Presuppositions of evidence law

Presuppositions of Evidence Law Rutgers University has made this article freely available. Please share how this access benefits you. Your story matters. [https://rucore.libraries.rutgers.edu/rutgers-lib/57243/story/] This work is an ACCEPTED MANUSCRIPT (AM) This is the author's manuscript for a work that has been accepted for publication. Changes resulting from the publishing process, such as copyediting, final layout, and pagination, may not be reflected in this document. The publisher takes permanent responsibility for the work. Content and layout follow publisher's submission requirements. Citation for this version and the definitive version are shown below. Citation to Publisher Leubsdorf, John. (2006). Presuppositions of Evidence Law. Iowa Law Review 91(4), 1209Version: 1258.https://heinonline.org/HOL/P?h=hein.journals/ilr91&i=1221. Citation to this Version: Leubsdorf, John. (2006). Presuppositions of Evidence Law. Iowa Law Review 91(4), 1209-1258. Retrieved from http://dx.doi.org/doi:10.7282/T3WQ0760. Terms of Use: Copyright for scholarly resources published in RUcore is retained by the copyright holder. By virtue of its appearance in this open access medium, you are free to use this resource, with proper attribution, in educational and other non-commercial settings. Other uses, such as reproduction or republication, may require the permission of the copyright holder. Article begins on next page SOAR is a service of RUcore, the Rutgers University Community Repository RUcore is developed and maintained by Rutgers University Libraries LEUBSDORF_FINAL 7/19/2006 7:36:19 AM Presuppositions of Evidence Law John Leubsdorf∗ I. INTRODUCTION .....................................................................................1210 II. THE ATOMIC THEORY OF EVIDENCE .....................................................1213 A. ITEMIZING TESTIMONY ....................................................................1213 B. SPLITTING THE ATOM: INFERENCES ..................................................1218 C. THE TRANSACTION IN QUESTION .....................................................1221 D. AN OPPOSITE REACTION: GENERALIZATIONS ....................................1227 III. MYTHS OF PRESENCE: VOUCHING FOR NONTESTIMONIAL EVIDENCE ...1234 A. THE TESTIFYING ESCORT .................................................................1234 B. RATIONALIZATIONS .........................................................................1237 C. HISTORY SPEAKS .............................................................................1241 IV. AMBIVALENCE AND TRIAL STRUCTURE .................................................1244 A. REASON AND EMOTION ....................................................................1245 1. Bad Emotion..........................................................................1245 2. Good Emotion.......................................................................1246 B. JURORS ...........................................................................................1248 1. Bad Jurors ..............................................................................1248 2. Good Jurors ...........................................................................1251 C. JURORS, EMOTIONS, AND THE STRUCTURE OF THE TRIAL ..................1253 V. CONCLUSION ........................................................................................1257 ∗ Professor of Law, Rutgers School of Law-Newark. Many thanks to Craig Callen, Sherry Colb, Dale Nance, Roger Park, and the participants in a Rutgers colloquium for their helpful comments. Thanks to John Mansfield for introducing the author to the law of evidence and to the Dean’s Research Fund of Rutgers School of Law-Newark for support. 1209 LEUBSDORF_FINAL 7/19/2006 7:36:19 AM 1210 91 IOWA LAW REVIEW I. [2006] INTRODUCTION The land of evidence has a weird logic or illogic that is all its own. This is a realm in which excitement makes people more reliable,1 in which one may accept reports of what someone said to establish what she planned to do but not what she had already done,2 and in which those considering whether someone robbed a bank may not be told that he is a professional bank robber.3 It is a realm founded on the untrustworthiness of jurors, in which jurors are nevertheless trusted to follow instructions to disregard obviously relevant uses of evidence.4 It is a world in which the Supreme Court declines to reform a group of rules, not because they make sense, but because they are so feeble that “[t]o pull one misshapen stone out of the grotesque structure is more likely simply to upset its present balance between adverse interests than to establish a rational edifice.”5 Not many fields of law can thus be described as too irrational to be improved. This Article demonstrates the incompleteness of the common view that evidence law’s strangeness results from its failure to pursue in the best way the goal of securing accurate adjudication of facts.6 To be sure, much can be said for that goal as a normative matter. And it is sufficiently hard to warrant the conclusion that any system pursuing the goal will be both complex and open to criticism. Sometimes it makes sense to exclude relevant evidence because it is more likely to mislead than to help the trier of fact7 or because its exclusion may stimulate the introduction of superior evidence.8 Sometimes specific rules may promote the goal,9 while in other situations judicial discretion may be preferable. Lawmakers may have relied on generalizations about human behavior that can be shown to be 1. FED. R. EVID. 803(2); see, e.g., Richard D. Friedman & Bridget McCormack, Dial-In Testimony, 150 U. PA. L. REV. 1171, 1234 (2002) (criticizing rule). 2. FED. R. EVID. 803(3); see, e.g., Paul S. Milich, Hearsay Antinomies: The Case for Abolishing the Rule and Starting Over, 71 OR. L. REV. 723, 757 (1992) (criticizing rule); see also infra notes 108–14 and accompanying text (discussing Mutual Life Ins. Co. v. Hillmon, 145 U.S. 285 (1892)). 3. FED. R. EVID. 404. See generally Sparks v. Gilley Trucking Co., 992 F.2d 50 (4th Cir. 1993); Pankey v. Commonwealth, 485 S.W.2d 513 (Ky. 1972). Admittedly, this principle has been distinguished so often as to be almost nonexistent. See generally United States v. MojicaBaez, 229 F.3d 292 (1st Cir. 2001); United States v. Queen, 132 F.3d 991 (4th Cir. 1997). 4. See Shari Seidman Diamond & Neil Vidmar, Jury Room Rumination on Forbidden Topics, 87 VA. L. REV. 1857, 1860 (2001). 5. Michelson v. United States, 335 U.S. 469, 486 (1948). The rulemakers have subsequently implemented one tiny reform in the rules that Michelson describes. FED. R. EVID. 405(b) & advisory committee’s note. 6. See WILLIAM TWINING, THEORIES OF EVIDENCE: BENTHAM AND WIGMORE 17 (1985) (describing the “rationalist tradition” of evidence scholarship). 7. FED. R. EVID. 403. 8. See generally Dale A. Nance, The Best Evidence Principle, 73 IOWA L. REV. 227 (1988). 9. TWINING, supra note 6, at 66–75. LEUBSDORF_FINAL PRESUPPOSITIONS OF EVIDENCE LAW 7/19/2006 7:36:19 AM 1211 untrustworthy.10 They may have been ignorant of what probability theory can teach about the weight and cumulation of evidence.11 And forces of inertia, notably the resistance of the trial bar, may have kept evidence law behind the times. These reasons explain some, but by no means all, of the odd features of evidence law. Likewise, this Article shows that deeper presuppositions are at work in evidence law beneath the clash between truth-seeking and competing goals. That clash undoubtedly exists.12 Efforts to limit the cost and delay of litigation may exclude relevant evidence.13 Under a regime of party autonomy, the parties may specify not just the facts to be proved but the evidence to prove them, opening the door to evidence otherwise inadmissible14 or excluding evidence by stipulation.15 Various policies may supervene: privileges protect privacy or promote professional prestige;16 facilitating the punishment of crime must be reconciled with maintaining safeguards for defendants;17 judges inflect evidentiary rules to reach substantive goals;18 and considerations of governmental structure and 10. This point dates back at least to 1928. Robert M. Hutchins & Donald Slesinger, Some Observations on the Law of Evidence, 28 COLUM. L. REV. 432 (1928). For more modern discussions, see generally ELIZABETH F. LOFTUS & JAMES M. DOYLE, EYEWITNESS TESTIMONY (3d ed. 1992); LUCY S. MCGOUGH, CHILD WITNESSES: FRAGILE VOICES IN THE AMERICAN LEGAL SYSTEM (1994); Richard O. Lempert, Built on Lies: Preliminary Reflections on Evidence Law as an Autopoietic System, 49 HASTINGS L.J. 343 (1998); Olin Guy Wellborn III, Demeanor, 76 CORNELL L. REV. 1075 (1991). 11. See generally Richard Friedman, Character Impeachment Evidence: Psycho-Bayesian [!?] Analysis and a Proposed Overhaul, 38 UCLA L. REV. 637 (1991). 12. See generally Michael L. Seigel, A Pragmatic Critique of Modern Evidence Scholarship, 88 NW. U. L. REV. 995 (1994); Symposium, Truth & Its Rivals: Evidence Reform and the Goals of Evidence Law, 49 HASTINGS L.J. 289 (1998). 13. See FED. R. EVID. 403, 408, 608(b), 613(b) (excluding evidence on certain grounds, including economy). See generally Richard A. Posner, An Economic Approach to the Law of Evidence, 51 STAN. L. REV. 1477 (1999). Excluding evidence to reduce cost or delay may be consistent with the search for truth: the jury may not be able to handle too much evidence. See generally Craig R. Callen, Rationality and Relevancy: Conditional Relevancy and Constrained Resources, 2003 MICH. ST. L. REV. 1243. 14. See FED. R. EVID. 103(a)(1), 404(a)(1)–(2), 608(a)(2) (outlining rules under which one party’s decisions affect what others may do); United States v. Piccinonna, 885 F.2d 1529, 1536 (11th Cir. 1989) (allowing admission of polygraph evidence by stipulation); JOHN W. STRONG ET AL., MCCORMICK ON EVIDENCE §§ 55–57, 93 (5th ed. 1999). 15. See FED. R. CIV. P. 36(b); Old Chief v. United States, 519 U.S. 172, 190–92 (1997) (requiring the prosecution to accept the defendant’s offer to stipulate in certain cases). 16. See Jaffee v. Redmond, 518 U.S. 1, 15 (1996) (protecting disclosures to psychotherapists). 17. See FED. R. EVID. 404, 413, 609, 704(b) (describing the admissibility of past bad acts and convictions, and expert opinions as to the insanity defense); Kenneth W. Graham, Jr., “There’ll Always Be an England”: The Instrumental Ideology of Evidence, 85 MICH. L. REV. 1204, 1213– 14 (1987) (describing evidence law as political and privileging the powerful). 18. Thus Daubert v. Merrell Dow Pharmaceutical, Inc., 509 U.S. 579 (1993), has been used to bar tort claims, but more leniently applied to the prosecution’s evidence in criminal cases. Margaret A. Berger, Upsetting the Balance Between Adverse Interests: The Impact of the Supreme Court’s LEUBSDORF_FINAL 1212 7/19/2006 7:36:19 AM 91 IOWA LAW REVIEW [2006] constitutional law affect the allocation of power between juries and judges.19 And because trials are (in addition to much else) ritual and theater, the search for dramatic coherence and spectatorial impact may inflect the search for truth.20 Without scoffing at these insights, this Article tries to trace some patterns of thinking that underlie and shape evidence law at a deeper level. The way in which one sees and analyzes trials must affect how one tries to govern them, no matter how one conceives their goals. That is not to say that the choice of approach is value free. On the contrary, the commitments of evidence law have not been chosen at random. And commitments have consequences. Although none of us has consciously created the world of evidence law, it has helped to create us. This Article seeks to show the implications of three commitments of evidence law, each of them a presupposition that underlies many evidentiary rules. The first is methodological: evidence is to be analyzed particle by particle and inference by inference. The second might be called epistemological: all evidence must be grounded on the testimony of a witness present in court. The third presupposition might better be described as a psychological field of force, in which law is justified by ambivalent and contrary distinctions between reason and emotion, and between the strengths and weaknesses of jurors, distinctions that turn out to be based less on reality than on the structural requirements of trials. Despite its somewhat anthropomorphic references to what “evidence law” does, this Article does not claim that law speaks with one consistent voice or derives from one consistent set of values or axioms. What it does claim is that these and perhaps other assumptions help shape the ways in which people perceive and create evidence law, and not always for the best. Indeed, even the twentieth-century trend to increase judicial discretion to admit or exclude evidence21 has left older assumptions in place and perhaps protected them from challenge by blurring their practical impact. Trilogy on Expert Testimony in Toxic Tort Litigation, 64 LAW & CONTEMP. PROBS. 