You are on page 1of 21

Torture and Plea Bargaining Author(s): John H. Langbein Source: The University of Chicago Law Review, Vol. 46, No.

1 (Autumn, 1978), pp. 3-22 Published by: The University of Chicago Law Review Stable URL: Accessed: 20/10/2008 11:29
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about JSTOR, please contact

The University of Chicago Law Review is collaborating with JSTOR to digitize, preserve and extend access to The University of Chicago Law Review.

Tortureand Plea Bargaining*
John H. Langbeint In this essay I shall address the modem American system of plea bargaining from a perspective that must appear bizarre, although I hope to persuade you that it is illuminating. I am going to contrast plea bargainingwith the medieval Europeanlaw of torture. My thesis is that there are remarkableparallels in origin, in function, and even in specific points of doctrine, between the law of torture and the law of plea bargaining. I shall suggest that these parallels expose some important truths about how criminal justice systems respondwhen their trial proceduresfall into deep disorder.

For about half a millennium, from the middle of the thirteenth century to the middle of the eighteenth, a system of judicial torture lay at the heart of Continental criminal procedure.In our own day the very word "torture" is, gladly enough, a debased term. It has come to mean anything unpleasant, and we hear people speak of a tortured interpretation of a poem, or the torture of a dull dinner party. In discussions of contemporarycriminal procedurewe hear the word applied to describe illegal police practices or crowded prison conditions. But torture as the medieval European lawyers understood it had nothing to do with official misconduct or with criminal sanctions. Rather, the application of torturewas a routine and judicially supervised feature of European criminal procedure. Under certain circumstances the law permitted the criminal courts to employ physical coercion against suspected criminals in orderto induce them to confess. The law went to great lengths to limit this technique of extorting confessions to cases in which it was thought that the accused was highly likely to be guilty, and to surroundthe use of torture with other safeguards that I shall discuss shortly. This astonishing body of law grew up on the Continent as an adjunct to the law of proof-what we would call the system of
* The William Crosskey Lecture in Legal History, The University of Chicago Law School, October 19, 1978. Suggestions and criticisms from Albert Alschuler, Paul Bator, RichardEpstein, Thomas Green,Emile Karafiol,RichardLempert,Norval Morris,Richard Posner, GeoffreyStone and Peter Westen are gratefullyacknowledged. t Professorof Law, The Universityof Chicago. ' This section of this article is based upon and reproduces some languagefromJ. LANGIN BEIN,TORTUREANDTHELAW OFPROOF:EUROPEANDENGLAND THEANCIENRIGIME 1-16 (1977).


LANGBEIN.the Italian Glossatorswho designedthe system developedand entrenchedthe rule that conviction had to be based upon the testimony of two unimpeachable eyewitnesses to the gravamenof the crime-evidence that was. for example. because circumstantialevidence depends for its efficacy upon the subjective persuasionof the trier who decides whetherto drawthe inferenceof guilt from the evidence of circumstance.4 The Universityof ChicagoLaw Review [46:3 trial-in cases of serious crime (for which the sanction was either death or severe physical maiming2). 151-64. Thus. They had set the level of safeguardtoo high.3The ordeals purportedto achieve absolute certainty in criminaladjudicationthroughthe happyexpedientof having the judgmentsrenderedby God. Accordingly. at 27-44. at 5-7. it would not have mattered in this system that the suspect was seen running away from the murdered man's house and that the bloody dagger and the stolen loot were found in his possession. 35 in supra note 1. in the famous phrase. Since no eyewitness saw him actually plunge the weapon into the victim. who could not err. which the Roman Churcheffectively destroyed in the year 1215. . Another way to appreciatethe purposeof these rules is to understand their corollary:conviction could not be based upon circumstantial evidence. In the historyof Westerncultureno legal system has ever made a morevaliant effortto perfectits safeguardsand therebyto exclude completely the possibility of mistaken conviction." Without these two eyewitnesses.substantiallyreproduced J.The HistoricalOriginsof the Sanctionof Imprisonmentfor SeriousCrime.The medieval law of proofwas designed in the thirteenth century to replace an earlier system of proof. They had constructeda system of proof 2 The use of imprisonmentas a sanction for serious crime was a developmentof the Renaissanceand later times.The replacement system of the thirteenth century aspired to achieve the same level of safeguard-absolute certainty-for human adjudication. LANGBEIN. a criminal court could not convict an accused who contested the chargesagainst him. the ordeals.5 J. But the Europeans learned in due course the inevitable lesson. Only if the accused voluntarily confessed the offense could the court convict him without the eyewitness testimony. the court could not convict him of the crime. "clear as the noondaysun. Langbein. LEGAL STUD. 3 J. supranote 1. Although human judges were to replace God in the judgment seat. they would be governedby a law of proofso objective that it would make that dramatic substitution unobjectionable-a law of proof that would eliminate human discretionfrom the determination of guilt or innocence. (1976).

something had to be done to extend the system to those cases. Vast legal treatises were compiled on this jurisprudence of tortureto guide the examining magistrate in determiningwhether there was probable cause for torture. The two-eyewitness rule was hard to compromise or evade. ' These works are canvassed in 1 & 2 P. That meant either one eyewitness. FIORELLI.W. accordingto a fairly elaboratetariff. each of those indicia would be a quarterproof. designed to assure that only persons highly likely to be guilty would be examined under torture. 1898). but a nominally subsidiaryplace. Because society cannot long tolerate a legal system that lacks the capacity to convict unrepentant persons who commit clandestine crimes. HISTORY ENGLISH THE 660 (2d ed. torture was permitted only when a so-called "half proof' had been established against the suspect. which was sufficient to permit the authorities to dispatch the suspect for a session in the local torture chamber. The law of torture grew up to regulate this process of generating confessions. but on a question of interlocutoryprocedure-whether or not to examine the accused under torture. ostensibly objective criteria for evaluating the indicia and assigning them numerical values (quarter proofs. The largest chapter of the European law of torture concernedthe prerequisitesfor examination under torture.6 OF & LAW ' Maitland's famous term. or circumstantial evidence of sufficient gravity. and the like).1978] Tortureand Plea Bargaining 5 that could as a practical matter be effective only in cases involving overt crime or repentant criminals. Circumstantial evidence was not consulted directly on the ultimate question. TORTURA LA NEL GIUDIZUIAA DIRITO COMUNE (1953-54). To go from accepting a voluntary confession to coercing a confession from someone against whom there was already strong suspicion was a step that began increasingly to be taken. MAITLAND.5In the example where a suspect was caught with the dagger and the loot. at 14.Together they cumulated to a half proof. Thus. . POLLOCK F. LANGBEIN. Even there the law attempted to limit judicial discretion by promulgating predetermined. half proofs. The spirit of safeguardthat had inspiredthe unworkableformal law of proof also permeated the subterfuge. but the confession rule seemed to invite the "subterfuge"'that in fact resulted. The law of torture found a place for circumstantial evidence. The European jurists devised what AngloAmerican lawyers would today call a rule of probable cause. guilt or innocence. * J. 2 F. In this way the prohibition against using circumstantial evidence was overcome. supra note 1.

