©   1999  Kenneth Pennington  

Innocent Until Proven Guilty: The Origins of a Legal Maxim

  Ken Pennington

In spite of what you have been reading and hearing in the media during the last year, the presumption of innocence, or even its existence, has not been on the lam. In fact, one might argue it's done quite well in the second half of the twentieth century. The United Nations placed the principle that a defendant is innocent until proven guilty in its Declaration of Human Rights in 1948 under article eleven, section one. The maxim also found a place in the European Convention for the Protection of Human Rights in 1953 [as article 6, section 2] and was incorporated into the United Nations International Covenant on Civil and Political Rights [as article 14, section 2].

The United Nations' declaration established "Innocent until proven guilty" as a right in the modern world. Yet there are few maxims that have a greater resonance in Anglo-American jurisprudence. But, in spite of our reverence for it, the maxim cannot be found in Magna Carta, the English Bill of Rights of 1689, the Declaration of Independence, or the Constitution of the United States; or, I might add, in the works of the great English jurists, Bracton, Coke, or Blackstone. Nevertheless, some scholars have claimed that the maxim has been firmly embedded in English jurisprudence since earliest times.

Claims about the maxim's Anglo-Saxon roots are sometimes quite stirring and display a particularly British capacity to create intellectual Camelots -- on their side of the Channel. An English scholar named Clementi gave a talk on the maxim at Göttingen, Germany in 1974.(1) He informed his learned, continental audience about the maxim's unique Anglo-Saxon origins. When a fourteenth-century English Parliament declared its complete independence from Roman law, the principle on which Parliament stood, was, he said, "that an accused person must be deemed innocent until such time as his or her guilt has been proved conclusively in court." Clementi declared that he would not have bothered his continental colleagues with "intimate details of English pragmatism" before England had entered the Common Market. Now, however, he assured them that "there is good reason why you should become better acquainted with us." English devotion to the principle of 'Innocent until proven guilty' served to "emphasize a separation between England and its European mainland in matters of law." With missionary zeal, Clementi propounded the virtues of innocence while being guilty of explicating a text in which the concept was completely absent.

Clementi did not know that the maxim "Innocent until proven guilty" cannot be found in any English court case or any tract on jurisprudence before ca. 1800 --- at least I have not yet found it in one. He also did not seem to know that the French, in spite of their legal system's being based on rebarbative Roman jurisprudence, did include an article in the French Declaration of the Rights of Man and Citizen of 1789 stating that "every man is presumed innocent until declared guilty." These facts raise two questions that will be the subject of this talk: how did this piece of English pragmatism become a part of the Romanist French tradition and how and when did the maxim surface in the Anglo-American tradition?

We can know exactly when the maxim formally entered American law: through a Supreme Court decision of 1894, Coffin vs. U.S. The court had heard a case from Indiana in which Francis A. Coffin and Percival A. Coffin had been convicted of having aided and abetted the President of the Indianapolis National Bank in the commission of misapplication of funds and making false entries in the bank's books. In other words, they defrauded the bank. The forty-fourth charge in the case was rejected by the lower court in its instruction to the jury. The appeal was based in part on the lower court's refusal.(2)

The forty-fourth charge stated: "The law presumes that persons charged with crime are innocent until they are proven by competent evidence to be guilty." Although the court rejected this charge, it did instruct the jury that "Before you can find any one of the defendants guilty you must be satisfied of his guilt as charged in some of the counts of the indictment beyond a reasonable doubt." The lower court then instructed the jury at great length on the doctrine of reasonable doubt and its relationship to evidence. The Supreme Court saw its task as determining whether the lower court had violated the defendants' rights by not instructing the jury on presumption of innocence and whether reasonable doubt was essentially the same as presumption of innocence.

Justice Edward Douglas White wrote the majority opinion. For a legal historian, his analysis is a dazzling display of legal history --- even if most of it is not correct. To prove the antiquity of "Innocent until Proven Guilty" White cited a story from the late antique Roman historian, Ammianus Marcellinus, Justinian's Digest and Code, Pope Gregory IX's Decretales, a decretal of Pope Innocent III,(3) and Giuseppe Mascardi's De probationibus, all, except for Ammianus, from the continental law.(4) None of the texts, unfortunately, contained the maxim. Not one of them, you may note, was from English law. Puzzling.

When White turned to the Anglo-American tradition, he found the principle clearly articulated in a number of nineteenth-century treatises on evidence and criminal law. The jurists White cited were William Wills', († 1860) On circumstantial Evidence, Simon Greenleaf's, On the Law of Evidence (1783-1853), and William Best's, (1809-1869) On Presumptions. Of these jurists Best is the only one who explicitly states that it is a "maxim of law, that every person must be presumed innocent until proven guilty."

Justice White did try and trace the maxim in the English common law tradition but could only find one piece of evidence. He cited an anonymous author of an article in the North American Review of 1851 who stated that the maxim is first found in a treatise on evidence by an Irish jurist named Leonard MacNally. White concluded that even "if the principle had not yet found formal expression in the common law writers at an earlier date, yet the practice which flowed from it has existed in the common law from earliest time."(5) Another piece of evidence that Clementi's faith was not unique.

Through Coffin v. U.S. Justice White ordained Leonard MacNally (1752-1820) as the midwife of "Innocent Until Proven Guilty's" entrance into the American common law tradition.(6) Who was he? He was born in Dublin in 1752. After his father died in 1756, he spent a part of his childhood in Bordeaux. At 19 he opened a grocery shop in Dublin. An ambitious sort, he was called to the Irish bar in 1776 and to the English in 1783. At the same time he began to write lyrics for musicals, some of which were performed in Covent Garden and other London theaters. In 1779 "The Apotheosis of Punch: A Satirical Masque" was performed, followed by thirteen other plays between 1779 and 1789. In anticipation of the pullulation of romantic medieval themes in the nineteenth century, he entitled one play "Robin Hood, or Sherwood Forest, a comic opera" and another "Richard Coeur de Lion: An Historical Romance." Although light fare, sort of a Kmart Gilbert and Sullivan, MacNally does merit a mention in The Grove Dictionary of Music.