289, 325 (2001); Paul C. Giannelli, The Supreme Court’s “Criminal” Daubert Cases, 33 SETON HALL L. REV. 1071, 1072–77 (2003); Michael J. Saks, The Legal and Scientific Evaluation of Forensic Science (Especially Fingerprint Expert Testimony), 33 SETON HALL L. REV. 1167, 1173–76 (2003). 19. See FED. R. EVID. 104 (governing which admissibility issues are decided by judges); Daubert, 509 U.S. at 595–96; Jackson v. Denno, 378 U.S. 368, 377–81 (1964). The examples given in the text by no means exhaust the instances in which policies unrelated to the ascertainment of truth influence evidence law. See FED. R. EVID. 407–412, 610. 20. See Desmond Manderson, Et Lex Perpetua: Dying Declarations and the Terror of Süssmayr, in COURTING DEATH: THE LAW OF MORTALITY 34, 48 (Desmond Manderson ed., 1999); Mark Cammack, Evidence Rules and the Ritual Functions of Trials: “Saying Something of Something,” 25 LOY. L.A. L. REV. 783, 788–91 (1992); Graham, supra note 17, at 1214; Charles Nesson, The Evidence or the Event? On Judicial Proof and the Acceptability of Verdicts, 98 HARV. L. REV. 1357, 1367–68 (1985). 21. Eleanor Swift, One Hundred Years of Evidence Law Reform: Thayer’s Triumph, 88 CAL. L. REV. 2437, 2441 (2000). LEUBSDORF_FINAL 7/19/2006 7:36:19 AM PRESUPPOSITIONS OF EVIDENCE LAW 1213 II. THE ATOMIC THEORY OF EVIDENCE What students in evidence courses learn is, in large part, a method of particularization. Testimony is considered answer by answer. Each answer is considered inference by inference. Moreover, it is assumed that jurors will also particularize and can therefore be asked to consider a given morsel of evidence for one purpose while disregarding its other uses. If you can think of something that is inextricably connected to something else without thinking of the thing to which it is connected, you may be an evidence scholar.22 But you will probably not be much of a trial lawyer, because the atomized approach of evidence law must be deployed within a trial that makes sense to its participants and observers only from a very different perspective. Indeed, evidence law itself mingles particularization with overgeneralizations about classes of evidence. Those who apply it are thus called to shift back and forth between microscope and telescope. A. ITEMIZING TESTIMONY Analyzing testimony answer by answer is a relatively recent development. Even in the eighteenth century, Anglo-American evidence law was mainly a law concerning the exclusion of witnesses. Those disqualified from testifying included parties, persons with an interest in the result, convicted felons, and atheists.23 Such persons could not testify at all, so it was unnecessary to consider their possible testimony item by item. It is true that more narrowly defined evidence rules also existed, for example, the attorney-client privilege and the emerging exclusion of hearsay.24 But rules excluding witnesses constituted the great bulk of evidence law and must have far overshadowed other rules in their practical effect. As legislatures repealed witness-incompetency rules during the nineteenth century, judges expanded other evidence rules, following the law of the conservation of complexity that seems to be inherent in our legal 22. This sentence paraphrases one attributed to Thomas Reed Powell. See A. Simpleman, Jr. [Peter R. Teachout], Sentimental Metaphors, 34 UCLA L. REV. 537, 545 n.17 (1986) (book review) (discussing the source of the statement). 23. Omychund v. Barker, (1744) 26 Eng. Rep. 15, 32 (Ch.); SIR GEOFFREY GILBERT, THE LAW OF EVIDENCE 86–104 (1754); 2 JOHN HENRY WIGMORE, A TREATISE ON THE ANGLOAMERICAN SYSTEM OF EVIDENCE IN TRIALS AT COMMON LAW §§ 515–531, 575–577, 579–580, 600– 603 (3d ed. 1940); James Oldham, Truth-Telling in the Eighteenth-Century English Courtroom, 12 LAW & HIST. REV. 95, 96 (1994). In the United States, race could be another ground for disqualification. See generally People v. Hall, 4 Cal. 399 (1854). Gender did not disqualify, but marriage disqualified testimony by one spouse for or against the other. WILLIAM BLACKSTONE, 1 COMMENTARIES *443. 24. JOHN H. LANGBEIN, THE ORIGINS OF ADVERSARY CRIMINAL TRIAL 178–252 (2003); T.P. Gallanis, The Rise of Modern Evidence Law, 84 IOWA L. REV. 499, 531–36 (1999); John H. Langbein, Historical Foundations of the Law of Evidence: A View from the Ryder Sources, 96 COLUM. L. REV. 1168, 1186 (1996). LEUBSDORF_FINAL 1214 7/19/2006 7:36:19 AM 91 IOWA LAW REVIEW [2006] system.25 The result was that, although more witnesses were allowed on the stand, individual questions and answers could more often be challenged on such grounds as hearsay, privilege, the best-evidence rule, the opinion rule, or the limits on cross-examination. The body of law seeking to define these doctrines steadily expanded. By 1904, when John Wigmore published the first edition of his treatise, he took four volumes to state the law of evidence, which William Evans had covered in 160 pages a century before.26 The main reason for this change seems to have been that the participation of lawyers in the examination and cross-examination of witnesses steadily increased, especially in criminal cases. Instead of simply and quickly telling their stories, witnesses answered lawyers’ questions. Between the efforts of the questioning lawyer to probe deeper and those of opposing counsel to limit the harm, objections multiplied. Ruling on objections, judges shaped the law of evidence to focus on the propriety of individual questions and the admissibility of individual answers.27 This development paralleled an increasing tendency to rely on cross-examination rather than the oath as the main guarantee for the trustworthiness of testimony.28 It may also have developed in part from an eighteenth-century English preoccupation with “facts,” thought of as nuggets of ascertainable truth—a preoccupation that was also reflected in the rise of science, newspapers, and the novel.29 Whatever the history of the assumption that each crumb of evidence is to be considered separately, its effects pervade evidence law. For example, when the admissibility of evidence depends on its reliability under the Confrontation Clause, courts typically decline to consider whether corroborating evidence might affect that reliability: each statement must 25. See generally C.J.W. ALLEN, THE LAW OF EVIDENCE IN VICTORIAN ENGLAND (1997); George Fisher, The Jury’s Rise as Lie Detector, 107 YALE L.J. 575 (1997); Gallanis, supra note 24. 26. Compare William Evans, Of the Law of Evidence, in M. POTHIER, A TREATISE ON THE LAW OF OBLIGATIONS app. XVI (W.D. Evans trans., 2d ed. 1839) (1806), with WIGMORE, supra note 23. Likewise, compare THOMAS PEAKE, A COMPENDIUM OF THE LAW OF EVIDENCE (1801) (190 pages), with CHARLES F. CHAMBERLAYNE, A TREATISE ON THE MODERN LAW OF EVIDENCE (1911– 16) (five volumes). 27. On the assertions in this paragraph, see LANGBEIN, supra note 24, at 178–250; ALLYSON N. MAY, THE BAR AND THE OLD BAILEY, 1750–1850, at 87–118 (2003); J.M. Beattie, Scales of Justice: Defense Counsel and the English Criminal Trial in the Eighteenth and Nineteenth Centuries, 9 LAW & HIST. REV. 221, 239–45 (1991); Gallanis, supra note 24, at 537–50. For a colorful example of testimony in narrative form dating from 1726, see J.M. BEATTIE, POLICING AND PUNISHMENT IN LONDON, 1660–1750: URBAN CRIME AND THE LIMITS OF TERROR 202–03 (2001). 28. Stephan Landsman, From Gilbert to Bentham: The Reconceptualization of Evidence Theory, 36 WAYNE L. REV. 1149, 1151 (1990); see also sources cited supra note 24. See generally Fisher, supra note 25; supra note 24 (citing authority). 29. See generally BARBARA J. SHAPIRO, A CULTURE OF FACT: ENGLAND, 1550–1720 (2000). The developments that Shapiro describes antedate the shift of evidence law, but this would not be the only instance in which law has lagged behind other disciplines. LEUBSDORF_FINAL PRESUPPOSITIONS OF EVIDENCE LAW 7/19/2006 7:36:19 AM 1215 stand on its own circumstances.30 Likewise, if a statement is inadmissible hearsay because it was made out of court, the fact that the declarant is now on the stand and subject to cross-examination does not automatically render the statement admissible.31 Nor does it help that the testimony of other witnesses might provide enough information about the making of the hearsay statement to permit the jury to appraise its reliability.32 Similarly, a striking coincidence between a hearsay statement and a statement made independently by someone else does not render the hearsay admissible.33 Even when some of a declarant’s statements are admitted under exceptions to the hearsay rule, the jury is often not allowed to use other hearsay statements to gain a complete picture of how the declarant’s story evolved.34 Evidentiary analysis through atomization is not limited to issues of confrontation and hearsay. The limits on leading questions presuppose that one can find and prevent improper suggestions to witnesses by parsing a particular question, without regard to the course of the examination as a whole (including previous leading questions disallowed by the court but heard by the witness) and the examiner’s opportunities to “prepare” the witness before trial.35 Indeed, one judge required a plaintiff representing himself to question himself and then upheld objections to some questions 30. Idaho v. Wright, 497 U.S. 805, 819–24 (1990) (holding that the admissibility of a child’s statement to a pediatrician is unaffected by corroborating evidence). But see Chambers v. Mississippi, 410 U.S. 284, 298–303 (1973) (considering other evidence in appraising a due process claim). The validity and scope of cases like Wright have recently been thrown into question by Crawford v. Washington, 541 U.S. 36, 53–56 (2004). 31. See FED. R. EVID. 801(c); People v. Johnson, 441 P.2d 111, 114–17 (Cal. 1968). There are, however, exceptions. FED. R. EVID. 801(d)(1). 32. See generally Eleanor Swift, A Foundation Fact Approach to Hearsay, 75 CAL. L. REV. 1339 (1987). 33. E.g., Williams v. Alexander, 129 N.E.2d 417, 421–22 (N.Y. 1955) (excluding hearsay report of plaintiff’s alleged hearsay statement despite its striking coincidence with defendant’s story). In other circumstances, evidence law relies on just such coincidences. E.g., United States v. Muscato, 534 F. Supp. 969, 976 (E.D.N.Y. 1982) (holding that evidence that the witness described the gun before it was found is admissible to corroborate his testimony that he got the gun from defendant); see also Westfield Ins. Co. v. Harris, 134 F.3d 608, 613–15 (4th Cir. 1998) (finding plaintiff’s seven questionable past fire claims admissible to show likelihood of arson in that case); Edward J. Imwinkelried, The History of the Uses of the Doctrine of Chances as Theory of Admissibility for Similar Fact Evidence, 22 ANGLO-AM. L. REV. 73 (1993). 34. See generally Tome v. United States, 513 U.S. 150 (1995). Otherwise inadmissible portions of the same statement might, however, be admissible to explain or qualify the admitted portions. See FED. R. EVID. 106; see also United States v. Stover, 329 F.3d 859, 866–68 (D.C. Cir. 2003) (allowing use of otherwise inadmissible evidence to show that admissible prior consistent statements were, in fact, made). 35. Federal Rule of Evidence 611(c) maintains but weakens the traditional limits that John Wigmore has described. 