and courtfunctionaries. 1562). clerk. De corfiue repetitionequaefionis turx.) . fromJoost Damhouder'sPraxis RerumCriminalium91 (Antwerp ed.The illustration. which discusses the rules for repeating the infliction of torture on an accused who has previously resisted confession despite examination under torture.6 The University of Chicago Law Review [46:3 CAPVT XXXVIII. appears in chapter 38. .v" j I 3 5Q 3St: 1g! I g MS I ii This woodcut from a leading sixteenth-century European criminal proceduremanual shows the accused being examined under torture in the presence of the court. (Reproduced with permission from the rare book collection of The University of ChicagoLaw Library.

"8 was forbidden to engage in so-called suggestive questioning. If he said he buried it under the old oak tree. LANGBEIN. J. "no innocent person can know. unsubstantiated confession of guilt. DAMHOUDER. ch.supra note 1. translatedin J. An accused who resisted any confessionunder torturewas supposedto be set free (subjectto rules permittingfurthertortureif new evidence was thereafter discovered). hence ineffective. PROSECUTING 8 ConstitutioCriminalisCarolinaart. the innocent might (as one sixteenth-century commentator put it) yield to "the pain and torment and confess things that they never did. CRIMINELLES 39. at 15-16. at 44 (Antwerp JUDICIAIRE CAUSES ES PRACTIQUE ed."9If the examining magistrate engaged in suggestive questioning.GERMANY.1978] Torture and Plea Bargaining 7 In orderto achieve a verbal or technical reconciliationwith the requirementof the formal law of proofthat the confessionbe voluntary.FRANCE282 (1974). unless the accused repeated it free from tortureat a hearingthat was held a day or so later. 1564) (first edition published as PRAXIS CRIMINALIUM RERUM (Louvain 1554)). and then found himself tortured anew.even accidentally. These were rules designed to enhance the reliability of the resulting was said that the accused had "purged"the incriminating evidence when he enduredthe torturewithout confession. more substantial safeguardswere devised to govern the actual application of torture. Because torture tests the capacity of the accused to endure pain rather than his veracity. torture was supposed to be employed in such a way that the accused would disclose the factual detail of the crime-information which. what he did know might be enough to give his confession verisimilitude. Rather. learned quickly enough that only a "voluntary"confession at the ratification hearing would save him from further agony in the torture chamber.Id. Moreover. the magistrate was supposed to send someone to dig it up. Torture was not supposed to be used to elicit an abject. he was supposed to be asked where he put the dagger. at 16. these safeguards never proved adequate to overcome the basic flaw in the system. If the accused confessed to the slaying. Often enough the accused who had confessed under torture did recant when asked to confirm his confession. the medieval lawyers treated a confessionextracted under torture as involuntary. in which the examiner supplied the accused with the detail he wanted to hear fromhim. in the words of a celebrated German The examining magistrate statute.the informationadmitted undertorture was supposed to be investigated and verified to the extent feasible.7 Fortunately. . But seldom to avail: the examination under torture could thereupon be repeated. ' J. If the accused knew something about the crime. recanted. but was still innocent of it. CRIMEIN THE RENAISSANCE: ENGLAND. his lapse could not always be detected or prevented. 54 (1532). An accused who confessed under torture. Alas. LANGBEIN.

and the least that I hope to achieve in this essay is to unburdenmyself somewhatby sharingthe disturbingvision that I think would come to any Americanwho had spent time studying the Europeanlaw of torture. I have lived with them for some years now. Having entrenchedthis unattainable level of safeguardin their formal trial procedure. In the eighteenth century. and the court is sparedhaving to adjudicateit. THELAW PLEA OF BARGANING I am now going to cross the centuriesand cross the Atlantic in order to speak of the rise of plea bargainingin twentieth-century America. but they were latecomersto a critical legal literature nearly as old as the law of torture itself.8 The University of Chicago Law Review [46:3 In some jurisdictions the requirementof verification was not enforced. Beccariaand Voltairebecame famous as critics of judicial torture. the prosecutoris relieved of the need to prove the accused'sguilt. in the form of a recommendationfor reduced sentence that the judge will follow.The two-eyewitnessrule had left Europeancriminal procedurewithout a tolerable alternative. . II. in the form of a chargereduction. The descriptionof the Europeanlaw of torturethat I have just presented has been meant to stir among Americanreadersan unpleasant sensation of the familiar. Plea bargainingoccurs when the prosecutorinduces a criminalaccused to confess guilt and to waive his right to trial in exchange for a more lenient criminal sanction than would be imposed if the accused were adjudicatedguilty following trial. The court condemns the accused on the basis of his confession.without independent adjudication. Judicial torture survived the centuries not because its defects had been concealed.the Europeansfound themselves obliged to evade it througha subterfugethat they knew was defective.but in spite of their having been long revealed. The prosecutoroffers leniency either directly. In exchange for procuringthis leniency for the accused. These shortcomingsin the law of torturewere identified even in the Middle Ages and were the subject of emphatic complaint in Renaissanceand early moder times. The parallels between the modern American plea bargaining system and the ancient system of judicial torture are many and chilling.throughthe connivanceof the judge. By way of preface.or indirectly. or was enforcedindifferently.let me set forthbrieflysome of the rudiments of our plea bargaining system. as the law of torture was finally about to be abolished along with the system of proofthat had requiredit. The coercedconfessionhad to replace proofof guilt.