The anonymous author of the Dictionary of National Biography's article on MacNally alleged that he was "no great lawyer" but an "astute and eloquent advocate." His dismissal of MacNally's legal skills does the Irish barrister a grave disservice. The DNB author did not realize that MacNally's The Rules of Evidence on Pleas of the Crown illustrated from Printed and Manuscript Trials and Cases, published in Dublin and London 1802 was immediately transported across the Atlantic and printed in Philadelphia 1804 and reprinted in 1811. One cannot read American treatises on evidence and presumption in the first half of the nineteenth century without stumbling over MacNally.

MacNally was particularly important for the development of rules governing evidence and procedure in criminal cases because he had represented a number of United Irishmen accused of treason. He quotes a large number of his own cases in his book. It is no fluke that treason led MacNally to consider the rules of evidence more carefully than previous writers. The cases that society has found most heinous have always been those in which the rules of fair and just procedure have come under attack.

The rules of procedure for cases of treason were still substantially different from the normal rules of criminal procedure. Even during MacNally's lifetime the same rules of due process were not extended to Irish defendants in trials of treason. Although two statutes of King Edward VI and another of William III required two witnesses for any conviction of treason, this procedural nicety was not extended to Ireland. MacNally emphasized the presumption of innocence for those accused of treason and justified applying the same rules of due process to them as to other defendants of criminal offences. His defense of Irish rights was rhetorically compelling:

If these English statutes were enacted because in cases of treason the oath of allegiance counterpoises the information of a single witness, is not an Irishman intitled(sic) to the benefit of that reason? --- or, if the principal reason for enacting those statutes was, as sir William Blackstone states, "To secure the subject from being sacrificed to fictitious conspiracies which have been the engines of profligate and crafty politicians in all ages," why should not Irishmen be granted the same security, from such conspiracies and the machinations of such politicians? The imperial parliament have to discuss and determine those questions at a future day.(7)

He never used the maxim "Innocent until Proven Guilty," but he argued vehemently for the rights of defendants, often using examples from cases in which he had participated. MacNally came very close to stating the principle when he discussed the two witness rule for cases of treason by citing Cesare Beccaria.

In Beccaria's judgment, one witness is not sufficient; for whilst the accused denies what the other affirms, truth remains suspended, and the right that every one has to be believed innocent turns the balance in his favour.(8)

A century later Justice White may have used this passage from MacNally (he does not give a specific citation) to plant the doctrine of presumption of innocence in American jurisprudence.

MacNally's story does however have a darker side. After his death in 1820 the English press revealed that MacNally had played the role of a double agent since at least 1794. At the same time he represented Irish revolutionaries in court, he was betraying them to the government by passing on key information about their accomplices and their organization. He relayed all the details about the revolutionary activities that he learned from his clients to the government prosecutors. From 1800 until his death he received 300l. a year for his trouble. Of this side of MacNally, Justice White knew nothing.(9)

One may ask, from where did MacNally get his principles? MacNally acknowledged Beccaria, and, indeed, Cesare did extoll presumption of Innocence several times in his famous treatise, Dei delitti e delle pene (On crimes and punishments). He argued for always having two witnesses before one could be condemned for a criminal offence:

More than one witness is needed, because, so long as one party affirms and the other denies, nothing is certain and the right triumphs that every man has to be believed innocent.(10)

A few pages later, Beccaria repeated the same argument when, in the most passionate page of his tract, he assailed torture.(11)

either the crime is certain or it is not; if it is certain, then no other punishment is called for than what is established by law and other torments are superfluous because the criminal's confession is superfluous; if it is not certain, then according to the law, you ought not torment an innocent because such is a man whose crimes have not been proven.(12)

MacNally relied on Cesare Beccaria to justify presumption of innocence.(13) The short version of the story is that presumption of Innocence entered American law thanks to a traitorous Irishman who learned it from a reclusive Italian. But the story is much longer and more complicated than the obvious link that I have shown between Beccaria and MacNally. Presumption of innocence had a long history that stretches back to the twelfth century. To learn about this history we must turn to the jurisprudence of the Ius commune.

The Ius commune was the common law of Europe from the twelfth to the seventeenth centuries. It was formed by the fortuitous and contingent conjuncture of Roman law, canon law, and, later, feudal law in the schools and courts of medieval Europe. Its birth took place in an age when momentous changes in the practice of law were taking place. Law was evolving from unwritten customary usages to written customary and legislated law. Judicial procedure was in a state of great flux. Prior to the twelfth century the judicial ordeal was a pervasive mode of proof. During the course of the twelfth century, particularly in Southern Europe, the ordeal was replaced by the ordo iudiciarius , a mode of proof that was based on Roman law. The consequences were unsettling.(14) Procedure is the central part of any legal system. A society's sense of justice is intimately linked to its modes of proof. As the ordo iudiciarius was imposed on Europe's courts by ecclesiastical and secular authorities, there is clear evidence that all strata of society had questions about its legitimacy.