3 WIGMORE, supra note 23, §§ 769–779. On preparing witnesses, see generally John S. Applegate, Witness Preparation, 68 TEX. L. REV. 277 (1989); Symposium, Witness Preparation, 30 TEX. TECH L. REV. 1333 (1999). LEUBSDORF_FINAL 1216 7/19/2006 7:36:19 AM 91 IOWA LAW REVIEW [2006] on the ground that the witness was leading himself.36 Common-law rules governing lay-opinion evidence and opinion evidence on “ultimate issues”— which, like the leading-question rules, have now been somewhat relaxed— likewise sought to prevent confusion and deception through minute regulation of the wording of questions and answers, without much regard to the rest of the witness’s testimony and the opportunities for clarification through cross-examination.37 Other rules regulate the form of questions to ensure that evidence will be presented in sufficiently small lumps, for example, by barring questions that are compound or that assume facts not in evidence.38 Needless to say, presenting testimony in narrative form is normally forbidden, although judges may allow it if they choose.39 All of these various rules thus implement a commitment to analyze testimony bit by bit. By describing how the law breaks down testimony into atomic particles, this Article does not mean to pass judgment. Minute examination is not always the same as myopia. Often, it is desirable. For example, comparing the details of witnesses’ testimony has long been recognized as a good way to probe their veracity.40 Yet, allowing a witness to tell his or her story before being questioned also has its advantages and is the usual practice in civil-law systems when oral testimony is presented.41 A witness who testifies in this way may be less likely to be led by the questioners and more likely to add useful details. The trier may grasp and appraise the witness’s story more easily if it is not broken up by lawyers’ questions and arguments.42 Should the witness omit or distort anything, later questioning can bring that to light. Even from an analytic perspective, starting from the whole rather than considering each part by itself may provide a sounder basis for deciding what a jury should be allowed to hear. American evidence law does not always pursue particularization to the bitter end. For example, judges have the discretion to exclude cumulative 36. Hutter N. Trust v. Door County Chamber of Commerce, 467 F.2d 1075, 1078 (7th Cir. 1972) (ordering a new trial on appeal). 37. FED. R. EVID. 701, 704; 7 WIGMORE, supra note 23, chs. 67–70; id. §§ 1920–1921. 38. For common courtroom objections to the form of questions, see STEVEN GOODE & OLIN GUY WELLBORN III, COURTROOM EVIDENCE HANDBOOK 291–97 (4th ed., student ed. 2001). 39. 4 JACK B. WEINSTEIN, WEINSTEIN’S FEDERAL EVIDENCE § 611.01[2] (Joseph M. McLaughlin ed., 2d ed. 2002); see also Evidence Act, 1995, § 29 (Austl.) (providing that a judge may allow narrative testimony). In cases tried by judges, expert testimony is sometimes presented in report form, subject to cross-examination. See FED. R. CIV. P. 26(a)(2)(B) (requiring pretrial disclosure of expert reports). 40. See Susanna 1:50–:51. 41. MIRJAN R. DAMAŠKA, EVIDENCE LAW ADRIFT 92–93 (1997); PETER L. MURRAY & ROLF STÜRNER, GERMAN CIVIL JUSTICE 295–96 (2004). 42. William T. Pizzi, Crime Victims in German Courtrooms: A Comparative Perspective on American Problems, 32 STAN. J. INT’L L. 37, 42–44 (1996). LEUBSDORF_FINAL PRESUPPOSITIONS OF EVIDENCE LAW 7/19/2006 7:36:19 AM 1217 evidence and also to admit certain hearsay when its probative value exceeds that of other reasonably available evidence, which in each case requires reference to other evidence.43 Likewise, the relevance of one item of evidence often depends on the relevance of another item that has been, or will be, introduced.44 Further, the rule of completeness may authorize admitting parts of a document that are otherwise inadmissible to explain or qualify other parts that have already been admitted.45 Thus, particularization is a strong tendency rather than an unwavering rule. In the courtroom, moreover, the particularized analysis of evidence law conflicts with the needs of litigants, lawyers, and judges to paint a picture embracing the evidence as a whole.46 A large amount of detail will not likely persuade a jury or judge, and trial lawyers are more likely to paint with a broad brush than to practice pointillism. Having analyzed and argued the admissibility of evidence item by item, lawyers must then present it to the trier of fact in a radically different way. In recent decades, much of the brilliant evidence scholarship sought to bridge the gap between bits of evidence and the trier of fact’s ultimate conclusions. In the real world, however, the gap may be too broad to cross. Students of Bayesian probability theory have considered how an ideal trier of fact could calculate how new pieces of evidence impact the overall strength of a party’s case.47 Unfortunately, no trier of fact can apply this analysis in a real case of any complexity, although Richard Friedman and others have shown that it can yield practical insights into evidence law and how this law might be improved.48 Starting from the other side of the gap, social scientists 43. FED. R. EVID. 403, 807; Old Chief v. United States, 519 U.S. 172, 185–86 (1997) (finding that when a prior felony offense is an element of a crime, a defendant’s willingness to stipulate to it ordinarily bars the prosecution from disclosing the name or nature of the prior offense). 44. STRONG ET AL., supra note 14, § 58. 45. FED. R. EVID. 106. See generally Dale A. Nance, A Theory of Verbal Completeness, 80 IOWA L. REV. 825 (1995); Dale A. Nance, Verbal Completeness and Exclusionary Rules Under the Federal Rules of Evidence, 75 TEX. L. REV. 51 (1996). 46. See generally, e.g., ANTHONY G. AMSTERDAM & JEROME BRUNER, MINDING THE LAW (2000); ROBERT P. BURNS, A THEORY OF THE TRIAL (1999); JANET MALCOLM, THE CRIME OF SHEILA MCGOUGH (1999); John Leubsdorf, The Structure of Judicial Opinions, 86 MINN. L. REV. 447 (2001). 47. See generally, e.g., PROBABILITY AND INFERENCE IN THE LAW OF EVIDENCE: THE USES AND LIMITS OF BAYESIANISM (Peter Tillers & Eric D. Green eds., 1988); Symposium, Decision and Influence in Litigation, 13 CARDOZO L. REV. 253 (1991). Wigmore’s complex-chart method might be considered a precursor of this approach. See generally TERENCE ANDERSON & WILLIAM TWINING, ANALYSIS OF EVIDENCE: HOW TO DO THINGS WITH FACTS BASED ON WIGMORE’S SCIENCE OF JUDICIAL PROOF (1991). 48. See generally, e.g., Richard Friedman, Route Analysis of Credibility and Hearsay, 96 YALE L.J. 667 (1987); Friedman, supra note 11. Bayesian analyses have been presented to juries to establish paternity, but it is not yet clear that this can be done in a way that is simple enough to understand without being misleading. See, e.g., State v. Skipper, 637 A.2d 1101, 1105–08 (Conn. 1994); State v. Spann, 617 A.2d 247, 257–59 (N.J. 1993). LEUBSDORF_FINAL 1218 7/19/2006 7:36:19 AM 91 IOWA LAW REVIEW [2006] and others have shown that triers of fact proceed holistically, framing narratives that embrace and organize evidentiary details.49 Whether the results of that procedure are reliable is another question. What is also unclear is how the insights-of-the-story approach should lead us to modify the details of a law of evidence, when that law is based on an entirely different perspective.50 B. SPLITTING THE ATOM: INFERENCES Having divided testimony into a series of questions and answers, evidence law proceeds to treat each inference separately, depending upon how that inference is based on a given answer. Under the standard approach, if one inference is permissible, the evidence should be admitted. This is true even though other inferences are improper. The admittance of evidence, however, remains subject to the opponent’s right to a limiting instruction and the court’s discretion to exclude evidence found to be more prejudicial than probative.51 Again, the point is not that this approach is always misguided. It is that choosing this mode of analysis is not inevitable. Such emphasis gives our evidence law some of its peculiar characteristics and has been done by assumption rather than after consideration. Separately considering inferences from the same testimony pervades the law of evidence and is perhaps the main art that its students must master. For example, that a defendant made a repair after an accident may not be used to show his previous negligence, but may nevertheless be introduced to show that the repair was feasible should this be controverted;52 and a transcript is inadmissible to show the contents of a recorded conversation that it transcribes, but is admissible to help the jury understand the recording itself. Similarly, in many jurisdictions, a trier who is given a “view” of the scene of an event is not supposed to consider as evidence what she sees with her own eyes, but is free to use it to understand witness testimony.53 The prosecution may not use a defendant’s past sales of stolen property to prove that he sold such property in the case before the court, 49. See generally W. LANCE BENNETT & MARTHA S. FELDMAN, RECONSTRUCTING REALITY IN COURTROOM: JUSTICE AND JUDGMENT IN AMERICAN CULTURE (1981); REID HASTIE ET AL., INSIDE THE JURY (1983); supra note 46 and accompanying text. 50. For some suggestions on demeanor, probability, and character evidence, see generally Ronald J. Allen & Brian Leiter, Naturalized Epistemology and the Law of Evidence, 87 VA. L. REV. 1491 (2001). For responses, see generally John Leubsdorf, Stories and Numbers, 13 CARDOZO L. REV. 455 (1991); Dale A. Nance, Naturalized Epistemology and the Critique of Evidence Theory, 87 VA. L. REV. 1551 (2001). 51. FED. R. EVID. 105, 403. 52. FED. R. EVID. 407. See generally Ross v. Black & Decker, Inc., 977 F.2d 1178 (7th Cir. 1992). 53. 2 STRONG ET AL., supra note 14, § 216, at 29. Compare United States v. Gray, 199 F.3d 547 (1st Cir. 1999) (rejecting rule), with State v. Mouzon, 485 S.E.2d 918 (S.C. 1997) (following rule). THE LEUBSDORF_FINAL PRESUPPOSITIONS OF EVIDENCE LAW 7/19/2006 7:36:19 AM 1219 but may use them to show that the defendant knew the present property was stolen.54 The hearsay rule gives rise to equally piquant distinctions. The jury may use a party’s statement against that party, but not against its employee.55 That a witness who now testifies to a defendant’s guilt formerly said he was innocent may not be used to show innocence, but may be used to impeach the witness’s testimony about guilt.56 The deceased’s statements that the defendant was going to kill her are not admissible to show that he did so,57 but may be admitted to show that the nature of their relationship was inconsistent with the defendant’s version of the facts.58 When a witness refreshes his recollection by examining a document in court, the jury should treat his testimony as resting only on his own credibility, not on that of the document.59 The separate appraisal of inferences gives rise to a recurring doctrinal pattern centering on a forbidden inference. Typically, a rule forbids the admission of certain evidence for one purpose, but allows it for others.60 Courts and lawyers then multiply the other uses, so that ultimately the evidence is almost always admissible.61 The main effect of the original prohibition is thus to multiply technicalities and penalize the clients of less skillful lawyers. The prohibition acts like a grain of sand around which the legal system secretes a pearl of doctrine, albeit a baroque and asymmetrical pearl. Inference-by-inference analysis is so ingrained among those trained in evidence law that, even when lawmakers enact what purports to be a relatively rigid exclusionary rule, courts tend to turn it into a prohibition limited to one forbidden inference. This may be happening with rape-shield laws and rules. Many of those laws and rules were designed to prohibit the admission of a complainant’s sexual history in sexual-assault cases except for 54. See generally Huddleston v. United States, 485 U.S. 681 (1988). 55. See generally Mahlandt v. Wild Canid Survival & Research Ctr., Inc., 588 F.2d 626 (8th Cir. 1978). 56. Federal Rule of Evidence 801(d)(1)(A) makes the former statement admissible to prove innocence, but only if it was given under oath at a “hearing, or other proceeding, or in a deposition.” FED. R. EVID. 801(d)(1)(A). 57. E.g., Commonwealth v. Levanduski, 2005 PA Super. 117 (excluding a written statement). 58. E.g., People v. Pinn, 94 Cal. Rptr. 741, 744–45 (Cal. Ct. App. 1971); State v. Alston, 461 S.E.2d 687, 703 (N.C. 1995). See generally United States v. Brown, 490 F.2d 758 (D.C. Cir. 1973). 