. and the television transmits a steady flow of dramas in which a courtroomcontest for the verdict of the jury leads inexorably to the disclosure of the true culprit.S. . you will find-in no less hallowed a place than the Bill of Rights-an opposite guarantee. Trial Judge's Role in Plea Bargaining(] Torture and Plea Bargaining 9 Plea bargaining is. 1063 n. I). Understandingthe Short History of Plea Bargaining. 76 COLUM. (forthcoming). See VERAINSTITUnT THEIR ARRESTS: AND 7 FELONY PROSECUTION DISPOSITION NEWYORK IN CrrY'sCOURTs JUSTICE. CONST. OMY LAWENFORCEMENT 4. . id.PLEA PLEAS 13 (1975). ProfessorAlschulerreports95 percent for Houston in 1975. 545 n.3 (1978). chs. . see Baldwin & McConville. 28 (1979) (forthcoming)." This nontrial procedure has become the ordinary dispositive procedureof American law. Alschuler. Friedman. In truth. . If you turn to the American Constitution in search of authority for plea bargaining. ANAT- OF New York in 1971. by an impartial jury . ZEISEL. jury trial continues to occupy its central place both in the formal law and in the mythologyof the law. 13 LAW Soc'YREV.In a largestatistical study of New YorkCity felony data about to be published. therefore. The constitutions have not changed. .Hans THE Zeisel computes that 98 percentof convictionswere obtained by plea. VI (emphasis added). criminal jury trial has largely disappeared in America. rendering it absolutely unworkable as an ordinary dispositive procedureand requiringthe development of an altema" U. BOND. a nontrial procedurefor convicting and condemning people accused of serious crime."0 In our day. (forthcoming). . Why? Why has our formal system of proofand trial been set out of force? What has happened in the interval of less than two centuries between the constitutionalization of jury trial in 1791 and the present day to substitute this nontrial system for the trial procedure envisaged by the Framers? Scholars are only beginning to investigate the history of plea bargaining. H. 12.The L.Sentencing ProblemsRaised by GuiltyPleas: An Analysis of NegotiatedPleas in the BirminghamCrown L. 544. the courts pretend to enforce the defendant's right to jury trial. the accused shall enjoy the right to . The Sixth Amendment provides: "In all criminal prosecutions. Instead.20 (1976). a vast transformationovercame the Anglo-American13 institution of criminal jury trial. & (1979)(forthcoming). 97 percent of felony convictions were by plea. REV. The criminal justice system now disposes of virtually all cases of serious crime through plea bargaining. as many as 99 percent of all felony convictions are by plea.41 MOD. AND BARGAINING GUILTY "The figure of 99 percent is given for Detroit in J. . OF 2 See Alschuler. a guarantee of trial. trial .Plea Bargaining and its History. Court. amend. Depending on the jurisdiction.'2 but enough is known to permit us to speak with some confidence about the broad outline. In the two centuries from the mid-eighteenth to the mid-twentieth.Langbein. In (1977). REV.1059. you will look in vain. '1 On the extent of plea bargainingin England. Plea Bargaining in Historical Perspective.

14In the Old Bailey in the 1730swe know that the court routinelyprocessedbetween twelve and twenty jury trials for felony in a single day.REP. Hardy. The Criminal Trial before the Lawyers. DE ANGLORUM (London 1583). The Law and the ENGLAND 307 (A. 263 (1978). 45 U. Cm. 6 & 7 Will. 199 (1794)). OFFICE "7[1976] ADMINISTRATIVE OFTHE UNITED ANN. 335."'16By contrast. and conditional nonprosecution."9 Appellate reviewwas very re4 See generally Langbein. Tr.'5 In the 1790s. and the courtseriouslyconsideredwhetherit had any powerto adjourn. The trial judge called the witnesses (whomthe local justice of the peace had bound over to appear). Lawyerswere not employed in the conduct of ordinary criminal trials. it was still rapid and efficient. Lawyers. Only five percent of the trials lasted more than five days. The word "altercation" is the famous term of Sir Thomas Smith. L. REPORT 's The figure for Los Angeles appears in SANFRANCISCO ON THE CRIMINAL COURTSOF SAN FRANCISCO. at 272-84. we may note that the trial of Patricia Hearst for bank robberyin 1976 lasted forty days'7and that the averagefelonyjury trial in Los Angelesin 1968required7.2 days of trial time.10 The University of Chicago Law Review [46:3 tive procedure. see the valuable compilation in the appendix to Note. 1976-1977. bench trial. namely adversaryprocedureand the exclusionaryrules of the law of criminal evidence. CONST. . An Historical Argument for . 114 (1836).REP. STATES COURTS ON A COMMITTEE CRIME. and 11 percent from three to five DIRECTOR THE OF ANN. and the proceedingtranspiredas a relatively unstructured "altercation" between the witnesses and the accused. A single jury would be impaneled and would hear evidence in numerous unrelated cases before retiringto formulate verdicts in all. REV. OF COURTS. whereas 49 percent lasted from one to three days.) " U. in 2 JOHNSON'S berville ed. PART I: THE SUPERIOR COURTBACK1 AND REMEDIES (1970). Alschuler LOG-CONSEQUENCES for the reference. In eighteenth-centuryEngland jury trial was still a summary proceeding. 80 SMITH. 1933) (referring to Rex v. T. .S. ADMINISTRATIVE at F-2 (1978). when the Americans were constitutionalizing English jury trial. were still primitive and uncharacteristic.S.which we now recognizeto be the plea bargaining system. 15 See id. although they hardly detract from the contrast with eighteenth-century trial duration figures. either for the prosecutionor the defense. Tur287. (I owe this reference to Professor Jerome Israel.amend. For the American colonies. this figure must reflect the diversion of most easy cases into nonjury-trial channels such as plea bargaining. days.'8In the eighteenth centurythe most characteristic (and time-consuming)features of modernjury trial. 24 St. (I am grateful to Professor Albert W. REPUBLICA " MacKinnon. IV.The accused's right to representationby retained counsel was not generalizedto all felonies until the end of the eighteenth century in Americaand the nineteenth century in England. ch.) Of course. "The trial of Hardy for high treason in 1794was the first that ever lasted more than one day. VI (1791). STATE NEWJERSEY. 35 percent less than a day. Reliable figures for New Jersey criminal trials conducted in 1976-77 (bench and jury) are less spectacular.