Medieval conceptions of rights were intimately connected with judicial procedure. Even in the world of the ordeal, a man or woman had the right to prove his or her innocence. Although the ordeal lacked a jurisprudence and a set of written norms, literary sources give us insight into their presumptions. In the Romance of Tristan that is part of the Arthurian cycles, King Mark condemned Tristan and Isolt to death without a trial when they were caught in flagrante delicto. The poet tells us that the people of the Kingdom were dismayed because they had not been tried by the proper judicial procedure, in this case an ordeal. The people cried out: "King, you would do them too great a wrong if they were not first brought to trial. Afterwards put them to death." Tristan and Isolt had been caught with their clothes off and their defenses down. Neither they nor King Mark had ever formulated the thought: Ah, yes, I'm innocent until proven guilty, but the story of Tristan and Isolt pinpoints a crucial problem that has hindered the development of a presumption of innocence in every legal system. Do the perpetrators of flagrant, notorious, public, and observed crimes have a right to a trial? Are they whom everyone presumes guilty because every knows they are guilty still deserving of a presumption of innocence?

The jurists of the Ius commune did create norms that a defendant must be canonically summoned and publicly convicted. A few texts in Roman law supported a defendant's right to be heard in court. In his famous decretal Venerabilem, Pope Innocent III stated that if defendants had not been cited, witnesses could not present testimony against them.(15) In another decretal, cited by Justice White in Coffin vs. U.S., the same Pope Innocent formulated a general presumption that a person should always be presumed to be of good reputation.(16)

Although the general principle of presumption of innocence was well established in the jurisprudence of the Ius commune by the beginning of the thirteenth century, the right was not absolute. Notorious crimes provided the most clear infringement of the right. The jurists agreed that when a crime was heinous and notorious a judge could render a decision against a defendant without a trial. In the middle of the thirteenth century, one of the most distinguished jurists of the age, Henricus of Segusio, declared that notorious crimes, especially those committed against the Church, needed no formal juridical examination.

At the same time as jurists were creating a jurisprudence that recognized a presumption of innocence for defendants, Europe's courts were making a transition from the ordeal to the ordo iudiciarius in the twelfth century. Although founded on Roman law, the ordo was new. It takes a leap of our imaginations to understand the turmoil this change must have created. We might project this turmoil into our own lives if we could imagine how we would react if the court system suddenly replaced juries with panels of judges. Jurists of the twelfth century needed to justify the radical changes in the modes of proof that were taking place. The jurists found their justification in the Old Testament and ingeniously traced the origins of the ordo iudiciarius to God's judgment of Adam and Eve in paradise. By doing so, they created a powerful myth justifying the ordo that retained its explanatory force until the seventeenth century.

Around 1150 a jurist named Paucapalea was the first to link the ordo iudiciarius to A am and Eve. He noted that the ordo originated in paradise when Adam pleaded innocent to the Lord's accusation. In Genesis 3.9-12, the Lord burst into Paradise and shouted: Adam ubi es? One may note that for a deity his question was not particularly omniscient. Adam responded to the Lord's accusation of illegal apple picking by complaining to God that "My wife, whom You gave to me, gave <the apple> to me, and I ate it." God had, in other words entrapped Adam by giving him a wife (not the only time that idea has popped into the heads of our dumber gender). Paucapalea's point is subtle but would not be lost on later jurists. Although God is omniscient, he too must summon defendants and hear their pleas. Paucapalea added another piece of evidence that the ordo arose from the Bible. When Moses decreed that the truth could be found in the testimony of two or three witnesses, he pronounced a basic rule of evidence and confirmed the antiquity of a system of procedure accepted by God himself (Deuteronomy 19.15). A few years later (ca. 1165) Stephen of Tournai further dissected the "trial" of Adam and Eve finding even more evidence that this event marked the beginning of the ordo iudiciarius. He pointed out that Adam raised a formal objection (exceptio), to the Lord God's complaint (actio) and shifted the blame on his wife or the serpent. "Exceptio" and "actio" were technical terms that were fundamental parts of the ordo iudiciarius. Stephen was the first jurist to define the ordo iudiciarius:

The defendant shall be summoned before his own judge and be legitimately called by three edicts or one peremptory edict. He must be permitted to have legitimate delays. The accusation must be formally presented in writing. Legitimate witnesses must be produced. A decision may be rendered only after someone has been convicted or confessed. The decision must be in writing.

This litany of admonitions indicates that by the second half of the twelfth century, the jurists were conscious of a defendant's right to a trial and his right to have his trial conducted according to the rules of the ordo iudiciarius --- limiting to right of the judge to act arbitrarily. Most importantly for our story, the subtext clearly implies that if God must summon litigants to defend themselves, mere humans must also, presuming that every defendant is innocent until proven guilty in court.

However, presumption of innocence was not yet a right. Notorious crimes still trumped innocence in the jurisprudence of the Ius commune. One more crucial change had to occur. The biblical origins of the ordo iudiciarius enabled that change to occur. Before the middle of the thirteenth century jurists accepted the right of the prince to subvert the judicial process because they considered legal procedure to be a part of the civil law, that is positive law, and, therefore, completely under the prince's authority. They traced the idea that procedural rights were derived from positive law back to ancient Roman jurisprudence. Paucapalea and Stephen of Tournai, however, introduced a different story and a different paradigm. If the ordo iudiciarius can be first found in the Old Testament, and if God had to respect the rights of defendants, then the rules of procedure must transcend positive law.

The jurists were not slow to see the implications of Paucapalea's and Stephen's new paradigm. During the second half of the thirteenth century they began to argue that the judicial process was not derived from civil law, but from natural law or the law of nations, the ius gentium. A paradigm shift occurred. The story is interesting but complicated. So I'll skip most of it.