59. See, e.g., United States v. Harris, 908 F.2d 728, 738 (11th Cir. 1990) (instructing the jury that a document used to refresh recollection is not evidence); United States v. Rinke, 778 F.2d 581, 588 (10th Cir. 1985) (finding testimony of a witness who read from his notes admissible as refreshed recollection). 60. See, e.g., FED. R. EVID. 404(a)–(b), 407–408; see also supra notes 52–59 (citing authority); infra notes 62–64 (same). 61. See generally EDWARD J. IMWINKELRIED, UNCHARGED MISCONDUCT EVIDENCE (rev. ed. 1999). LEUBSDORF_FINAL 1220 7/19/2006 7:36:19 AM 91 IOWA LAW REVIEW [2006] two or three carefully limited purposes, with an unavoidable exception when admissibility is constitutionally required.62 Relying on that exception, courts have allowed defendants to introduce sexual-history evidence for a variety of other purposes.63 In effect, they treat the rape-shield laws as forbidding the use of sexual history to show that a complainant was likely to consent to sex, but allowing its use when it has any other relevance, always subject to a limiting instruction and the usual balancing of prejudice versus probative value. Federal courts grounding this forbidden-inference approach in the Constitution can then impose it on state courts.64 The standard criticism of admitting evidence for one use but not another is that jurors will neither understand nor accept the restriction.65 Substantial empirical evidence supports this criticism, and in rare instances judges have adopted it by excluding the evidence.66 Usually they do not, 62. E.g., FED. R. EVID. 412(b)(1)(C); CONN. GEN. STAT. § 54-86f(4) (1958); FLA. STAT. § 794.022 (2005). 63. See, e.g., Redmond v. Kingston, 240 F.3d 590, 592–93 (7th Cir. 2001) (finding that a prior false sexual-assault charge bore on the complainant’s motives); United States v. Platero, 72 F.3d 806, 814–15 (10th Cir. 1995) (finding that the jury should be allowed to consider whether an extramarital affair gave the complainant motive to fabricate a rape claim to preserve a relationship with his lover); State v. DeJesus, 856 A.2d 345, 356 (Conn. 2004) (finding evidence that complainant was a prostitute was admissible to protect the defendant’s right to cross examination); People v. Slovinski, 420 N.W.2d 145, 153–54 (Mich. Ct. App. 1988) (finding evidence that complainant was a prostitute admissible to show that she consented to sex with defendant for money); State v. Lessley, 601 N.W.2d 521, 527–28 (Neb. 1999) (finding that prosecution’s reference to complainant as a lesbian made evidence of her sex with males admissible because of defendant’s constitutional right to confront witnessess); State v. Colbath, 540 A.2d 1212, 1216–17 (N.H. 1988) (Souter, J.) (finding complainant’s public advances to other men in the same tavern on the same evening admissible); see also Doe v. United States, 666 F.2d 43, 48 (4th Cir. 1981) (finding that the rule does not forbid showing defendant’s knowledge of complainant’s reported promiscuity to show defendant’s intent); Commonwealth v. Harris, 825 N.E.2d 58, 70 (Mass. 2005) (finding that a judge has discretion to allow use of complainant’s nightwalking conviction as impeachment); Chew v. State, 804 S.W.2d 633, 638 (Tex. Crim. App. 1991) (finding that the rule does not forbid showing complainant’s later sex with others to prove her “nymphomania”). There are, however, many other decisions excluding sexual-history evidence, sometimes in circumstances similar to those in some of these cases. 64. E.g., Olden v. Kentucky, 488 U.S. 227, 232–33 (1988); Redmond v. Kingston, 240 F.3d 590, 592–93 (7th Cir. 2001). The Supreme Court of Canada likewise struck down as unconstitutional a rape-shield statute without a catch-all exception. R. v. Seaboyer, [1991] 2 S.C.R. 577, 583. 65. E.g., United States v. Grunewald, 233 F.2d 556, 573–74 (2d Cir. 1956) (Frank, J., dissenting) (stating the criticism), rev’d, 353 U.S. 391 (1957). Sometimes, limiting the acceptable uses of evidence results in there not being enough evidence to go to the jury, in which case whether jurors would follow a limiting instruction becomes irrelevant. See, e.g., Allstate Ins. Co. v. Sprout, 782 F. Supp. 999, 1006 (M.D. Pa. 1991). 66. Bruton v. United States, 391 U.S. 123, 128–37 (1968) (relying on the Confrontation Clause); Shepard v. United States, 290 U.S. 96, 104 (1933); AMINA MEMON ET AL., PSYCHOLOGY AND LAW: TRUTHFULNESS, ACCURACY AND CREDIBILITY 162–64 (2d ed. 2003) (citing mixed evidence); Diamond & Vidmar, supra note 4, at 1863–65; Andrew J. Wistrich et al., Can Judges Ignore Inadmissible Information? The Difficulty of Deliberately Disregarding, 153 U. PA. L. REV. 1251, 1270–76 (2005) (citing mixed evidence); see also FED. R. EVID. 903(4) advisory committee’s note LEUBSDORF_FINAL 7/19/2006 7:36:19 AM PRESUPPOSITIONS OF EVIDENCE LAW 1221 either because they think that, on balance, admitting it is a good thing (and a good way to get around an exclusionary rule)67 or because recognizing the practical impossibility of using evidence for one purpose but not another would undermine the assumption of particularized analysis on which evidence law has been built. The law thus systematically ignores the way in which jurors, and for that matter judges, usually appraise evidence. Analyzing evidence inference by inference has a deeper defect to which evidence scholars have devoted little attention. That way of analyzing evidence would often be undesirable even if jurors were able to use it. When the goal is to appraise a witness’s credibility, pursuing each inference separately is often a distraction. It might be better to listen to the witness’s testimony as it evolves, using what psychoanalysts call “evenly hovering attention,” always on guard for the gap, inflection, or phraseology that gives a clue to what is really going on.68 And it would help to concentrate on just what the witness is saying about the transactions in dispute rather than to diverge into parsing the inferences that might be drawn from one smidgeon of evidence. Admittedly, jurors may not engage in inferential analysis while they are hearing the evidence, even when the judge instructs them at that time about how it may or may not be used. But even after they have heard all the witnesses and are considering their verdict, the analytic technique that instructions inculcate seems no more useful than possible to implement. In any event, other approaches are possible. Evidence law’s use of inference tracing results from a choice and is not an inevitable consequence of the nature of evidence or of human reasoning. C. THE TRANSACTION IN QUESTION Evidence law particularizes its analysis in yet a third way. It draws a number of distinctions between the acts directly before the court and those that may cast light on those acts. Some of these distinctions have more substantial justifications than the features of evidence law that have already (rejecting the common-law distinction limiting use of statements to physicians because distinction “was one most unlikely to be made by juries”). But see generally Kamala London & Narina Nunez, The Effect of Jury Deliberations on Jurors’ Propensity to Disregard Inadmissible Evidence, 85 J. APPLIED PSYCHOL. 932 (2000) (concluding that the jury deliberation removes the impact of evidence that jurors are told to disregard). Judges also exclude evidence as unduly prejudicial when an improper inference substantially outweighs the proper one. FED. R. EVID. 403; e.g., King v. Ahrens, 16 F.3d 265, 270 (8th Cir. 1994); United States v. Beasley, 809 F.2d 1273, 1278– 80 (7th Cir. 1987). 67. See, e.g., Richardson v. Marsh, 481 U.S. 200, 211 (1987); Nash v. United States, 54 F.2d 1006, 1006–07 (2d Cir. 1932) (L. Hand, J.). 68. Sigmund Freud, Recommendations for Physicians on the Psychoanalytic Method of Treatment, in THERAPY AND TECHNIQUE 117, 118 (Philip Rieff ed., 1963). But see Wellborn, supra note 10, at 1091–94 (citing evidence that demeanor evidence does not improve the performance of triers of fact). LEUBSDORF_FINAL 1222 7/19/2006 7:36:19 AM 91 IOWA LAW REVIEW [2006] been discussed. A rational procedural system must identify the facts in dispute and appraise the bearing that other facts may have on them. An efficient procedural system must impose some limits on the scope of the matters that it will consider in one litigation.69 But here too, evidence law does not simply respond to considerations of rationality and efficiency, but rather assumes without discussing that, absent special considerations, the narrowest focus is the best. Just because it is usually considered defunct, the res gestae doctrine furnishes an especially clear illustration of evidence law’s focus on the event in question. The common law admitted any relevant statement made in the course of the transaction in dispute under an exception to the hearsay rule.70 Whether a statement was admissible could hence turn on its time and place rather than on any functional reason to accept or reject it. Scholars have long criticized this preoccupation with the scope of the transaction, seeking to replace the res gestae hearsay exception with a family of narrower exceptions, each with its own rationale.71 The accepted wisdom is that they have succeeded.72 This may not be entirely true, however: a computer search reveals 186 uses of the phrase “res gestae” in reported opinions of courts in the United States in 2004 alone.73 Statutes in three states still embody the res gestae doctrine in modified form.74 Additionally, the justifications for the 69. See Mary Kay Kane, Original Sin and the Transaction in Federal Civil Procedure, 76 TEX. L. REV. 1723, 1735–47 (1998) (comparing Federal Rules of Civil Procedure that use terms such as “transaction” and “occurrence”). 70. E.g., Ins. Co. v. Mosley, 75 U.S. (8 Wall.) 397, 408 (1868); Lund v. Tyngsborough, 63 Mass. (9 Cush.) 36, 41–43 (1851); Teper v. Reginam, [1952] 2 All E.R. 447, 449 (P.C.) (appeal taken from Brit. Guiana) (U.K.); Friedman & McCormack, supra note 1, at 1212–18. 71. For classic critiques, see generally Edmund M. Morgan, A Suggested Classification of Utterances Admissible as Res Gestae, 31 YALE L.J. 229 (1922), James B. Thayer, Bedingfield’s Case—Declarations as a Part of the Res Gestae. I, 14 AM. L. REV. 817 (1880); James B. Thayer, Bedingfield’s Case—Declarations as a Part of the Res Gestae. II, 15 AM. L. REV. 1, 71 (1881). See also 6 WIGMORE, supra note 23, § 1767. 72. See FED. R. EVID. 801(d)(2)(D)–(E), 803(1)–(3), 804(b)(2) (stating exceptions that cover much evidence formerly considered res gestae, a term that does not appear in the Rules); Ratten v. Regina, [1971] 3 All E.R. 801, 805–08 (P.C.) (appeal taken from Vict., Austl.) (U.K.). 73. These results are based on a Lexis search conducted on July 12, 2005. A comparable search revealed forty-one Commonwealth cases using the phrase during the same year. Most of the United States cases do not deal with the traditional hearsay exception but with other issues, such as character-trait evidence. 