “prefumptive evidences fhould be warily preffed.” What exactly a perfon is may be a fruitful fubject for further refearch.
Increase Mather wrote that he “would rather judge a Witch to be an honest woman, than judge an honest woman as a Witch,” but this is because most witches at the time seem to have been female. One article, on the appropriateness of punishing people with multiple personalities (only one of which may be guilty), points out that “unless one could devise punishments that punished only the guilty self in some body, multiple selves within the same body would face the legal system with the choice between a radically extended system of vicarious responsibility, or not punishing anyone.” The author suggests the maxim, “Better to let ten guilty selves go free than to punish one innocent self.” …
Jeremy Bentham, founder of utilitarianism, warned against the warm fuzzy feeling that comes from large values of n:
We must be on guard against those sentimental exaggerations which tend to give crime impunity, under the pretext of insuring the safety of innocence. Public applause has been, so to speak, set up to auction. At first it was said to be better to save several guilty men, than to condemn a single innocent man; others, to make the maxim more striking, fix the number ten; a third made this ten a hundred, and a fourth made it a thousand. All these candidates for the prize of humanity have been outstripped by I know not how many writers, who hold, that, in no case, ought an accused person to be condemned, unless evidence amount to mathematical or absolute certainty. According to this maxim, nobody ought to be punished, lest an innocent man be punished.
Some less theoretical minds went somewhat further in their skepticism toward the n maxim. German chancellor Otto von Bismarck is said to have remarked that “it is better that ten innocent men suffer than one guilty man escape.” Feliks Dzerzhinsky, founder of the Soviet secret police, saw Bismarck’s motto and raised him an execution: “Better to execute ten innocent men than to leave one guilty man alive.” Dzerzhinsky apparently did not elaborate on the rationale for this sort of treatment. Nor did Nikolai Yezhov, one of his like-minded successors, except to quote the Russian proverb, “When you cut down the forest, woodchips fly.” For more information, the interested reader is referred to Major Nungo, a Colombian military prosecutor, who said, “For us military men, everybody is guilty until proved otherwise. . . . Better to condemn an innocent man than to acquit a guilty one, because among the innocent condemned there may be a guilty man.”
[Volokh now presents a lengthy and detailed analysis of the values of n in American state and Federal law, and then proceeds:]
Advice for Criminals
Criminals, therefore, are advised to go to New Mexico (n = 99) or Oklahoma (n = 100). Criminals who had planned on going to Ohio or Virginia, hoping to find n = 99 there, may want to reconsider. Criminals who like to live on the edge may want to take their chances with n = “many” in Idaho, Kentucky, New Jersey, or Rhode Island. They may also want to try out New York, but this could be risky.
Criminals who are planning to violate federal law should go to the D.C. Circuit or the Second Circuit (which contains Connecticut, New York, and Vermont). The abnormally high crime rate in Washington, D.C. (41% higher than that of its nearest competitor, Florida, and 72% higher than the ninety-fifth percentile of state crime rates) suggests that many criminals already have gone there. To be on the safe side, criminals who go to D.C. to violate federal law should run for public office.
Whether there really is a relationship between high values of n and high crime rates is controversial. “Tough-on-crime” types believe that there is a positive relationship between n and c, or that high values of n — a high presumption of innocence — lead to high values of c — an increased incidence of crime. Others believe, however, that n and c are negatively related — that punishment may be counterproductive, and that low values of n can lead to high values of c. We would like to find a mathematical function that relates n and c, for example, c = an + b, to know precisely how different values of n will affect c. Unfortunately, we do not know at the outset what form this function will take. It may, for instance, look like 1 / c = an squared + b, or ln c = a / n + b. Standard statistical programs can tell us the values of a and b that best fit the data, but first we must guess at the precise form of the function. We tried different possible mathematical functions until we found a functional form that fit the data better than others. Using values of n as revealed by state court opinions, and FBI data on crime rates in different states, 205 our trial-and-error method yields the following possible model:
c squared = -1,251,677 ln n + 30,217,466
c is measured in cases per 100,000 population. The numbers — in this case, -1,251,677 and 30,217,466 — were chosen to make the equation fit the real-world numbers as closely as possible. In this model, n and c are negatively related — as n (the presumption of innocence) goes up, c (the crime rate) goes down. 206 Knowing n allows us to predict c, and, conversely, knowing current crime rates, c, allows us to know n, that is, to tell how strong the presumption of innocence is.
The current national crime rate is 5,482.9 crimes per 100,000. Setting c = 5,482.9, we find that n = 1.132 — a slight presumption of innocence. If Blackstone were in charge of the criminal justice system, of course, n would be 10, and so c would be 5,228.3 crimes per 100,000, about 4.6% better than the current rate.
The data, however, are inconclusive. Another model, which fits the data about as well, instead shows n and c positively related — with crime rates going up as the presumption of innocence increases: 1 / c squared = – (5.6 ‘ 10^-10 ln n) + (4.75 ‘ 10^-8)
In this model, the Blackstonian crime rate — the value of c corresponding to n = 10 — is 4,651.9. Setting c = 5,482.9, the national average, we find that the corresponding value of n is 1.0967 ‘ 10^11, or approximately 109,670,000,000 (109.67 billion guilty men acquitted for every one innocent man convicted). That’s a lot of presumption of innocence. …
Juvenal and Goethe believed that the guilty are never really acquitted. To such people, the question is somewhat moot. Franz Kafka doubted the existence of guilt; Archibald MacLeish doubted the existence of innocence. Albert Camus doubted both, and Pontius Pilate doubted truth. To these, too, our question cannot help but be a little beside the point. Still others sidestep the issue, as in the case of former British prime minister John Major, who told a Tory party conference that it is better to put the guilty behind bars than to imprison the innocent in their homes.
Others take a still different tack. Just convict all the guilty and acquit all the innocent, say letter writers, state supreme courts, Ulysses S. Grant, and the Chinese. Blackstone’s maxim “supposes a dilemma which does not exist: the security of the innocent may be complete, without favouring the impunity of crime,” said Jeremy Bentham. “The law recognizes no such comparison of numbers,” the Alabama Supreme Court has reasoned (debatably), adding (perhaps more soundly) that “the tendency of such a charge, unexplained, is to mislead” jurors; it has often upheld trial courts’ decisions not to offer the maxim, holding that the maxim is “merely argumentative.” Other state supreme courts, including those of California and Illinois, have agreed. Thomas Starkie called the notion that “moral probabilities could ever be represented by numbers . . . and thus subjected to arithmetical analysis . . . chimerical,” but that did not stop him from developing such an analysis anyway.
The story is told of a Chinese law professor, who was listening to a British lawyer explain that Britons were so enlightened, they believed it was better that ninety-nine guilty men go free than that one innocent man be executed. The Chinese professor thought for a second and asked, “Better for whom?”
Unfortunately, the Formulation, although high-flown and noble sounding, is fundamentally flawed to the point of practical uselessness, although it may still serve as an inspirational guidepost. As Bentham points out, assuming infinitely large n yields the reductio ad absurdum that all (capital) punishment is illegitimate, but in the real world it is pretty clearly impossible to actually specify a meaningfully precise, finite value for n, so we are left with just a vague exhortation to err on the side of acquittal, but not much guidance in determining exactly how far to go in that direction.