445-46 (1974). REV.439. the Right to Counsel During Police Interrogation. In the middle of the 19th century.What we will not understand until there has been research directed to the question is why the pressurefor greatersafeguardled in the Anglo-American procedureto the law of evidence and the lawyerizationof the trial. but they led to reforms in nonadversarial procedure that preservedthe institution of trial. 7 GERICHTSSAAL1. In particular.g. 1055-57 (1964). it will be necessary to investigate the influence of the rise of professional policing and prosecution and the accompanying changes in the levels of crime reporting and detection. 21 In isolating the transformation of jury trial as the root cause of plea bargaining. but not Germany. there will certainly be other chapters.21 Similar for safeguardwere being felt on the Continent in the same pressures period. The practices that so protractmodernAmericanjury trial-extended voir dire. German scholars routinely studied English procedure as a reform model. 265. Ueber die processualische pt. L. .1978] Torture and Plea Bargaining 11 stricted into the twentieth century. Controlling Prosecutorial Discretion in Germany. exclusionary rules and other evidentiary barriers. and the intellectual influence of the marketplace model in an age when laissez faire was not an epithet. France. Anyone looking beyond the uniquely Anglo-American procedural development that we have emphasized needs to explain why plea bargaining has characterized lands with such disparate social composition as the United States and England. DES Wirkung des Gestandnisses im Schwurgerichtsverfahren. e.J. Das Schwurgericht: Gestiindniss und Verdikt und Kollision zwischen 530 ARCHIV (1870).. It desperately needed reform. 18 GOLTDAMMERS 22 This paragraph (with note 21) is derived from Langbein. 73 YALE 0 The latter portion of this paragraph is derived from Langbein. They found much to admire and to borrow (including the principle of lay participation in adjudication and the requirement that trials be conducted orally and in public).22 L. but they were unanimous in rejecting the guilty plea. reformsthat ultimately destroyed the system in the sense that they made jury trial so complicated and time-consuming that they rendered it unworkableas the routine dispositive procedure.counsel for indigent accused was not requireduntil the middle of this century. Gestandniss statt des Verdicts. supra note 12. beiden. changes in the social composition of victim and offender groups. 275 (1855). However. 41 U. von Arnold. The level of safeguard against mistaken conviction was in several respects below what civilized peoples now require. Walther. See. It was wrong for a court to sentence on "mere confession" without satisfying itself of the guilt of the accused. motions designed to provoke and preserve issues for appeal.20 Nobody should be surprisedthat jury trial has undergonegreat changes over the last two centuries. when German criminal procedure was being given its modem shape. Cm. maneuvers and speeches of counsel-all are late growths in the long history of common law criminal procedure. these other phenomena were largely experienced in Continental countries that did not turn to plea bargaining. changes in the rates and types of crime. we do not mean to imply that this procedural development is the sole cause of a practice so complex. 1851 ARCHIV CRIMINALRECHTS (Neue Folge) 225. 1000. When the history of plea bargaining is ultimately written. or the other major European states.

we make it terribly costly for an accused to claim his right to the constitutional safeguard of trial. The starting point. Once probable cause had been determined. This sentencing differentialis what makes plea bargainbetween having your ing coercive.There is. The law of torture served legal systems whose only sanctions for . Each system found itself unable to recant directly on the unrealisticlevel of safeguard to which it had committed itself. But like the Europeans of distant centuries who did employ those machines. safeguard was more important). In twentieth-centuryAmerica we have duplicated the central experience of medieval European criminal procedure: we have moved from an adjudicatoryto a concessionarysystem. We coerce the accused against whom we find probable cause to confess his guilt.The medieval Europeans were trying to eliminate the discretionof the professionaljudge by requiringhim to adhere to objective criteria of proof.12 The University of Chicago Law Review III. THEPARALLELS [46:3 Let me now turn to my main theme-the parallelsin function and doctrine between the medieval European system of judicial torture and our plea bargainingsystem.or twowordverdict that they do not explain or justify.the accused was made to concede his guilt rather than his accusersto prove it. To be sure. no thumbscrew. the law of tortureworkedan absolutelyfundamentalchangewithin the system of proof:it largelyeliminated the adjudicativefunction. But in the more difficult cases (where. The European law of torture preservedthe medieval law of proofundisturbedfor those easy cases in which there weretwo eyewitnesses or voluntary confession. and each then concentratedon inducing the accused to tender a confession that would waive his right to the safeguards. We threaten him with a materially increased sanction if he avails himself of his right and is thereafter convicted. a difference23 n This difference is related to differences in the sanctions that characterize the medieval and the modem worlds. Both the medieval law Europeanlaw of proofand the modem Anglo-American of jury trial set out to safeguardthe accused by circumscribing discrethe tion of the trier in criminal adjudication. of course. is that each of these substitute proceduralsystems arose in response to the breakdown of the formal system of trial that it subverted. I might add. in which laymen ignorantof the law return a one. The AngloAmericantrial system has been caught up over the last two centuries in an effortto protectthe accusedagainstthe dangersof the jury system. our means are much politer. we use no rack. which will be obvious from what I have thus far Spanish boot to mash his legs.

The Prosecutor'sRole in Plea Bargaining. CRIMINAL JUSTICE (1970). The Constitutiondoes not grant citizens any immunity from criminal prosecution."Alschuler. The prosecutor. typically some formof imprisonment. but not contraryto his rights. Convictionon this chargewould not have led to a greater sentence than thirty days' imprisonment. as in AttorneyDavis's example.who seems to have shared the defense attorney'sopinion on this point. Coercingpeople to stand trial is different from coercingthem to waive trial and to bring upon themselves sanctions that should only be imposed after impartialadjudication. CH.g. or sufferingsome extra years of imprisonment if you refuse to confess. at 183. 60 (1968). is coercive. at 60. a ratheroppositeargumentis made in behalf of plea bargaining-not that everythingis coercive. torture. at 61. susceptible to empirical testing."Id. but that a mere sentencing differential is not seriousenoughto be reckonedas coercion. therefore. I think that the answerto this argumentis straightforward. but the difference is of Like the degree.but it does grant them the safeguardof trial.althoughnot in the same degree. I do not think that great numbers of American defendants plead guilty to offenses committedby strangers. BLUMBERG. so is virtually every exerciseof criminaljurisdiction. he emphasized that conviction at trial seemed highly improbable.One can test this point simply by imagining a differential so great (e. The objectionis sometimesvoiced that if an accusedis innocent. Coercionauthorizedby law is differentfromcoercionmeant to overcomethe guarantees of law. See J. Davis recently representeda man chargedwith kidnappingand forciblerape.since few criminaldefendantsare genuinevolunteers. death versus a fifty-cent fine) that any reasonabledefendant would waive even the strongest defenses.. When Davis informedhis client of this offer. note 8. it stands to reasonthat ." Id. Like exchange for his confession.36 U. See A. The torturevictim was coercedinto a confession that condemned him to the most severe of punishments.or when the offensehas a complicatedconceptualbasis.The defendant'sreply was simple: "I can't take the chance. the greater the severity of the sanction that the accused's confession will bring down upon himself.and there was every likelihoodthat the defendant would be granted probation.(The law of torturewas also not supposedto apply in circumstances wherethe accused could explain away the evidence that might otherwisehave given cause to examine him under torture. See also text 89-92 at note 29 infra (discussingNorth Carolinav. Alford). as in tax and otherwhite collar crimes.Davis says. the greaterthe coercionthat must be broughtto bear upon him to wringout the confession. L. offeredto permit a guilty plea to simple battery. at 59-62 for 50.24 serious crime were severe physical maiming and death. evidence that the threatened"sentencedifferentialbetweenguilty-pleaand trial defendants to increasesin direct proportion the likelihoodof acquittal. See id. Alschulerreportsone case that resemblesthe hypotheticalchoice between death penalty and fifty-cent fine: San Francisco defense attorney Benjamin M. REV. 24Defendersof plea bargainingsometimes try to minimize the forceof this point with a reductio-ad-absurdum argument: granted that plea bargainingis coercive.The defendantwas innocent.) I do believe that plea supra bargainingis used to coercethe waiverof tenable defenses. The question whether significant numbers of innocent people do plead guilty is not. Obviously. whereas the plea bargainrewardsthe accusedwith a lesser sanction. Alschulerthinks that "the greatest pressuresto plead are broughtto bear on defendantswho may be I have mentionedin the text. the sentencing differentialin plea bargainingelicits confessions of guilt that would not be freely tendered. accused is made a criminal The defendantagainst his wishes. not kind. It is knownthat many of those who plead guilty claim that they are innocent. coercivein the same sense as torture. of course. It is. Sometimes. Plea bargaining.Plea bargainingis as coerciveas it has to be for the modernsystem of sanctions. LANGBEIN.1978] Torture and Plea Bargaining 13 limbs crushed if you refuse to confess. and after investigating the case Davis was confident of an acquittal.