The main point is that once the jurists decided that the norms of procedure were part of natural law, they quickly saw that essential rights of defendants could not be transgressed. The most sophisticated and complete summing up of juristic thinking about the rights of defendants in the late thirteenth and early fourteenth centuries is found in the work of a French canonist, Johannes Monachus who died in 1313. While glossing a decretal of Pope Boniface VIII (Rem non novam) he commented extensively on the rights of a defendant. He began by asking the question: could the pope, on the basis of this decretal, proceed against a person if he had not cited him? Johannes concluded that the pope was only above positive law, not natural law. Since a summons had been established by natural law, the pope could not omit it. He argued that no judge, even the pope, could come to a just decision unless the defendant was present in court. When a crime is notorious, the judge may proceed in a summary fashion in some parts of the process, but the summons and judgment must be observed. He argued that a summons to court (citatio) and a judgment (sententia) were integral parts of the judicial process because Genesis 3.12 proved that both were necessary. God had been bound to summon Adam; human judges must do the same. Then he formulated an expression of a defendant's right to a trial and to due process with the following words: a person is presumed innocent until proven guilty (item quilbet presumitur innocens nisi probetur nocens).(17) This sentence is the ultimate irony of the story: rather than a sturdy Anglo-Saxon, a cardinal of the Roman church, a Frenchman, a canonist, Johannes Monachus was the first European jurist to recognize the inexorable logic of God's judgment of Adam: God could not condemn Adam without a trial because even God must presume that Adam was innocent until proven guilty. Johannes' commentary on Rem non novam eventually became the Ordinary Gloss of the Extravagantes communes and was circulated in hundreds of manuscripts and scores of printed editions until the seventeenth century.(18) Johannes Monachus gave birth to the maxim, "Innocent until Proven Guilty." Since it was read by the jurists of the Ius commune to the time of Cesare Beccaria, his Ordinary Gloss was a primary vehicle for broadcasting the principle to later generations of jurists.

Roman law, canon law, the Ius commune: from these sources spring that great Anglo-Saxon principle: A person is innocent until proven guilty. The question remains, however, how deeply did this doctrine inform the jurisprudence and court practice of the late medieval and early modern Europe? Today I can give only a brief outline of the problem and rough sketch of the story's main features up to the time with which we began, the time of Beccaria and MacNally.

A glance at the standard accounts of procedure and law after the thirteenth century would seem to render the opinion risible that any conception of "innocent until proven guilty" existed before the eighteenth century in European jurisprudence. Inquisitorial courts searching out heresy seem the antithesis of due process and contrary to any conception of defendants' rights. Torture, secret accusations, and arbitrary procedural injustices seem the norm rather than the exception. Some scholars have argued that the courts had an obligation to punish crimes, it was a matter of public utility, and that procedural short cuts to the "truth" like torture were means through which the courts fulfilled their obligations. A legal maxim that paralleled "innocent until proven guilty" and that stated that "It is in the public interest that crimes not remain unpunished" might seem to reflect the essence of late medieval and early modern juristic thought (Rei publicae interest, ne crimina remaneant impunita).

It has been true in the past and remains true today that procedural rules are broken and rights violated most often when judges have faced crimes that strike society's most sensitive nerves. The cases in which I have found that the presumption of innocence discussed again and again are those that dealt with marginal groups, especially heretics, witches, and Jews. In these sources we find evidence that presumption of innocence was not reinvented in the eighteenth century.

Let me give a couple of examples. In 1398 or 1399, Salamon and his son Moyses, Jews living in Rimini, had been accused by several Christian women of having had sexual relations with them. The case was heard by a Franciscan inquisitor, Johannes de Pogiali. The case fell under the jurisdiction of the inquisition because Salamon and Moyses had used heretical arguments to seduce the women. When the Jews encountered virtuous resistance from the Christian women, they told them that Christian women who fornicated with Jewish men did not sin. The women testified before the Inquisition that they capitulated to Salamon and Moyses only after having been convinced by their clever arguments.(19)

We do not know the facts behind this case, only its outcome as reported in the papal court. Although the bare facts might make us think of this case as material for a Boccaccian farce, Salamon and Moyses did not think the accusation was amusing. The inquisitor's summary of the case is of great interest. He called witnesses before him, examined them, and took their oaths to tell the truth. In the end he did not find that the accusations against Salamon and Moyses were juridically and legitimately proven. It is not often that we find a judge justifying his decision in the Middle Ages. In this case, Johannes de Pogiali did. He examined the facts and concluded that "it was better to leave a crime unpunished than to condemn an innocent person."(20) Johannes had to choose between two conceptions of order: that crimes should be punished in the public interest or that defendants should be presumed innocent if proofs were insufficient, even in a delicate case where an outsider had violated more than just the public order.(21) Johannes also had to choose between a standard of justice for Christians and a standard for Jews. When judges and jurists asked themselves that question in the fifteenth and sixteenth century, the theoretical answer was invariably the same: Jews had the same rights of due process as Christians. And if proofs failed, they were presumed innocent. To be sure, the theory did not always find its way into the courtroom, but the rules were repeated again and again in papal mandates sent to local judges and inquisitorial courts. In 1469 Pope Paul II confirmed the petition of the Emperor Frederick III that absolved Christian judges, notaries, and scribes who participated in cases involving Christians and Jews from any wrong doing. Some Christian priests refused to absolve them from their sins unless they did penance for their roles in court aiding Jews. "Justice," Paul observed, "ought to be common to all, Christian or Jew."(22) Later popes issued decretals that specified in great detail the procedural protections that Jews must be given. A letter of Pope Sixtus IV in 1482 mandated that Jews should receive the names of their accusers, should be able to present legitimate exceptions, proofs, and defenses to the court, and, if these rights were violated, could appeal to Rome.(23) From the number of times the Roman curia repeated these admonitions over the next fifty years, theory and practice may not have always happily coincided.(24) Several sixteenth-century letters emphasized a Jew's right to a defense, to have an advocate, and to receive money from supporters for a defense in heresy and apostasy trials. As Pope Paul III declared in 1535, "no one should be deprived of a defense, which is established by the law of nature."(25) The right to a defense, a lawyer, and the means to conduct a defense was a necessary extension of the rights enshrined by the maxim "Innocent until Proven Guilty." By way of contrast, the common law did not recognize the right of a criminal defendant to counsel in treason trials until 1696.(26)