74. GA. CODE ANN. § 24-3-3 (1995); LA. CODE EVID. ANN. art. 801D(4) (1995); MONT. CODE ANN. § 26-1-103 (2005); see, e.g., Roby v. State, 614 S.E.2d 916, 918 (Ga. Ct. App. 2005); State v. Lowery, 781 So. 2d 713, 728–29 (La. Ct. App. 2001); State v. Maier, 977 P.2d 298, 304– 05 (Mont. 1999). Res gestae, as a hearsay exception, may survive in two other states. See Ex parte Hunt, 744 So. 2d 851, 856 (Ala. 1999) (“We need not decide today whether, under our Rules of Evidence, Alabama recognizes a ‘res gestae exception’ to the hearsay rule . . . .”); Swanson v. State, 666 N.E.2d 397, 398–99 (Ind. 1996); see also TEX. CODE CRIM. PROC. ANN. art. 38.22(5) (Vernon 2005) (providing that admissibility of a defendant’s statement may turn on whether it was part of the res gestae). LEUBSDORF_FINAL PRESUPPOSITIONS OF EVIDENCE LAW 7/19/2006 7:36:19 AM 1223 “excited utterance” exception to the hearsay rule75 are so feeble as to permit the conclusion that it exists only because it was invented to justify retroactively decisions reached under the res gestae exception: excitement may or may not reduce the ability to fabricate, but it surely reduces the ability to observe, recollect, and narrate with accuracy.76 Just as it welcomes evidence about the transaction in dispute, evidence law resists evidence from outside it. Today, this resistance appears most clearly in the proliferation of disputes about evidence of “other crimes” or character traits. In principle, and subject to exceptions, a party may not introduce such evidence to show that someone has a propensity to behave in a certain way and therefore probably did so in the transaction before the court.77 Yet, jurors are never forbidden to draw just this kind of inference from a party’s behavior during the transaction in question—for example, by concluding that a defendant who brutalized a bystander probably also committed the murder for which he is standing trial. Lawyers routinely use such admissible behavior to paint a portrait of a party as someone likely or unlikely to commit the disputed acts. Whether behavior is part of the transaction in question can therefore be a decisive factor in determining its admissibility in evidence. Judges must distinguish “other crimes, wrongs, or acts,”78 subject to the exclusionary rule from parts of the same act. Not surprisingly, the phrase res gestae has reemerged to characterize the latter.79 That the law clings to the rules purporting to exclude other crimes and character evidence, despite their almost total negation in practice, indicates the grasp of particularization in evidence analysis on the legal mind. As already noted, the rules bar the “forbidden inference” from past behavior to present behavior, while allowing the use of past behavior for virtually any other purpose.80 Scholars, moreover, continue to cast doubt on whether the 75. FED. R. EVID. 803(2). 76. For criticism, see generally Aviva Orenstein, “MY GOD!”: A Feminist Critique of the Excited Utterance Exception to the Hearsay Rule, 85 CAL. L. REV. 159 (1997). 77. See generally FED. R. EVID. 404; IMWINKELRIED, supra note 61. For the recent and wavering emergence of this rule, see generally Thomas J. Reed, The Development of the Propensity Rule in Federal Criminal Causes 1840–1975, 51 U. CIN. L. REV. 299 (1982); Thomas J. Reed, Trial by Propensity: Admission of Other Criminal Acts Evidenced in Federal Criminal Trials, 50 U. CIN. L. REV. 713 (1981); Julius Stone, The Rule of Exclusion of Similar Fact Evidence: America, 51 HARV. L. REV. 988 (1938); Julius Stone, The Rule of Exclusion of Similar Fact Evidence: England, 46 HARV. L. REV. 954 (1933). 78. FED. R. EVID. 404(b). 79. See, e.g., United States v. Holliman, 291 F.3d 498, 502 (8th Cir. 2002); United States v. Hardy, 228 F.3d 745, 748–49 (6th Cir. 2000); see also United States v. Lane, 323 F.3d 568, 579–80 (7th Cir. 2003) (using similar analysis but different phrasing—“inextricably intertwined evidence”). For criticism of this terminology, see United States v. Ameri, 297 F. Supp. 2d 1168, 1168–73 (E.D. Ark. 2004). 80. See FED. R. EVID. 404(a)(1)–(3), 406, 608(b), 609(a); IMWINKELRIED, supra note 61, § 8:06; Kenneth J. Melilli, The Character Evidence Rule Revisited, 1998 BYU L. REV. 1547, 1549; supra notes 54, 60–61 and accompanying text. Congress recently added a new exception allowing use LEUBSDORF_FINAL 1224 7/19/2006 7:36:19 AM 91 IOWA LAW REVIEW [2006] prohibition can be justified by the unreliability of inferences from character, the alleged tendency of jurors to overvalue them, or the policy of treating each act separately.81 Yet, the prohibition survives. The same focus on the transaction in question appears in the somewhat artificial rules governing impeachment.82 A party may inquire about specific incidents showing a witness’s past honesty or dishonesty only at the discretion of the court, and even then only by cross-examination rather than through the use of extrinsic evidence.83 But if a witness behaved dishonestly during the transaction before the court, the jury is free to consider his dishonesty in appraising his credibility as a witness, regardless of whether the dishonesty was proved out of his own mouth or by extrinsic evidence. Somewhat similarly, parties were traditionally forbidden to use extrinsic evidence to impeach a witness by contradiction or a prior inconsistent statement on a “collateral” assertion. And an assertion ceased to be collateral if it tended to show something else about the transaction in dispute,84 even if its use is to show that something else was forbidden by the law of evidence. For example, a prosecution witness could be impeached with his prior inconsistent statement that the victim told him that his shooting was accidental, even though the statement was double hearsay that could not be used to show that the shooting was accidental.85 In effect, mere connection to the transaction in dispute outweighed both the collateral-dispute problem and the hearsay problem. Judicial attachment to this stress on the transaction before the court is such that some federal courts continue to apply the rules about extrinsic evidence on collateral matters,86 even though of a defendant’s previous sexual assault or child molestation to prove similar behavior in the case at bar. FED. R. EVID. 413–415. 81. See Sherry F. Colb, “Whodunit” Versus “What Was Done”: When to Admit Character Evidence in Criminal Cases, 79 N.C. L. REV. 939, 970–71 (2001) (arguing that, when the actor’s identity is undisputed, character evidence should be admissible to show what he did); Chris William Sanchirico, Character Evidence and the Object of Trial, 101 COLUM. L. REV. 1227, 1295–99 (2001) (basing rules on the encouragement of defendants to reform). See generally Symposium, Truth & Its Rivals: Evidence Reform and the Goals of Evidence Law: Character Evidence Revisited, 49 HASTINGS L.J. 663 (1998) (presenting various views); H. Richard Uviller, Evidence of Character to Prove Conduct: Illusion, Illogic, and Injustice in the Courtoom, 130 U. PA. L. REV. 845 (1982) (criticizing the rules). 82. The use of criminal convictions to impeach a witness’s credibility is the most prominent exception to the statement in the text. See FED. R. EVID. 609. 83. FED. R. EVID. 608(b). This rule was amended in 2003 to clarify that it covers only incidents relating to the witness’s character for truthfulness, as opposed to other matters such as bias that also bear on credibility. 84. See, e.g., State v. Oswalt, 381 P.2d 617, 619 (Wash. 1963); Kevin C. McMunigal & Calvin William Sharpe, Reforming Extrinsic Impeachment, 33 CONN. L. REV. 363, 385 (2001). 85. Smith v. State, 328 A.2d 274, 279 (Md. 1974); accord United States v. Budzanoski, 331 F. Supp. 1201, 1204 (W.D. Pa. 1971); Commonwealth v. Basch, 437 N.E.2d 200, 203 (Mass. 1982). 86. See, e.g., United States v. Mulinelli-Navas, 111 F.3d 983, 988 (1st Cir. 1997); United States v. Payne, 102 F.3d 289, 294 (7th Cir. 1996). LEUBSDORF_FINAL PRESUPPOSITIONS OF EVIDENCE LAW 7/19/2006 7:36:19 AM 1225 there are strong arguments that the adoption of the Federal Rules of Evidence abolished them.87 These are not the only instances in which clinging to the transaction in dispute seems to become an end in itself.88 Indeed, even the objections to the adequacy of statistical evidence to support a verdict are, in substantial part, grounded in the belief that parties should rely on evidence of what happened in a particular case.89 Courts often overcome the resistance to looking at evidence beyond the event before them. My point is that there is indeed resistance to overcome. Whatever the arguments for trying to restrict jurors and judges to evidence directly reflecting the event in question—and there certainly are such arguments90—doing so is radically inconsistent with the way human beings usually proceed. Most of us, if we were trying to decide whether an individual had committed an offensive act, would want to find out about the individual’s character and previous behavior. Despite the asserted problems of burden, prejudice, and confusion, other legal systems have found it practicable to inquire into such matters.91 Even in the United States, the individual’s character and prior behavior are brought before the court 87. 4 WEINSTEIN, supra note 39, § 607.06[3][b]; McMunigal & Sharpe, supra note 84, at 385–405. 88. For example, an individual’s past practice is inadmissible as evidence of the individual’s present behavior unless it constitutes a “habit.” See, e.g., Camfield v. City of Okla. City, 248 F.3d 1214, 1232–33 (10th Cir. 2001) (excluding evidence that a police officer seized the same video from five other video rental stores to show that the police officer did the same with plaintiff); Weil v. Seltzer, 873 F.2d 1453, 1460–61 (D.C. Cir. 1989) (excluding evidence that a doctor lied to five other patients by concealing the fact that he was prescribing them steroids to show that he lied in the same way to plaintiff); Reyes v. Mo. Pac. R.R., 589 F.2d 791, 794 (5th Cir. 1979) (excluding an individual’s four prior convictions for public intoxication as insufficient to show habit); Brett v. Berkowitz, 706 A.2d 509, 516–17 (Del. 1998) (excluding defendant lawyer’s prior alleged sexual relations with many other clients as insufficient to show a habit of sexual harassment). 89. See generally McCleskey v. Kemp, 481 U.S. 279 (1987); Guenther v. Armstrong Rubber Co., 406 F.2d 1315 (3d Cir. 1969); Smith v. Rapid Transit, Inc., 58 N.E.2d 754 (Mass. 1945); Nesson, supra note 20, at 1377–85. But see generally Howard v. Wal-Mart Stores, Inc., 160 F.3d 358 (7th Cir. 1998); Jonathan J. Koehler & Daniel N. Shaviro, Veridical Verdicts: Increasing Verdict Accuracy Through the Use of Overtly Probabilistic Evidence and Methods, 75 CORNELL L. REV. 247 (1990). 90. Among the classic opinions are Michelson v. United States, 335 U.S. 469, 475 (1948), in which Justice Jackson stated that “[t]he state may not show defendant’s prior trouble with the law . . . even though such facts might logically be persuasive that he is by propensity a probable perpetrator,” and People v. Zackowitz, 172 N.E. 466, 468 (N.Y. 1930), in which Justice Cardozo stated that “[f]undamental hitherto has been the rule that character is never an issue in a criminal prosecution unless the defendant chooses to make it one.” 91. See generally Adriaan Lanni, “Verdict Most Just”: The Modes of Classical Athenian Justice, 16 YALE J.L. & HUMAN. 277 (2004); Bron McKillop, Anatomy of a French Murder Case, 45 AM. J. COMP. L. 527 (1997). LEUBSDORF_FINAL 1226 7/19/2006 7:36:19 AM 91 IOWA LAW REVIEW [2006] during the sentencing process,92 and indeed during trial when one of the many exceptions to the exclusionary rules applies. Other kinds of background material, such as similar past occurrences and events demonstrating the character of witnesses, are likewise among the circumstances that an inquirer untainted by the law of evidence would often want to know. In the end, no rule of evidence can prevent triers of fact and advocates from thinking about such matters. Justice Souter’s opinion in Old Chief v. United States93 dramatically illustrates the collision between courtroom reality and the particularization of evidence law. The collision concerns not the Court’s holding that evidence should have been excluded as prejudicial, but its explanation of why prosecutors should usually be free to introduce evidence carrying a potential for prejudice despite the defendant’s offer to stipulate to the facts that the evidence is supposed to prove. Litigants, Justice Souter explained, may seek not just to introduce evidence of disputed facts but to “tell[] a colorful story with descriptive richness.”94 Only such a story will “sustain the willingness of jurors to draw the inferences” of guilt or innocence and “to implicate the law’s moral underpinnings and a juror’s obligation to sit in judgment” as well as meeting “jurors’ expectations about what proper proof should be.”95 This is a vivid evocation of the trial process, but is the Court really prepared to accept the implications of Justice Souter’s opinion for the law of evidence?96 Accepting these implications would allow an admission of Balzacian detail about the circumstances and background of the deeds in question. Acceptance would also allow portrayal of each participant’s psyche and morals, in defiance of the rules limiting character evidence. It would make the motivation of jurors a central feature of the law and recognize juror expectations as a ground for admitting evidence. Furthermore, a plaintiff in a civil action would then be allowed to present the whole story of a case in vivid detail even though the defendant’s answer had admitted many 92. E.g., U.S. SENTENCING GUIDELINES MANUAL §§ 4A1.1–4B1.5, 5H1.8 (2005), available at http://www.ussc.gov/2005guid/gl2005.pdf (discussing consideration of individual’s criminal history for sentencing); FED. R. CRIM. P. 32(c)(1)(A), 32(d)(2)(A)(i) (discussing the requirement of a presentence report that must include any prior criminal history); FED. R. EVID. 1101(d)(3); Eddings v. Oklahoma, 455 U.S. 104, 110–16 (1982) (discussing the need to consider defendant’s character in sentencing); Williams v. New York, 337 U.S. 241, 250–51 (1949) (stating that the sentencing judge is not limited to information heard in open court to make the sentencing decision). But see Kate Stith, Crime and Punishment Under the Constitution, 2004 SUP. CT. REV. 221, 225–26, 230–39, 261–65 (discussing the recent narrowing of Williams). 