. The American counterpartis Rule 11(d) of the FederalRules of CriminalProcedure."but under the threat of being tortured anew if he recanted." D nothingto do but give judgmentand sentence. I have said that Europeanlaw attempted to devise safeguards for the use of torture that proved illusory. confession. can trial procedure do not contradiction.. The architectsof the Europeanlaw of torturesoughtto enhance the reliability of a torture-inducedconfessionwith other safeguards designed to substantiate its factual basis. he must necessarily be calculatingthat there is a significantprobabilitythat the trier will fail to recognizehis innocencedespite the great safeguardsof trial designedto preventsuch error.If trials were perfectlyaccurate. STUD. determining that the plea is voluntaryand not the result of force or threats or of promisesapart from a plea agreement. now aptly describesAmericanlaw: confessioest regina pean26 probationum.The Europeansmade the torturevictim repeat his confession "voluntarily.S.This was the lesson of the medievalEuropeanlaw. We have said that they required a probable cause determination for investigation under torture and that they directed the court to take steps to verify the accuracy of the confession by investigating some of its detail.14 The University of Chicago Law Review [46:3 medieval Europeans. 146-47(1972)."Kerchevalv. 220. Lay Judges in the German Criminal Courts. the court has 135. United States.The social cost of a rule of absolutecertainty-massive releaseof the culpable-would be intolerable.these measuresbear an eerie resemblanceto the supposed safeguardsof the Americanlaw of plea bargaining. the plea agreement is the source of the coercionand already embodies the involuntariness.Ironically. .R. plea bargainingwould be perfectlyaccurate. long as humanjudgmentis fallible. CRIM. and verification. 11(d) (emphasis added). n See Casper & Zeisel.if an innocentaccuseddoes pleadguilty.of course. P. "27 Of course. anyonewhowoulddenigrate becauseit infringesthe rightto trial must also assumethat the trial itself is plea bargaining The response. More is not required.even he will presshis defenseat trial.Foremostamong the illusorysafeguardsof both systems is the doctrinal preoccupationwith characterizingthe induced waiversas voluntary. no workable So morethan minimizeerror.therefore. " FED.. Probablecause is not the same as guilt. 1 J. 223 (1927).which forbidsthe court from accepting a guilty plea without first "addressingthe defendant personally in open court. 274 U.the Americansare now operatinga procedural The system that engages in condemnationwithout adjudication.since no innocent person wouldhave an incentiveto accusehimself..25 maxim of the medieval Glossators. We have explained why these measures were inadequate to protect many innocent suspects from torture. and that paradoxis to some extent recognizedto be mistake-prone. is itself a conviction . and it explains why the standard of our law is not "beyonddoubt" but "beyondreasonable longer applicable to Eurolaw. LEGAL "A plea of guilty .confessionis the queen of proof.

I just pleaded guilty because they said if I didn't they would gas me for it I'm not guilty but I plead guilty. The function of trial.S. becausethey do not give adequateweight to the seriousness with which the law of tortureundertookto separate the guilty from the innocent. It is all falsehood and invention. at 28 n... the case of North Carolina v. in order to tell his daughter why he had confessed to witchcraft "forwhich I must die. so with the negotiated plea: any case that has resisted dismissal for want of probable cause at the preliminary hearingwill rest upon enough inculpating evidence to cast suspicion upon the accused. I think. markedly less reliable than the negotiated plea.2.S. .29decided in this decade. Alford's statement with the explanation of one Johannes Julius. They never cease to torture until one says something. is to requirethe court to adjudicate whetherthe facts proven support an inference of guilt beyond a reasonable doubt. however.). Alford.R.:2 2 2.."28 with the tortured confession. THE WITCH-CRAZE THE16THAND17TH OF Quoted in H. 400 U. 30Id. but I ain't shot no man .. . 32 Some of those who have favoredme with prepublication critiquesof this paper have resistedthis point-largely." Federal Rule 11(f) provides that "the court should not enter judgment upon [a guilty] plea without making such inquiry as shall As satisfy it that there is a factual basis for the plea. But the resulting moral quandaryis the same. 25 (1970). EUROPEAN 84 CENTURIES (1969) (Pelican ed.1978] Tortureand Plea Bargaining 15 if undertaken in good faith. An accused is more likely to bear false witness against himself in order to escape further hours on the rack than to avoid risking a longer prison term.. Supreme Court found it permissible to condemn without trial a defendant who had told the sentencing court: "I pleaded guilty on second degree murderbecause they said there is too much evidence. which plea bargainingeliminates. Consider. of course. who wrote from his dungeon cell where he was awaiting execution. either because the matters confessed were not susceptible of physical or testimonial corroboration. or because the accused might know enough about the crime to lend verisimilitude to his confessioneven though he was not in fact the culprit. My critics suggest that plea bargainingis in theory meant to have a differentialimpact upon the guilty and the innocent. 11(f). because the degree of coercion is greater. 3' Id. could easily fail as a safeguard. so help me God ."31 The tortured confession is. in which the U.. seventeenth-century burgomasterof Bamberg. The American law of plea bargaining has pursued a similar chimera: the requirementof "adequate factual basis for the plea. They contend that the plea bargaining . whereas torture was not. TREVOR-ROPER."30 invite you to compare I .