The sixteenth century became a great age for criminal law and procedure in the Ius commune. Earlier jurists had written tracts on torture, evidence, heresy and witchcraft trials, but none had written a detailed tract on criminal procedure. From the thirteenth to fifteenth centuries, treatises on criminal procedure were, with only a few exceptions, short and schematic.(27) During the sixteenth century, the jurists synthesized the jurisprudence of the Ius commune, and they wrote great tracts on the rights of criminal defendants. The names of these proceduralists are not well known: Giuseppe Mascardi, Giovanni Luigi Riccio, Giulio Claro, and Giacomo Menochio are not household names, even to legal historians. One of the great figures in this development was Prospero Farinacci (1544-1618). He was educated in Perugia and quickly discovered both sides of the bench. In 1567 he became the general commissioner in the service of the Orsini of Bracciano; the next year he took up residence in Rome as a member of the papal camera. However, in 1570 he was imprisioned for an unknown crime. Legal problems hounded him for the rest of his life. He lost an eye in a fight, was stripped of his positions, and accused of sodomy. In spite of his difficulties, Pope Clement VII reinstated him to the papal court in 1596. He began his most important work, Praxis et theorica criminalis, in 1581 and put the finishing touches on it by 1601.(28)

Farinacci's treatise bristles with the presumption of innocence. The issue arose in several different contexts. He insisted that the exception of innocence was privileged in law and could never be abolished by statute; if a statute would abolish a defendant's right to a defense, it should be interpreted as only being unjust or calumnious defenses.(29) Even the pope could not take away the right of a defendant to prove his innocence, since that right was grounded in the law of nature.(30) Like other jurists who wrote on criminal procedure, Farinacci distinguished between presumptions of law and of men: a presumption of man was, for example, that in doubt, a man was presumed to be good.(31)

Presumption of innocence was alive and well in the Ius commune during the sixteenth and seventeenth centuries. And the maxim, "Innocent until Proven Guilty" was flourishing as well. A great voice of reason in criminal procedure was Frederick von Spee (1591-1635). Spee was a jurist, Jesuit, poet --- literary critics are still spilling ink on his most important poem, Trutznachtigall --- he was a critic of intolerance and ignorance. As Beccaria would a century later, he condemned torture, the persecution of witches, and other crimes that enraged princes and the rabble. Unfortunately for him, Europe was not yet ready for his voice of reason. He was stripped of his academic positions after the publication of Cautio criminalis, his famous treatise on procedure in witchcraft trials, condemned by his order, and died young, helping to treat soldiers infected by the plague of 1635 in Trier.(32)

"Must we assume that witches are guilty?" he asked in Cautio criminalis. "That's a stupid question," he bellowed.(33) His condemnation of torture was absolute. He took his arguments from Farinacci.(34) His rhetoric inspired Beccaria a century later:(35)

I ask a judge who tortures a defendant, to what end? Is the torture a punishment for a crime? Or a road to the truth? It is against all law that torture is punishment, and precisely unheard of, for, of which crime is it punishment?

A few pages earlier he had already made the point explicit that implicitly underlies this passage. Can a defendant who does not confess under torture be condemned? "I assume," wrote Spee, "that no one can be condemned unless his guilt is certain; an innocent person ought not be killed. Everyone is presumed innocent, who is not known to be guilty."(36)

There is some irony in this part of the story too. Beccaria and Pietro Verri, Beccaria's muse who wrote a significant tract on torture published long after Dei delitti,(37) probably borrowed Spee's thought and adapted his words when they wrote about torture.(38) Yet Beccaria and Verri condemned Spee, Farinacci and other jurists at the same time that they appropriated their ideas, accusing them of being soft on torture.(39) Alessandro Manzoni, the mother of all modern Italian novelists, whose father was Verri's brother and whose mother was Beccaria's daughter, laid bare Verri's dishonesty. Verri, he wrote, overemphasized his own contribution to the intellectual arguments condemning torture and de-emphasized the contribution of earlier jurists. Verri had seriously distorted the legal tradition.

Manzoni's defense of Spee and the earlier proceduralists was as elegant as any passage in his great novel, I promessi sposi:(40)

From this evidence and from all we know of the practice of torture in their own time, one can undoubtedly conclude that the interpreters of criminal procedure left the theory and practice of torture much, but much, less barbarous than they found it. Of course it would be absurd to attribute this diminution of evil to one cause alone, but I think that among the many causes that it would be reasonable to count one as particularly significant: the repeated reproofs and warnings of jurists, who renewed then publicly, century after century and who exercised a de facto authority over the practices of the courts.

Manzoni had extraordinary insight into the evolution of norms in European jurisprudence. He perceived the complicated dialectic through which jurists argued, borrowed, and added to the thought of their predecessors and spoke to their peers and successors. This dialectic recognized no national laws, was not impoverished by territorial and linguistic boundaries, and did not try to conform to a particular set of customs and practices.

We have come full circle: from Justice White to MacNally to Beccaria to Johannes Monachus and back to Beccaria. The evolution of the norm that every person is presumed innocent until proven guilty is a case study of the long process through which principles of law emerge, slowly, hesitantly, sometimes painfully, in a legal system. The maxim, Innocent until proven guilty was born in the late thirteenth century, preserved in the universal jurisprudence of the Ius commune, employed in the defense of marginalized defendants, Jews, heretics, and witches, in the early modern period, and finally deployed as a powerful argument against torture in the sixteenth, seventeenth and eighteenth centuries. By this last route it entered the jurisprudence of the common law through a thoroughly disreputable Irishman's having read a book on criminal punishments by an Italian. In a world that is choked by the narrow horizons of legal systems imprisoned by national sovereignties, this story is the best argument I know for returning to a conception of law that looks outside the national, territorial boundaries that Cerberus, the great hound of Hell and defender of the Austinian doctrine of legal positivism, so assiduously guards today.