93. 519 U.S. 172 (1997). 94. Id. at 187. 95. Id. at 187–88. For comparable descriptions of trials, see supra note 46. 96. See generally D. Michael Risinger, John Henry Wigmore, Johnny Lynn Old Chief, and “Legitimate Moral Force”—Keeping the Courtroom Safe for Heartstrings and Gore, 49 HASTINGS L.J. 403 (1998) (criticizing Old Chief’s reasoning). LEUBSDORF_FINAL 7/19/2006 7:36:19 AM PRESUPPOSITIONS OF EVIDENCE LAW 1227 of the facts to be proven. The extent of these changes vividly demonstrates the extent to which the present law of evidence is built on the particularizing assumptions discussed above. D. AN OPPOSITE REACTION: GENERALIZATIONS After classifying testimony into the smallest possible particles, evidence law proceeds to judge testimony under rules based on broad and untenable generalizations about classes of evidence. The two phenomena are related. Given that each particle of evidence is to be appraised separately, courts would find it hard to proceed at all without a set of rules that is, in theory, easily applied as issues arise in the midst of trials. The system is something like a machine—perhaps a Rube Goldberg contraption97—meant to sort quickly a large volume of diverse objects. This intention may explain the use of generalizations but does not necessarily justify it. On the contrary, if bits of evidence are to be appraised with rules that are at the same time crude and complex, why should one bother to dissect evidence into bits to begin with? Charles McCormick explained the drawbacks of categorization, attributing it to a technique of eighteenth-century judges: That technique consisted of creating large, simple, but definite categories under which offered items of proof could be classified accurately and, above all, quickly. All the contents of each of these classes were either black or white, admissible or inadmissible. The largest of these categories of inadmissible evidence (though its recognition as such was later than we usually suppose) is that of hearsay. The advantages of these clear-cut rules of exclusion are obvious. They enable the lawyer preparing his case to know in advance with fair certainty what he can get in, and what he cannot. If a question as to admissibility does arise, the judge who has no time for subtle discrimination in the heat of trial can make a decision in his stride, as it were. This is splendid, and the only difficulty is that it does not work. The rule excluding all hearsay, clear and simple in its original form, when it was tested by the offer of particular hearsay evidence of a peculiarly indispensable or reliable kind cracked under the strain.98 Rules of the sort that McCormick described will be both overbroad and overnarrow, and may be no better than alternative rules based on alternative generalizations. In the case of hearsay, the generalization grounding the exclusionary rule is that statements made out of court without the 97. See generally The Official Rube Goldberg Website, http://www.rube-goldberg.com (last visited Feb. 25, 2005). 98. Charles T. McCormick, The Borderland of Hearsay, 39 YALE L.J. 489, 503 (1930) (footnote omitted). See generally Swift, supra note 32. LEUBSDORF_FINAL 1228 7/19/2006 7:36:19 AM 91 IOWA LAW REVIEW [2006] safeguards of oath and cross-examination are less reliable than those made in courtroom testimony. But one could say, with at least equal truth, that statements made close in time to the events that they describe are more likely to be accurate than those made later. A rule based on that generalization would find most hearsay better evidence than trial testimony.99 Or one could say that statements made by disinterested speakers are more reliable than those of interested speakers. This might lead to a rule admitting all statements, in- or out-of-court, by disinterested witnesses, while excluding (as was formerly done)100 statements by interested witnesses, with the exception of statements going against the speaker’s interests. The hearsay exceptions are likewise based on more or less dubious generalizations, carving out of the class of hearsay statements subclasses deemed to be either as reliable as courtroom testimony or at least the best available. Among the generalizations are some whose defects are familiar to all students of the law of evidence: excited people tell the truth;101 even children do not lie to doctors;102 employees making records do not deceive;103 imminent death ensures reliability;104 and so on. Because these generalizations have a narrower scope than the one condemning hearsay, some of them may correspond better to reality. But they are still crude because they fail to take into account many of the factors that affect the value of testimony. For example, although a statement may become admissible if it was against the interests of its speaker,105 that a statement promoted its speaker’s interest is not a ground for exclusion.106 The quality of a speaker’s eyesight, hearing, or memory likewise does not, except in 99. French courts hearing civil cases apply an approach much like this one. See infra Part III.B. Even United States courts have sometimes relied on a similar rationale to support hearsay exceptions. United States v. Owens, 484 U.S. 554, 562–63 (1988) (allowing identification evidence); United States v. Inadi, 475 U.S. 387, 395 (1986) (allowing co-conspirator statements); see FED. R. EVID. 803(1), 803(5) (allowing hearsay exceptions for present-sense impressions and recorded recollections). 100. See supra note 23 and accompanying text. 101. FED. R. EVID. 803(2). For a criticism of the excited-utterance rule, see generally Orenstein, supra note 76. 102. FED. R. EVID. 803(4); United States v. Norman T., 129 F.3d 1099, 1106 (10th Cir. 1997). But see United States v. Sumner, 204 F.3d 1182, 1185 (8th Cir. 2000) (finding that a child must be shown to understand the importance of being truthful). See generally Robert P. Mosteller, The Maturation and Disintegration of the Hearsay Exception for Statements for Medical Examination in Child Sexual Abuse Cases, 65 LAW & CONTEMP. PROBS. 47 (2002); John E.B. Myers et al., Hearsay Exceptions: Adjusting the Ratio of Intuition to Psychological Science, 65 LAW & CONTEMP. PROBS. 3, 8–19 (2002). 103. Cf. FED. R. EVID. 803(6). See generally Lewis v. Baker, 526 F.2d 470 (2d Cir. 1975) (finding that a railroad could introduce an accident report prepared by its trainmaster). 104. Cf. FED. R. EVID. 804(b)(2). 105. Cf. FED. R. EVID. 804(b)(3). 106. But see FED. R. EVID. 803(6) (noting that a business record may be excluded when “circumstances of preparation indicate lack of trustworthiness”). LEUBSDORF_FINAL PRESUPPOSITIONS OF EVIDENCE LAW 7/19/2006 7:36:19 AM 1229 extraordinary circumstances,107 affect whether his statement is admissible under a hearsay exception. Such matters would be said to bear on weight rather than admissibility. The Hillmon paradox of forward- and backward-looking statements strikingly demonstrates the perils of categorization and might be considered a reductio ad absurdum of a hearsay rule based on it. Hillmon allows a statement made out of court to be used to prove the declarant’s intent to support the inference that he acted on that intent.108 The statement is said to be reliable—and hence either nonhearsay or admissible under a hearsay exception—because a statement about one’s own present state of mind involves little danger of misperception or poor memory.109 The paradox is that statements about what the declarant has already done are, if anything, stronger evidence that he has done it than statements about what he intended to do are that he later did it; yet, the latter are admissible, and the former are not.110 The justifications that have been advanced for excluding backwardlooking statements, while admitting forward-looking statements, are unpersuasive. Justice Cardozo famously observed that “[t]here would be an end, or nearly that, to the rule against hearsay if the distinction were ignored.”111 This is correct and may be a good argument from authority, but it provides no policy support. If forward-looking statements made out of court are good evidence, and backward-looking ones are better, why not admit the latter and let the hearsay rule perish? A more subtle argument is that the dangers of backward-looking statements include poor memory and perception, which are dangers against which the hearsay rule is directed, while the danger of forward-looking statements is that the declarant may not consummate his intent, which is not a hearsay danger.112 That argument 107. Poor vision or hearing might affect whether a declarant “was perceiving the event or condition” as required by the present-sense-impression exception. FED. R. EVID. 803(1). A declarant’s vacant memory would deprive her of the “knowledge” needed to create a recorded recollection. FED. R. EVID. 803(5). A declarant whose capacities were so feeble as to leave him without the “personal knowledge” required for witnesses by Federal Rule of Evidence 602 might also be unable to make an admissible hearsay statement. 108. Mutual Life Ins. Co. v. Hillmon, 145 U.S. 285, 295–98 (1892). 109. FED. R. EVID. 803(3); see STRONG ET AL., supra note 14, §§ 274–275. 110. See Capano v. State, 781 A.2d 556, 602–08 (Del. 2001); cf. FED. R. EVID. 803(3). This does not refer to the second Hillmon problem—i.e., use of a statement that the declarant intends to do something together with someone else to show not just what the declarant did, but what the other person did. See generally United States v. Best, 219 F.3d 192 (2d Cir. 2000); People v. James, 717 N.E.2d 1052 (N.Y. 1999). 111. Shepard v. United States, 290 U.S. 96, 106 (1933); see FED. R. EVID. 803(3) advisory committee’s note (speaking of “virtual destruction of the hearsay rule”). 112. See John M. Maguire, The Hillmon Case—Thirty-Three Years After, 38 HARV. L. REV. 709, 727 (1925). Of course, one can make up claims about juror psychology that would justify the distinction, for example, that jurors will appreciate the weaknesses of forward-looking hearsay LEUBSDORF_FINAL 1230 7/19/2006 7:36:19 AM 91 IOWA LAW REVIEW [2006] explains the internal logic of the hearsay rule, but again provides no justification for barring the use of backward-looking statements to prove things remembered. Generalizing about classes of statements has tangled evidence law in this paradox. The cases giving rise to the Hillmon rule, including Hillmon itself, involved reliable forward-looking statements, usually given by declarants unavailable to testify under oath and under cross-examination.113 But when Hillmon became the Hillmon rule, far less useful statements became admissible because they could be classified as evidence of the declarant’s intent.114 Likewise, the class of backward-looking statements contains statements of varying value. Whether one of these classes is, on the whole, better evidence than the other is impossible to say. Unfortunately, the hearsay rule and its exceptions are based on precisely that kind of generalization. One can easily create more paradoxes like the Hillmon paradox. If a party may introduce into evidence its own self-serving business records,115 should not any document prepared by a disinterested nonparty be admissible? If a statement to the police by a crime victim that “X did it” is admissible as an identification, provided that the victim testifies at the trial,116 why should not all out-of-court statements by persons who later testify describing any aspect of the crime be admissible?117 Each of these paradoxes is produced by taking one of the more questionable examples of a class of hearsay statements admissible under a plausible exception and comparing that example to a class of inadmissible hearsay statements identified by a feature favoring admission. The paradoxes show that hearsay law regularly better than those of backward-looking hearsay. That claim seems no more plausible than its opposite. 