Zeisel. 206-07. Die Rolle der Geschworenenin den USA. " See..quite apart fromthe increaseddangerof condemningan innocent man. concurring) N Beccaria. Supp. but for something less opprobrious.W."33 Beccaria might as well have been speaking of the coercionof plea bargainingwhen he said of the violence of torture that it "confoundsand obliterates those minute differencesbetween things which enable us at times to know truth from falsehood.) moreseverethan The sentencesprescribed statute in the United States are markedly by forcomparable offensesin Europeanstates. but he could as well have been describingthe tortured confession when he said. When people who have murderedare said to be convictedof wounding.The severityof our prescribed 121. 787 (1968) (Levin. 12 Soc. themselves. "Confessor the pain will continue. as I have said in the text. LANGBEIN.cynicismabout the processesof criminaljustice is inevitably reinforced. 186. 158-59(1964) (child molestationcases resultingin no DefenderOffice. 12 Mich. the differenceis one of degreeand not kind. J. Grigson trans.). "there is no way of knowing whether a particular guilty plea was given because the accused believed he was guilty.2d 777.D. (italics deleted). In the plea bargainingthat takes the form of charge bargaining(as opposed to sentence bargaining). Sudnow.g. but I wouldbe surprisedif HansZeisel werefar wrongin his suggestion to that each month of imprisonmentin Continentalsentences corresponds a year in the United States. 151 n. 162 N. even if the offense was committed nowherenear a schoolyard). 502 (S.See J. 21 OSTERREICHISCHE sentencescontributesto the JURISTENZEITUNG 123 (1966).As for the law of torture. seriousharmto the victim regularlyreducedto the chargeof loiteringarounda schoolyard. 1976)." United States v.55."whereasthe law of torturegave the same messageto both the innocentand the guilty. or because of the promised concession. The factual unreliabilityof the negotiatedplea has furtherconsequences.Y."34 doctrine of adequate factual basis for The the plea is no better substitute for proof beyond reasonabledoubt than was the analogousdoctrine in the law of torture. at 16. Bergman. MANZONI. 3 People v. supra note 24.. and the law of torture made express provisionfor releasingthose who did not confess under torture.NormalCrimes: Featuresof the Penal Codein a Public Sociological 155.(I owe this referenceto ProfesorRichardLempert.416 F. supra note 1. reprintedin 33 OF preface to A.weredesignedforthe sole purpose separating innocentfrom the guilty. Byrd.oursis "a nation where prisonsentences of extravagantlength are morecommonthan they are almost anywhereelse.THECOLUMN INFAMY (1964). e. I have alreadypointedout.I shouldreiteratethat the safeguards of the see text at notes 6-9 supra. . 1764) (J.the culprit is convicted not for what he did. that plea bargainingcan and does induce innocent defendantsto convict discussedabove. In the wordsof Judge Frankel." I think that both branchesof the argumentare mistaken. App.35 This wilful mislabellingplays havoc with system meansto tell the accused.16 The University of Chicago Law Review [46:3 Judge Levin of Michiganwas speakingof the negotiatedguilty plea. I am very willing to concedethat the safeguardsof the law of torturewereeven less effectivethan those of plea but bargaining.N.Of Crimesand Punishment(1st ed. 496. or when those caught stealing are nominally convicted of attempt or possession. PROB.I knowof no study of the point."Don'tput us to the troubleof a trial unlessyou are really innocent.

supra note 11. and resultant Of course. and to increase the public's unease about the administrationof criminaljustice. The case of James Earl Ray is perhaps the best example of public dissatisfaction over the intrinsic failure of the plea bargainingsystem to establish the facts It about crime and guilt in the forumof a public trial. or retributionjustifies the enormousdifferentialsneeded to sustain plea bargaining. 23 LAW CONTEMP. we were not.when the Europeanswere amelioratingtheir of sentences. at 1082-87. plea bargainingis not responsiblefor all of the downcharging viewsthe statuengage. H. e. Those differentialsexist without a moral basis. In the nineteenth and twentieth centuries. ZEISEL. there is a profounddifferencein purposebetween civil and criminalsanctions.g. Henry Hart was surely correctthat "[t]he core of the difference"between a confined mental patient and an imprisonedconvict is "that the patient has not incurredthe moral condemnationof his community. Our law of plea bargaininghas not only recapitulated much of the doctrinal folly of the law of torture. 3 Of course. I think that the unreliability of the plea.If the prosecutor mislabellingin which moder Americanprosecutors without torily prescribedminimumsentence for an offense as too severe.1978] Torture and Plea Bargaining 17 our crime statistics. the mislabelling of the offense. It is for the same reasonthat the sentence differentialsrequiredby plea bargainingare so repugnantto any tenable theoryof sentencing. The Aims of the Criminal & PROB. . See N. The Sacco-Vanzettiand the Rosenbergcases continue to be relitigatedin the forumof plea bargainingleaves the public with what I believeto be a more popularopinion. 11.Y. pronouncedsense of unease about the justness of results.." Hart. 401. Just this concern appears to have motivated the briberycase of Vice-PresidentAgnewto take such extraorgovernmentin the plea-bargained case. culminating in an adjudicationof guilt. supra note 11. The moral force of the criminal sanction depends for part of its efficacy on the sanction having been imposed after rational inquiry into the facts. complete with the pathetic plea bargainingsystem by expanding the potential differentialbetween sentence following trial and sentence pursuantto plea bargain. he can downcharge exacting a concessionaryquid pro quo from the accused. Alschuler. 406 (1958). The public's hostility to plea bargainingshould suggest why some of my colleaguesin the law-and-economics fraternityare mistaken in their complacentassimilationof plea bargaining to the model of the negotiatedsettlement of civil disputes.'However. Law. It is tempting to wonderwhetherthe requirements the plea bargaining system have been somewhatresponsible. To assert the equivalencyof waiver and adjudicationis to overlookthe distinctive characteristicof the criminallaw. which explains in part why Americansuniquely among Western peoples-attach so much importance to arrest records rather than to records of conviction. at 35-38.whereasthe convict has. see.36 is interesting to rememberthat in Europe in the age of Beccaria and Voltaire. by avoidingthe open ventilation of evidence that characterizespublic trial. Times.not every trial resolvesthe questionof guilt of innocenceto public satisfaction. On their extent.Nothing in the theoryof deterrence. dinarysteps to assurethe disclosureof the substanceof the prosecution Oct. ? 1. the want of adjudication and the unreliability of the law of torture had bred a strangely similar cynicism towardsthat criminaljustice system.reformation. 1973. and the underlying want of adjudication all combine to weaken the moral force of the criminal law. Howevergreat the operational similarity.