1.  D. Clementi, "The Anglo-Saxon Origins of the Principle `Innocent until Proven Guilty'," Herrschaftsvertraege, Wahlkapitulationen, Fudamentalgesetze 68-76 at 69.

2. Coffin vs. U.S., 156 U.S. 432, 432-463 (1894).

3. Dudum (X 2.23.16), which was in fact used by the creator of the maxim in the thirteenth century to justify it; see below n. 18.

4. Joseph Mascardi (d. 1588), Josephi Mascardi ... conclusiones probationum omnium, quae in vtroque foro versantur continens, iudicibus, aduocatis, causidicis, omnibus denique iuris pontificij, Cesareique professoribus utiles, practicabiles, ac necessarias ...(3 Vols. Venice: 1609; 4 Vols. Frankfurt: 1661, 1703, 1727-1731)

5. Coffin vs. U.S., 156 U.S. 432, 455.

6. For a detailed discussion of its present meaning in American law, see William S. Laufer, "The Rhetoric of Innocence," Washington University Law Review 70 (1995) 329-421.

7. Leonard MacNally, The Rules of Evidence on Pleas of the Crown (London-Dublin: 1802) 33.

8. Leonard MacNally, The Rules of Evidence on Pleas of the Crown (London-Dublin: 1802) 19. Charles Louis de Secondat Montesquieu, De l'esprit des lois in Oeuvres complètes, ed. Daniel Oster (Paris: 1964) Book 12, chapter 3, p. 599.

9. Bernard Porter, Plots and Paranoia: A History of Political Espionage in Britain, 1790-1988 (London: 1989) 32-33 for a short account of MacNally's career. Porter misspells his name (McNally).

10. Cesare Beccaria, On crimes and punishments, translated by Richard Davies and edited by Richard Bellamy, with Virginia Cox and Richard Bellamy (Cambridge Texts in the History of Political Thought; Cambridge: 1995), chapter 13, 32. The Italian text reads: "Più d'un testimonio è necessario, perché fintanto che uno asserisce e l'altro nega niente v'è di certo e prevale il diritto che ciascuno ha d'essere creduto innocente," ed. Luigi Firpo (Milano: 1984) 56

11. F. Venturi, "Beccaria, Cesare," Dizionario biografico degli Italiani 7 (1965) 458-469, at 461: "Argomenti vecchi, derivanti in parte de Montesquieu, ed elementi nuovi, che sgorgavano dalla personalità stessa del Beccaria, confluivano in queste pagine, tra le più chiare e persuasive uscite della sua penna."

12. Cesare Beccaria, On crimes and punishments, translated by Richard Davies and edited by Richard Bellamy, with Virginia Cox and Richard Bellamy (Cambridge Texts in the History of Political Thought; Cambridge: 1995), chapter 16, 39 (I have corrected their translation). The Italian text reads: "o il delitto è certo o incerto; se certo, non gli conviene altra pena che la stabilita dalle leggi, ed inutili son i tormenti, perché inutile è la confessione del reo; se è incerto, e' non devesi tormentare un innocente, perché tale è secondo le leggi un uomo i di cui delitti non sono provati, " ed. Luigi Firpo (Milano: 1984) 62.

13. On Beccaria and eighteenth-century penal law, see Cesare Cantù, Beccaria e il diritto penale (Firenze: 1862) and Marcello Maestro, Cesare Beccaria and the Origins of Penal Reform (Philadelphia: Temple University Press, 1973) especially 3-19, 110-143. Also Francesco Corpaci, Ideologie e politica in Cesare Beccaria (Milano: 1965).

14. A short sketch of these developments can be found in K. Pennington, "Law, Procedure of, 1000-1500," The Dictionary of the Middle Ages 7 (New York: 1986) 502-506.

15. The idea was already found in Gratian's Decretum, C. 3 q.9 c.3: "Absente reo accusator non audiatur."

16. X 2.23.15 (Dudum).

17. Citing Innocent III's decretal Dudum (X 2.23.16) to justify his assertion.

18. Johannes Monachus to Extravag. com. 2.3.1 (Rem non novam) v. Non obstantibus aliquibus privilegiis, London, BL Royal 10.E.i., fol. 214r, London, Lambeth Palace 13, fol. 363v-364r: "Et Gen. xviii. ubi factum erat notorium attamen Deus uoluit probare quam iudicare . . . Nec obstat extra. de accus. c. Euidentia <X 5.1.9>, nec ibi tollitur citatio nec sententia quia Gen. iii. probatur utrumque necessarium . . . Hinc est quod iudiciorum ordo et placitandi usus in paradiso videtur exordium habuisse. Nam Adam de inobedientia a Domino redargutus, quasi actori exceptionem obiiciens, relationem criminis in coniugem, immo in coniugis actorem convertit, dicens: Mulier quam mihi sociam dedisti me decepit . . . Item quilibet presumitur innocens nisi probetur nocens, extra. de presum. c. Dudum <X 2.23.16>."

19. The Apostolic See and the Jew: Documents: 492-1404 (Studies and Texts, 94; Toronto: 1991) 527-529, ASV, Reg. Vat. 316, fol. 226r-226v and Bologna, Bibl. univers. Cod. Lat. 317, Vol. 5.1, fol. 277r-280v. Published by C. Piana, Cartularium Studii Bononiensis S. Francisci (Analecta Francescana 11; Assisi: 1970) 384ff. For another consilium treating a Jew's allegation that fornication was not a sin, see Bologna, Collegia di Spagna, MS 123, fol. 382r-416r: "Quidam Iudeus firmiter credit et publice aserit quod cohire solutum cum soluta non est peccatum mortale." This consilium was probably written between 1381 and 1387 by Guglielmo de Vallseca who was the chancellor to King Peter IV of Catalonia. See I codici del Collegio di Spagna di Bologna, ed. Domenico Maffei, Ennio Cortese, Antonio García y García et al. (Orbis Academicus: Saggi e documenti di storia delle università, 5; Milano: 1992) 400. On Muslim and Jewish miscegenation with Christian women, see David Nirenberg, Communities of Violence: Persecution of Minorities in the Middle Ages (Princeton: 1996) 127-165.