113. See Milich, supra note 2, at 736–38. But see generally Marianne Wesson, State of Mind: The Hillmon Case, the McGuffin, and the Supreme Court (Univ. of Colo. Law Sch. Legal Studies Research Paper Series, Working Paper No. 06-09, 2006), available at http://ssrn.com/abstract=813307 (arguing that the Hillmon letter was forged). 114. Compare United States v. Torres, 901 F.2d 205, 239 (2d Cir. 1990) (finding reliability irrelevant to admissibility), and United States v. Badalamenti, 794 F.2d 821, 826 (2d Cir. 1986) (discussing an informant’s statement to a government agent), with R. v. Starr, [2000] 2 S.C.R. 144, 147 (limiting the exception to statements with “indicia of reliability”). 115. FED. R. EVID. 803(6); see also Sea-Land Serv., Inc. v. Lozen Int’l, LLC, 285 F.3d 808, 819 (9th Cir. 2002). 116. FED. R. EVID. 801(d)(1)(C); United States v. Lopez, 271 F.3d 472, 485 (2d Cir. 2001); United States v. Davis, 181 F.3d 147, 149 (D.C. Cir. 1999); United States v. Brink, 39 F.3d 419, 426 (3d Cir. 1994). 117. If anything, statements not made to the police should be considered more reliable than those that are. See Crawford v. Washington, 541 U.S. 36, 61–62 (2004) (finding “testimonial” statements to the police barred by the Confrontation Clause, absent crossexamination or past cross-examination plus proof of unavailability); Miranda v. Arizona, 384 U.S. 436, 446–54 (1966) (describing dangers of police-interrogation techniques). LEUBSDORF_FINAL PRESUPPOSITIONS OF EVIDENCE LAW 7/19/2006 7:36:19 AM 1231 admits into evidence material more questionable than what it excludes because it proceeds by categorization. Evidence law not only relies on generalizations in establishing the hearsay rule and defining its exceptions, but adds a third and still stranger level of generalization: the rule that double or multiple hearsay is admissible if each level of hearsay falls within an exception.118 For example, a witness may testify that A said that B had just told A on the telephone what C said, if A’s statement qualifies as a present sense impression, B’s as a statement by an employee about a matter within the scope of employment, and C’s as a co-conspirator’s statement advancing the conspiracy.119 It is pure fiction to say that whether or not such testimony is good evidence can be evaluated by adding up exceptions. Not only do hearsay exceptions vary in their plausibility,120 but each covers a variety of statements that vary widely in their evidentiary value. In addition, three levels of hearsay might well be more than three times worse than one level, as the credibility of the original declarant becomes further removed from appraisal, or perhaps in some circumstances not much different from two levels. Reckoning on the assumption that each level of hearsay equals each other level, each exception equals each other exception, and each statement equals each other statement in the same exception, looks like a manic parody of efforts to apply mathematics to the evaluation of evidence.121 The lawmakers have not heeded Aristotle’s call to search only “for that degree of precision in each kind of study which the nature of the subject at hand admits.”122 In practice, the effects of categorization on the admission of hearsay are diluted by the catch-all provision allowing judges to admit hearsay not falling 118. FED. R. EVID. 805. See generally Neill v. N. Antrim Magistrates’ Court, [1992] 1 W.L.R. 1220 (H.L.) (appeal taken from N. Ir.) (U.K.). But see Evidence Act, 1995, §§ 62–68 (Austl.) (creating special exceptions for single hearsay); THE LAW COMM’N, EVIDENCE IN CRIMINAL PROCEEDINGS: HEARSAY AND RELATED TOPICS 97–100 (1997) (Law Comm’n No. 245) (recommending that double hearsay should ordinarily be inadmissible). 119. United States v. Portsmouth Paving Corp., 694 F.2d 312, 320–23 (4th Cir. 1982). 120. For example, the case described in the text involves both the present-sense-impression exception, which has strong foundations, and the co-conspirator exception, which has little to do with reliability and is based mainly on the need to secure evidence against alleged criminals. See United States v. Goldberg, 105 F.3d 770, 775 (1st Cir. 1997) (stating that “frankly, the underlying co-conspirator exception to the hearsay rule makes little sense as a matter of evidence policy”); United States v. DiDomenico, 78 F.3d 294, 303 (7th Cir. 1996) (questioning “the justification for the [conspirator exception to the hearsay] rule”); Joseph H. Levie, Hearsay and Conspiracy: A Reexamination of the Co-Conspirators’ Exception to the Hearsay Rule, 52 MICH. L. REV. 1159, 1159–60 (1954); Christopher B. Mueller, The Federal Coconspirator Exception: Action, Assertion, and Hearsay, 12 HOFSTRA L. REV. 323, 355 (1984). 121. See supra notes 47–48 and accompanying text (discussing works on Bayesian probability theory and evidence law). 122. ARISTOTLE, NICOMACHEAN ETHICS 1094b (Martin Ostwald trans., 1962). LEUBSDORF_FINAL 1232 7/19/2006 7:36:19 AM 91 IOWA LAW REVIEW [2006] within another exception in certain circumstances.123 But introducing evidence under this limited exception has not been frequent.124 The catchall provision of the Federal Rules of Evidence works only in one way. It allows a judge to admit otherwise inadmissible evidence on the basis of a particularized analysis of reliability and necessity, but—by contrast to the common law of Canada—not to exclude evidence otherwise admissible.125 And the great bulk of evidence challenged as hearsay is admitted or excluded on the basis of rules and exceptions based on generalizations about large classes of evidence. Although the use of generalizations may be most glaring in the law of hearsay, it also occurs elsewhere in evidence law. To determine when impeaching a witness’s credibility must yield to fears of distraction and wasted time, the law classifies impeachment evidence as intrinsic or extrinsic, with the latter divided into further categories: impeachment by contradiction; prior inconsistent statement; reputation for untruthfulness; and acts indicating untruthfulness, prior criminal convictions of various sorts, bias, and so forth. For each category of evidence, there is a rule specifying when it may be used.126 To the extent that these rules have justifications, they are based on the theory that the value or drawbacks of each bit of impeaching evidence can be specified in large part by identifying the category into which it falls. The attorney-client privilege likewise deploys a squadron of sub-rules in an effort to divide the instances in which evidence should be excluded to encourage the use of lawyers. Contrast this with those situations in which that goal yields to the desirability of obtaining evidence.127 This privilege law is far more intricate than work-product law, which deals with a somewhat similar problem but relies less on categorization than on ad-hoc 123. FED. R. EVID. 807. 124. 5 WEINSTEIN, supra note 39, § 807.02[1]. Using the catch-all exception against a criminal defendant may also involve Confrontation Clause difficulties. See Lilly v. Virginia, 527 U.S. 116, 123–24 (1999) (discussing Confrontation Clause issues regarding hearsay). Some states, including California and New York, do not recognize the catch-all exception. Idaho v. Wright, 497 U.S. 805, 816 (1990) (requiring “‘particularized guarantees of trustworthiness’” (quoting Ohio v. Roberts, 448 U.S. 56, 66 (1980))). 125. Compare R. v. Smith, [1992] 2 S.C.R. 915, 917 (admitting hearsay not within other exceptions), with R. v. Starr, [2000] 2 S.C.R. 144, 148 (excluding hearsay previously thought to be covered by exception). While judges in the United States can sometimes exclude otherwise admissible hearsay on the ground of prejudice, mere unreliability is not a ground for exclusion. 126. See generally EDWARD J. IMWINKELRIED, EVIDENTIARY FOUNDATIONS ch. 5 (5th ed. 2002); CHRISTOPHER B. MUELLER & LAIRD C. KIRKPATRICK, EVIDENCE §§ 6.21, 6.27, 6.28, 6.40, 6.47, 8.24 (2d ed. 1999); McMunigal & Sharpe, supra note 84. Some, but not all, of the rules may be found in Federal Rules of Evidence 608, 609, 613(b), and 801(d)(1)(A). 127. See generally PAUL R. RICE, ATTORNEY–CLIENT PRIVILEGE IN THE UNITED STATES (2d ed. 1999). LEUBSDORF_FINAL PRESUPPOSITIONS OF EVIDENCE LAW 7/19/2006 7:36:19 AM 1233 balancing.128 In principle, a system of precise rules enables people, or at least lawyers, to know the law governing them and promotes uniform court decisions. These goals are especially important in the law of privilege, which is more likely to encourage communications to lawyers and others if all concerned know in advance that what they say will not be admissible in evidence. The problem, as McCormick pointed out, is that the systems do not work because they rely on flimsy generalizations about the evidence assigned to a given category.129 For example, the rule that the privilege survives the client’s death (except as to will disputes and the like) may sometimes reflect the real needs of the client to have his disclosures remain secret,130 while in other instances it unnecessarily frustrates the discovery of the truth.131 Do these categorizations distinguish evidence law from other kinds of law? All law relies on rules, and almost all rules rest on justifications whose scope does not match precisely the scope of the rules based on them. Evidence scholars have long argued that evidence law uses too many rules and too little ad-hoc discretionary weighing, and have succeeded in increasing the scope of judicial discretion.132 There is certainly room to dispute whether the resulting system succeeds in reconciling sound results in specific cases with some measure of predictability and consistency, or on the contrary, requires lawyers to jump through intellectual hoops in order to argue for or against admissibility, while leaving judges free to decide more or less as they wish. In any event, the use of weak generalizations is just half the story: evidence law resorts to generalizations only after fractionating evidence witness by witness, answer by answer, and inference by inference. Both trends implement a misguided search for rigor, whether in slicing the 128. Compare RESTATEMENT OF THE LAW GOVERNING LAWYERS §§ 68–85 (2000) (attorneyclient privilege), with id. §§ 87–93 (attorney work product). Similarly, compare CAL. EVID. CODE §§ 950–962 (West 1995) (attorney-client privilege), with CAL. CIV. PROC. CODE § 2018 (West 1998) (attorney work product). 129. See supra note 98 and accompanying text. In the case of privilege, many of the rules do not, in any event, provide predictability because they are found in precedents that leave many issues open. Davis v. Alaska, 415 U.S. 308, 315–21 (1974) (discussing when the Confrontation Clause overrides privileges). See generally Jaffee v. Redmond, 518 U.S. 1 (1996); Upjohn Co. v. United States, 449 U.S. 383 (1981). 130. See Swidler & Berlin v. United States, 524 U.S. 399, 410 (1998) (protecting confidences of a client who committed suicide after consulting lawyer). 