Plea bargainingconcentrateseffective controlof criminal procedurein the hands of a single officer.38 the plea bargainingsystem has largely dissolved that safeguard. While on the subject of institutional factors. 1964). Our formal law of trial envisagesa division of responsibility. 1965. keep himself in his decisional function free from the predisposinginfluences of his own instigating and investigatingactivity. . but it has also repeated the main institutional blunder of the law of torture. The powerto prosecuteas we knowit containswithin itself the power to persecute. the judge and especially the jury to adjudicate. The modernpublic prosecutor commandsthe vast resourcesof the state for gatheringand generating accusing evidence. most recentlyby the Attorney-General Alaska. 13 LAW& Soc'Y REV. Feb. LEHRKOMMENTAR ZUR STRAFPROZESSORDNUNG UND ZUM GERICHTSVERFASSUNGSGESETZ (2d ed. We allowed him this power in large part because the criminal trial interposedthe safeguardof adjudication against the danger that he might bring those resourcesto bear against an innocent citizen-whether on account of honest error. I have one last comparisonto advance. hence to make himself judge in his own cause. 13. SCHMIDT. torture. convenience-or 37 1 E. Hoffa contended "that special investigatorsfrom the Justice Dept. But arbitrariness. ."37 The dominantversionof Americanplea bargainingmakessimilar demands: it requiresthe prosecutorto usurp the determinative and sentencing functions. " One need not necessarily accept Jimmy Hoffa's view that Robert Kennedy was con- hundreds of agents from the Federal Bureau of Investigation were used to satisfy 'a personal Gross. . investigation.18 The University of Chicago Law Review [46:3 safeguardsof voluntarinessand factual basis that I have just discussed. I cannot emphasize too stronglyhow dangerousthis concentration of prosecutorial powercan be. 48.Students of the history of the law of torture are remindedthat the great psychologicalfallacy of the Europeaninquisitorialprocedureof that time was that it concentratedin the investigatingmagistratethe powersof accusation. 197 ductinga personaland politicalvendettaagainst him in orderto appreciatethe dangerthat he might have been.or worse. (1979) (forth- coming). In short.The single inquisitor who wielded those powersneeded to have what one recent historian has called "superhumancapabilities [in order to] . and condemnation. WEEK. 3. and sanctional phases of the procedurein the hands of the prosecutor. and hate' of RobertKennedy. Plea bargaining merges these accusatory. prosecutors.We expect the prosecutor to make the chargingdecision. and defense counsel. and the judge to set the sentence. The point has been made. determinative.39 of that preparingand taking cases to trial is much harderworkthan plea bargaining-for police. Plea Bargaining: The Alaska Experience."Bus.

137.1978] Torture and Plea Bargaining 19 worse. ONTHELAWS ENGLAND COMMENTARIES OF *330 (1768). at 1103 n. IV. and plea bargaining lives on although its evils are quite familiar to us all. Why is it so hard for a legal system to reforma decadent system of proof?I think that there are two main reasons. BLACKSTONE.S. before it will admit the need for reform in its system of proof and trial.McNabb v. Here as elsewhere the old adage may apply that if necessity is the mother of invention. The most important is this: a legal system will do almost anything.1 (1883). for one where the law is doubted of. We suppose that this factor had a little to do with torture as well. at 147 n. sloth-is a factor that sustains plea bargaining.F."41Every institution of the legal system is geared to the system of proof. supra note 11. He said: "experience will abundantly shew. laziness is the father. virtually every vested interest. One is in a sense practical: nothing is quite so imbedded in a legal system as the proceduresfor proof and trial. . and it is even more true of criminal litigation. Alschuler. tolerate almost anything. The inertia. because most of what a legal system does is to decide matters of proof-what we call fact-finding.A HISTORY THE OF CRIMINAL OFENGLAND n. was this accused the man who fired the shot or robbed the bank?) Blackstone emphasized this point in speaking of civil litigation. What makes such shoddy subterfuges so tenacious is that they shield their legal systems from having to face up to the fact of breakdownin the formal law of proof and trial. the resistance to change that is associated with 40 1 J. Stephen's LAW 442 forcefulquotationhas been cited forthis point elsewhere. supra note 1. that above a hundred of our lawsuits arise from disputed facts. I want to draw some conclusionsabout what I regardas the lessons of the exercise. 318U."40If we were to generalize about this point. at one level or another. J. "It is far pleasanter to sit comfortablyin the shade rubbing red pepper into a poor devil's eyes than to go about in the sun hunting up evidence. THE JURISPRUDENCE OFCONCESSIONARY CRIMINAL PROCEDURE Having developed these parallels between tortureand plea bargaining.8 (1943). 332. As someone in India remarkedto Sir James Fitzjames Stephen in 1872about the proclivity of the native policemen for torturing suspects. (Was the traffic light green or red. we might say that concessionary criminal procedural systems like the plea bargaining system and the system of judicial torture may develop their own bureaucracies and constituencies. 41 3 W. 344 n.14. STEPHEN. United States. forthright reconstructionwould disturb. LANGBEIN. The law of torture endured for half a millennium although its dangers and defects had been understood virtually from the outset.