20. Ibid., p. 528: "non invenit contra ipsum Salomonem fore iuridice et legitime probatum, videlicet quod ipse talia verba protulerit aut alia supradicta commiserit vel fecerit, considerans fore melius facinus impunitum relinquere quam innocentem condempnare . . . declaravit dictum Salomonem de verborum predictorum prolatione et aliorum premissorum perpetratione fuisse et esse innocentem."

21. Richard M. Fraher has argued in a series of articles that medieval procedure between 1200 and 1500 was saturated with the idea that the Ius commune dictated that it was in the public interest that crimes not remain unpunished. This conception of judicial order led to the introduction of torture and deposited the burden of proof on defendants. I think that presumption of innocence played a greater role in theory and practice than Fraher would concede. See his articles "The Theoretical Justification for the New Criminal Law of the High Middle Ages: `Rei publicae interest ne crimina remaneant impunita'," University of Illinois Law Review (1984) 577-595 and 'Conviction according to Conscience: The Medieval Jurists' Debate concerning Judicial Discretion and the Law of Proof," Law and History Review 7 (1989) 23-88. Also see "Ut nullus describatur reus prius quam convincatur:  Presumption of Innocence in Medieval Canon Law,"  Proceedings of the Sixth International Congress of Medieval Canon Law, Berkeley, California (Monumenta iuris canonici, Series C, 7; Vatican City: 1985) 493-506.

22. Shlomo Simonsohn, The Apostolic See and the Jews: Documents: 1464-1521(Studies and Texts 99; Toronto: 1990) 1163-1164 at 1164: "cum iustitia, que omnibus communis esse debet, suos Iudeis ministratur."

23. Ibid. 1284-1287.

24. Ibid. pp. 1287-1288; The Apostolic See and the Jews: Documents: 1522-1538 (Studies and Texts 104; Toronto: 1990) 1693-1700 at 1696; The Apostolic See and the Jews: Documents 1539-1545 (Studies and Texts 105; Toronto: 1990) 2297-2298.

25. Ibid. 1991: "nos, volentes nemini defensionis munus, quod de iure nature est, tolli . . ." Also pp. 2078.

26. Theodore F.T. Plucknett, A Concise History of the Common Law (4th Ed. London: 1948) 410. Cf. James A. Brundage, Medieval Canon Law (London-New York: 1995) 149 n. 50. French jurisprudence continued to follow the dictates of the Ius commune: Procez verbal des conférences tenues par ordre du roi, entre messieurs les commissaires du conseil et messieruts les députés du parlement de Paris, reprinted in Code Louis: Ordonnance criminelle, 1670 (Testi documenti per la storia del processo, edd. Nicola Picardi and Alessandro Giuliani (Milano: 1996) 163: " ce conseil qu'on a acoûtumé de donner aux acusez n'est point un privilége acordé par les Ordonnances ni par les Loix que c'étoit une liberté aquise par le droit naturel, qui est plus ancien que toutes les Loix humaines. Que le nature enseignoit à l'homme d'avoir recours aux lumieres des autres, quand il ne n'en avoit pas assez pour se conduire et d'emprunter du secours, quand il ne se sentoit pas assez fort pour se défendre. . . . Que si on vouloit comparer nôtre procédure criminelle à celle des Romains et des autres Nations, ou trouveroit qu'il n'y en voit point de si rigoureuse, que celle qu'on observe en France, particuliérement dépuis l'Ordonnance de 1539."

27. Tancred of Bologna wrote one the first tracts dealing solely with criminal procedure ca. 1210-1215, see Richard M. Fraher, "Tancred's `Summula de criminibus': A New Text and a Key to the Ordo Iudiciarius," Bulletin of Medieval Canon Law 9 (1979) 23-35, edited on pp. 29-35. The most complete discussion of criminal procedure during this period was the third part of Guilielmus Durantis' Speculum iuris (2 vols. Basel: 1574; reprinted Aalen: 1975). Other influential tracts on criminal procedure can be found in volume 11 of Tractatus universi iuris (Venice: 1584).

28. A. Mazzacane, "Farinacci, Prospero (1544-1618)," Juristen: Ein biographisches Lexikon von der Antike bis zum 20. Jahrhundert, ed. Michael Stolleis (München: 1995) 199-200; N. Del Re, "Prospero Farinacci giureconsulto romano (1544-1618), Archivio della Società Romana di Storia Patria 00 (1975) 000. Mazzacane writes that he completed it in 1614, but an edition of Praxis et theoricae criminalis was published in Venice: apud Georgium Variscum, 1603 (in fine 1601), which is described as the third edition, with additions made by the author to the first and second editions, see Antichi testi giuridici (secoli XV-XVIII) dell'Istituto di Storia del Diritto Italiano, ed. Giuliana Sapori (Università degli Studi di Milano, Pubblicazioni dell'Istituto di Storia del Diritto Italiano, 7; Milano: 1977) 1.242, no. 1162.

29. Prospero Farinacci, (Prosperus Farinacius), Praxis et theoricae criminalis libri duo in quinque titulos distributi (Francofurto: 1606) Liber I, tit. I, quaestio XI, p. 141, num 9: "Quarto urget quia exceptio innocentiae a iure priuilegata est, et taliter, ut numquam a statuto alias exceptiones tollente sublata censeatur, l. Quoniam, C. ad leg. Iul. de adult. Anan. in c. Si adversus . . . immo si statutum tollit reo defensiones, intelligitur de defensionibus iniustis et calumniosis, secus de iustis, et quae respiciunt illius innocentiam."

30. Ibid., p. 143-144, num. 15: "Et si tu subtilis diceres ergo Princeps isto casu <cases where the pope has judged someone contumacious> tollit exceptionem, et defensionem innocentiae? Qua tamen eum sit de iure naturae, nec a principe, nec a statuto tolli potest . . . Respondeo duobus modis. Primo, quod ex caussa (sic) publicae utilitatis multa potest princeps contra generales iuris regulas, praesertim ne delicta remaneant impunita, ita in his terminis respondet Carer. d. num. 89 et 99 poret Matth. et Andr. in constitu. Paenam eorum, in i. notab. . . . Respondeo secundo, quod isto casu summus pontifex non prohibuit exceptionem innocentiae, sed illius admissionem sibi reservavit, cum dicit "Nisi habita desuper asignatura nostra speciali gratia" (Cited bull of Pius V, dated to the fifth year of his pontificate on p. 143). Advertat ergo iudex isto casu, ne sit velox ad exsequendum sententiam quia si reus offert innocentiam suam docere per contrarias probationes debet supersedere et principem consulere vel exspectare, quod pro parte rei adeundum principem recurratur, isto praesertim casu, in quod prout infra dicetur, facilis esse debet idem princeps in admittendo reum ad defensiones, et si iudex aliter faceret, male faceret."

31. Ibid. Lib. I, tit. V, p. 563 num. 94: "Praesumtio hominis . . . est quidam conceptus caussatus in mente ab aliqua probabili coniectura . . . quod in dubio quilibet praesumitur bonus et non malus, c. Dudum, de praesum. et similia [citing Giuseppe Mascardi, Conclusiones probationum, Lib. I, q.10, num. 24 and 25].

32. Wilhelm Kosch, Deutsches Literaturlexikon, ed. Bruno Berger (Bern-München: 1963) 408-409; Karl Keller, Friedrich Spee von Langenfeld (1591-1635): Seelsorger, Dichter, Humanist (S.l.: 1969); G. Richard Dimler, Friedrich Spee's Trutznachtigall (German Studies in America, 13; Bern-Frankfurt am Main: 1973) 11-13.

33. Friedrich von Spee (published anonymously), Cautio criminalis seu De processibus contra sagas liber (Rinthelii(Rinteln an der Weser): 1631) Dubium XIX, p. 101: "An de captis nomine Veneficii mox praesumendum sit eas necessario reas esse? Stulta quaestio videatur." A modern edition edited by Theo G.M. van Oorschot (Friedrich Spee, Sämtlcihe Schriften, 3; Tübingen-Basel: Francke Verlag, 1992), including the pagination of the 1631 edition.

34. Ibid. p.264, Dubium XXXIX, Ratio i. "Si convicta esset, torta non esset, est autem torta, non igitur convicta: Constat ex supra dictis et vide Farinacium q.38 n. 4."

35. Ibid., pp. 264-265: "Dubium XXXIX, Ratio ii. Quaero ex judice quem in finem ream torserit? An, ut tortura paena delicti esset? An, ut, via ad veritatem? Contra jura est, ut tortura paena sit, et prorsus inauditum, nam cujus criminis erit paena? . . . Ratio iii. Quero item ex judice an haec reae confessio necessaria ad condemnandum fuit, an non fuit? Si fuit, cur igitur sine ea nihilominus damnavit? Si non fuit, crudelitas est, quam morte puniendam seu confessam seu inconfessam destinaverat, tam gravibus torturis lacerasse, ne, quae unit tantum destinata morte erat, una morte morteretur."

36. Ibid., p. 262-263: "Dubium XXXIX. An, quae nihil in torturis confessa est, damnari possit? Suppono damnari neminem posse nisi certo de eo constet culpam sustinere; neque enim innocens occidi debet. Innocens autem omnis praesumitur, qui nocens esse nescitur."

37. Pietro Verri (1728-1797), Osservazioni sulla tortura (Milano: 1804). Most recent edition: PietroVerri, Osservazioni sulla tortura (Volti e anime ; 3 Milano : C. Gallone, 1997).

38. See the comparison of texts printed in the edition of Beccaria, ed. Luigi Firpo (Milano: 1984), p. 62 n. 1 and p. 63 n. 1.

39. Pietro Verri, Osservazioni sulla tortura: E singolarmente sugli effetti che produsse all'occasione delle unzioni malefiche alle quali si atribui la pestilenza che devasto Milano l'anno 1630 (Milano: 1804).

40. Alessandro Manzoni, Storia della colonna infame: Testo del 1840, ed. Alberto Chiari and Fausto Ghisalberti (Verona: 1963) 702: "Da queste testimonianze, e da quello che sappiamo essere stata la tortura negli ultimi suoi tempi, si può francamente dedurre che i criminalisti interpreti la lasciarono molto, ma molto, men barbara di quello che l'avevan trovata. E certo sarebbe assurdo l'attribuire a una sola causa una tal diminuzione di male; ma, tra le molte, me par che sarebbe anche cosa poco ragionevole il non contare il biasimo e le ammonizioni ripetute e rinnovare pubblicamente, di secolo in secolo, da quelli ai quali pure s'attribuisce un'autorità di fatto sulla practica de' tribunali." Manzoni's father was probably Giovanni Verri, the brother of Pietro, and his mother was Giulia Beccaria, the daughter of Cesare, see Alessandro Manzoni, Storia della colonna infame: Testo definitivo e prima redazione, ed. Renzo Negri (Milano: Marzorati, 1974) 5.