I EDWARDAND LAW See T. We have respondedto the breakdown of our formal system of proof by taking steps to perpetuate the ideologyof the failed any society the adjudicativepowermust be rested on a theoreticalbasis that makes it palatable to the populace. The Anglo-American jury system invoked the inscrutable wisdom of the folk to justify its results. For example.20 The University of Chicago Law Review [46:3 such deep-seated interests.especially criminaladjudication. Adjudication. The medieval Europeanlaw of proof assured people that. I have elsewhere pointed out how in the nineteenth centurythe ideologicalattachmentto the jury retardedexperimentationwith juryless trial-that is.F.precisely because they are not framed in a 42 CRIMINAL 74-75 (1960). plays a central role in legitimatingthe entire legal system. and we have arrangedto place defendants under fierce pressure to "choose"the substitute. supra note 12.the system will respondby developing subterfugesthat overcomethe formal law. Consequently. . there are intrinsic limits to the level of complexityand safeguardthat even a civilized people can tolerate. But subterfugesare intrinsically overbroad. we have devised a substitute nontrialprocedureto subvertthe formalprocedure.43 Like the medieval Europeanlawyersbeforeus.42 Each of these theories was ultimately untenable-the European theoryvirtuallyfromits inception. we have been unable to admit that our theory of proof has resulted in a level of proceduralcomplexity and safeguardthat rendersour trial procedureunworkable in all but exceptional cases. 43 Langbein.Because the theoryof proof purportsto govern and explain the application of the adjudicative power.the legal system would achieve certainty.involves a profound intrusion into the lives of affected citizens.Like the medievalEuropeans. steps that closelyresemblethose taken by the architectsof the law of torture. That this script could have been played out in a pair of legal cultures so remote from each other in time and place invites some syssuggestionsabout the adaptive processesof criminalprocedural tems. we have preservedan unworkabletrial procedurein form. is inevitably reinforcedby the powerful ideological component that underlies a system of proof and trial. what we now call bench trial-while the plea bargaining system of juryless nontrial procedurewas taking shape out of public sight.Yet the ideological importanceof these theories preventedeither legal system from recanting upon them. If those limits are exceeded and the repressive capacity of the criminal justice system is thereby endangered.T. the Anglo-American theoryafter a centuries-longtransformationof jury procedure. First.

the presumptionof innocence. 257 (1971). In cases like Griffinv.46 I have made no secret of and my admiration for the brilliant balance that it strikes between safeguard and procedural effectiveness. In its sad plea bargaining opinions of the 1970s.the sacrifice of our fundamental values through plea bargaining is needless. the prosecutorialburdenof proofbeyond reasonable doubt.1978] Torture and Plea Bargaining 21 careful. Chief Justice Burger explained that plea bargaining "is to be encouraged" because "[i]f every criminal charge were subjected to a full-scale trial." Id.a middle path between the impossible system of routine adversaryjury trial and the disgraceful nontrial system of plea like fashion.S. I have described the German system in detail elsewhere. while in the plea bargaining system-which is our routine procedurefor processingcases of serious crime-we have eliminated practically every trace of the privilege. Not the least of its achieve44 380 U.COMPARATIVE . Furthermore. the privilege against self-incrimination. the right to confrontand cross-examineaccusers.45 the Court has simply assumed that the present nontrial plea bargainingprocedureis the inevitable alternative. LANGBEIN. The upshot is that the criminaljustice system is saddled with a lowerlevel of safeguard than it could and would have achieved if it had not pretended to retain the unworkableformal system. Especially in its handling of the privilege against self-incrimination does American criminal procedure reach the outer bounds of incoherence. at 260. 46 GERMANY PROCEDURE: CRIMINAL (1977). the States and the Federal Government would need to multiply by many times the number of judges and court facilities. 404 U. The medieval Europeans insisted on two eyewitnesses and wound up with a law of torture that allowed condemnationwith no witnesses at all. See generally J. The contemporarynonadversarial criminal justice systems of countries like West Germany have long demonstrated that advanced industrial societies can institute efficient criminal procedures that nevertheless provide for lay participation and for full adjudication in every case of serious crime.That path is a streamlined nonadversarialtrial procedure. There is. sacrifices just those values that the unworkablesystem of adversaryjury trial is meant to serve: lay participation in criminal adjudication. explicit. Americanplea bargaining. New York.S. 609 (1965). 45 In Santobello v. and principled manner directed to achieving a proper balance between repression and safeguard. the Supreme Court has effectively admitted that for reasons of expediency American criminal justice cannot honor its but promise of routine adversarycriminal trial. California4 we have exaggeratedthe privilege to senseless lengths in formal doctrine. however.

nor does a confession excuse the trial court from hearing sufficient evidence for conviction on what amounts to a In standard of proof.whoseordinary is to dispositive is withoutsuch evasions. but because there is no advantage to be wrung from the proceduralsystem by withholdingthem. 457 n. supra note 46. at 77. ContinentalCriminalProcedure: L. I believe that we will find in modernContinentalcriminalprocedure an irresistible model for reform. 12. 87 YALE 1549. 1561-65(1978). supranote 20. Study. they are not and cannot be but bargained for. REV. 73-74. "The resignationoccurred as part of a 'cow-trade. I hope that over the coming decades we who still live under criminaljustice systems that engagein condemnationwithout adjudication will face up to the failure of adversarycriminalprocedure. LANGBEIN. at 66-67. " See J.22 The University of Chicago Law Review [46:3 ments is that in cases of serious crime it functions with no plea bargaining whatsoever. 121 U.peopleswho live in contentment with their criminaljustice systems look out across the sea in disbelief5' to the spectacle of plea bargainingin America.Today in lands wherethe law of tortureonce governed. he cannot inflict significant costs upon the prosecution by contesting an overwhelming case.44 (1974): Plea bargaining all but incomprehensible the Germans.'as it can only in the United States be imagined. 51 See Langbein.50 be put to trial whether he confessesor not. That. 552 (1973). at 150. 17For recent discussion see Langbein & Weinreb. Confessions are tendered at trial not because they are rewarded. I am broughtto concludewith a paradox. 4 Casper& Zeisel.In the Germanpressthe judicialprocedure procedure workable the of surrounding resignation Vice PresidentAgnewwas viewedwith the sortof wonder normallyinspiredby reportsof the customs of primitivetribes. Casper& Zeisel." BadischeZeitung. "Myth"and Reality. supra note 26. at 146-47. col. supra note 26. at See id.47Confessions are still tendered in many cases (41 percent in one sample48).150-51.49 a trial procedure beyond-reasonable-doubt shorn of all the excesses of adversary procedureand the law of evidence. however. .Oct. 3. 1973. Thus. the time differencebetween trial without confessionand Because an accused will trial with confessionis not all that a theme about which I can say no more if I am to remainwithin the propersphere of the CrosskeyLecture in Legal History.Evidentiary A of CriminalProcedure: Comparative 506. PA.L. Damaska. while American tourists come by the thousands each year to gawk in disbelief at the decaying torture chambersof medieval castles. For the observationthat modem Continentalnonadversarial confersmoresafeguardsupon the accusedthan does procedure American see Barriers Conviction TwoModels to and plea bargaining.