131. See State v. Macumber, 544 P.2d 1084, 1086 (Ariz. 1976) (finding that a prosecutor could prevent a lawyer from testifying that a dead client confessed to the crime for which another was being tried); see also RESTATEMENT OF THE LAW GOVERNING LAWYERS § 77 cmt. d & reporter’s note. 132. See, e.g., WILLIAM TWINING, THEORIES OF EVIDENCE: BENTHAM & WIGMORE 66–75 (1985); Michael Ariens, Progress Is Our Only Product: Legal Reform and the Codification of Evidence, 17 LAW & SOC. INQUIRY 213 (1992). See generally Swift, supra note 21; Jack B. Weinstein, Probative Force of Hearsay, 46 IOWA L. REV. 331 (1961). LEUBSDORF_FINAL 1234 7/19/2006 7:36:19 AM 91 IOWA LAW REVIEW [2006] evidence too finely, or in trying to devise rules precise enough to be applied mechanically and complex enough to do justice to the variety of evidentiary problems. Extreme particularization results in a mountain of evidentiary fragments that calls for a mechanical sorting system so that it can be handled under the time pressure of trials. At the same time, because each fragment is viewed in isolation, the sorting system necessarily relies on circumstances— the particular inference drawn from a particular assertion—that do not embrace the full bearing and value of that fragment. The system must therefore use rules that are similarly constricted. The result is evidence law that is simultaneously shortsighted and blurred, because it looks both too narrowly and too broadly. III. MYTHS OF PRESENCE: VOUCHING FOR NONTESTIMONIAL EVIDENCE Evidence law presupposes that virtually all evidence that is not itself courtroom testimony by a live witness will be escorted into court by such a witness. This presupposition is related to, but different from, the hearsay rule: it applies regardless of whether the evidence in question is hearsay. This presupposition differs radically from normal human practice outside the courtroom, which nowadays bases beliefs about past events on all sorts of materials without demanding a living witness to vouch for them. The policy arguments for such a demand are more than dubious, and even resorting to history fails to provide an adequate explanation for it, much less a justification. Rather, the escort requirement seems to embody a privileging of live testimony that structures evidence law without itself emerging into the light of examination and critique. No doubt its impact on evidence law is far less massive than that of the techniques of atomization already discussed, but like them, it is part of the deep structure of the subject. A. THE TESTIFYING ESCORT The requirement that a live witness escort all evidence other than that of another live witness should not be confused with a belief that live testimony is more probative than other evidence such as documents. AngloAmerican evidence law leaves the weighing of evidence to the trier of fact without any intimations that the trier should give greater weight to any one kind of evidence.133 And if any belief that oral evidence is superior to documents had ever existed,134 it had disappeared by the early eighteenth century, before most of our present evidence law had developed, when the first English evidence treatise declared that documents are better evidence 133. See Desert Palace, Inc. v. Costa, 539 U.S. 90, 92 (2003) (rejecting any distinction in probative value between direct and circumstantial evidence). 134. See infra Part III.C (discussing this historical thesis). LEUBSDORF_FINAL PRESUPPOSITIONS OF EVIDENCE LAW 7/19/2006 7:36:19 AM 1235 than testimony.135 Today, trial lawyers still assert that “a judge or jury will often believe the barest scrap of paper over the memory of any witness.”136 The primacy of live testimony in evidence law is not a question of probative value but rather of what might almost be called epistemology. Nontestimonial evidence may be of great value, but the jury may perceive it only if it is, as it were, incorporated by reference in live testimony. Like a character in the social world portrayed by Proust, it must be introduced to the jurors by someone they already know before they will acknowledge its presence. The most obvious example of this requirement is the demand for authentication of documents and other tangible things introduced into evidence. Each such thing must be warranted by “evidence sufficient to support a finding that the matter in question is what its proponent claims.”137 The decisive point for our purposes is not what must be shown to authenticate—handwriting analysis, for example, usually suffices despite its feeble credentials138—but how it must be shown. Authentication must be by “evidence”: a live witness or a document that has in turn been authenticated, or the admission of the party against which the evidence is admitted.139 If authenticating evidence is lacking, the document or thing must be excluded. The Federal Rules do modify previous law by recognizing an exception for a few kinds of evidence that are usually referred to as selfauthenticating.140 That description is inaccurate in a significant way: almost 135. BARON GEOFFREY GILBERT, THE LAW OF EVIDENCE 4–5 (1st ed. 1754); accord Evans, supra note 26, at 115 (“Written evidence is superior to verbal, as it is by no means equally liable to misconception, or misrepresentation.”); see ALEXANDER WELSH, STRONG REPRESENTATIONS: NARRATIVE AND CIRCUMSTANTIAL EVIDENCE IN ENGLAND, at ix (1992) (describing eighteenthcentury praise of circumstantial evidence as preferable to testimony). For a discussion of the continental view, see LORRAINE DASTON, CLASSICAL PROBABILITY IN THE ENLIGHTENMENT 320– 21 (1988); Tony Honoré, The Primacy of Oral Evidence?, in CRIME, PROOF AND PUNISHMENT: ESSAYS IN MEMORY OF SIR RUPERT CROSS 172–73 (1981). But see 1 THOMAS STARKIE, A PRACTICAL TREATISE OF THE LAW OF EVIDENCE 108–09 (1824) (preferring oral to written evidence); Fisher, supra note 25, at 643–44 (concluding that eighteenth-century juries found circumstantial evidence less persuasive than sworn eyewitnesses). 136. Elliot G. Sagor, Victory Is in the Documents, LITIG., Winter 1997, at 36, 36. 137. FED. R. EVID. 901(a); see 2 STRONG ET AL., supra note 14, §§ 218–228; 7 WIGMORE, supra note 23, § 2129. The authentication requirement is not limited to tangible things but includes testimony about voices heard by telephone. FED. R. EVID. 901(b)(5)–(6). 138. D. Michael Risinger & Michael J. Saks, Science and Nonscience in the Courts: Daubert Meets Handwriting Identification Expertise, 82 IOWA L. REV. 21, 21–22 (1996); see TAMARA PLAKINS THORNTON, HANDWRITING IN AMERICA: A CULTURAL HISTORY 88–92, 101–05 (1996) (comparing the early nineteenth-century view that someone who had seen the author write could authenticate his handwriting with later reliance on comparison by jurors and experts). 139. See FED. R. CIV. P. 36 (requiring notice to admit genuineness of document); In re Japanese Elec. Prods. Antitrust Litig., 723 F.2d 238, 285 (3d Cir. 1983) (holding that authentication must be by admissible evidence), rev’d on other grounds, 475 U.S. 574 (1986). 140. FED. R. EVID. 902; see also FED. R. EVID. 901(b)(3) (allowing authentication by comparison with document already authenticated). LEUBSDORF_FINAL 1236 7/19/2006 7:36:19 AM 91 IOWA LAW REVIEW [2006] all the “self-authenticating” documents do have to be warranted, typically by another document or in some cases a chain of documents.141 The point is that they do not have to be warranted by a live witness, and it is the need for that warrant that lies at the heart of the traditional authentication requirement. The development of the law of photographic evidence shows just how far evidence law’s insistence on a live witness escort has extended. As Jennifer Mnookin recounts, for many decades courts received photographs in evidence on the theory that they merely illustrated a witness’s testimony.142 In Wigmore’s words, a photograph “is, for evidential purposes simply nothing, except so far as it has a human being’s credit to support it. It is mere waste paper—[a] testimonial nonentity.”143 In reality, of course, any juror would give far more weight to a witness’s description backed by a photograph than to the description alone. Yet, it was not until the rise of automatic cameras in banks and stores that courts became willing to admit photographs without testimony that the scene in the photograph accurately recorded what the witness had seen.144 Even then, an authenticating witness was required, as is true today.145 The only change is that the witness need not be able to say that the photograph records what she saw, but may instead testify that the photograph was recorded and preserved in such a way as to produce an accurate image of whatever was before the camera. The principle of evidentiary escort reaches its weird extreme in the rule, still followed in many jurisdictions, that a judge’s or jury’s view of the scene of relevant events is not itself evidence and may be used only to help the jury understand the testimony of witnesses.146 The appraisal by a trier of fact of what she sees at the scene in question is usually at least as valuable as her appraisal of the testimony of a witness about what he saw there. Yet the law in many jurisdictions accepts the testimony and rejects—or pretends to reject—the trier’s own observations. The argument that an appellate court reviewing the record will not be able to take into account what the trier has seen can be met by taking photographs at the time of the view, which will give the court at least as good an idea of what the trier saw as a transcript gives of a witness’s testimony. Many courts accept these arguments and 141. E.g., FED. R. EVID. 902(3) (allowing attestation by foreign official, whose execution is in turn verified); FED. R. EVID. 902(11) (allowing written declaration authenticating business record). But some documents, such as newspapers and labels, are indeed self-authenticating. FED. R. EVID. 902(6)–(7). 142. Jennifer L. Mnookin, The Image of Truth: Photographic Evidence and the Power of Analogy, 10 YALE J.L. & HUMAN. 1, 43–59 (1998). 143. 3 WIGMORE, supra note 23, § 790 (emphasis omitted). 144. See State v. Pulphus, 465 A.2d 153, 157–58 (R.I. 1983) (citing authority). 145. MUELLER & KIRKPATRICK, supra note 126, § 9.14. 146. See supra note 53 (citing authority). LEUBSDORF_FINAL 7/19/2006 7:36:19 AM PRESUPPOSITIONS OF EVIDENCE LAW 1237 accept views as evidence.147 That many still do not shows the strength of the primacy that live testimony retains. Although the main effect of the escort principle is to exclude evidence when no witness endorses it, courts also use the principle to expand the scope of what the jury is, in fact, likely to consider. When a photograph is admitted into evidence on the theory that it merely illustrates a witness’s testimony, the jury is, in reality, free to give it greater weight than that of its sponsor and scan it for details that the sponsor might not remember. When a jury is duly instructed that a view is not evidence, it will surely consider it anyway, and an appellate court may well rely on the instruction to reject a claim that the judge erred by allowing the view.148 Another example falls outside the subject of authentication: when a witness refreshes his recollection by consulting a document that is not in evidence, a jury that has seen the witness read the document before testifying may well take this into account as increasing the weight of his testimony.149 In each of these examples, the theory that the jury considers only that which is warranted by a witness is used as a Trojan horse to smuggle into the jury room material that is not so warranted. Even though the witness-escort principle bears on a fraction of today’s law of evidence, it still has fundamental importance because it helps define trials as basically oral events during which any non-oral evidence requires special oral support. This definition is so deeply rooted in our conception of trials as to escape almost all conscious appraisal. Were it absent, lawyers and judges would be more likely to treat live testimony, documents, photographs, views, and other modes of proof simply as alternatives of generally equal status, to be favored or disfavored in light of each individual item’s strengths and weaknesses taking into account the needs of the case. B. RATIONALIZATIONS Whether the witness-escort principle makes sense is irrelevant as to whether it underlies much of evidence law; the less sense the principle makes, the more striking is its prevalence. Examination of five possible rationales for the witness-escort principle suggests that it makes very little sense. Ultimately, perhaps the best that can be said for it is that it is harmless 147. E.g., Barron v. United States, 818 A.2d 987, 991–92 (D.C. 2003) (accepting a jury view as evidence as long as both parties were given an opportunity to present arguments relating to the evidence); State v. Pauline, 60 P.3d 306, 325 (Haw. 2002) (arguing that not admitting into evidence large objects that could only be viewed outside of the courtroom would be unjust). 148. See generally State v. Francisco, 26 P.3d 1008 (Wash. Ct. App. 2001) (finding that defendant's absence from jury view was not error because the view was not evidence). 149. See United States v. Johnson, 4 F.3d 904, 915 (10th Cir. 1993) (finding it proper for the prosecutor to refresh a witness’s recollection by reading the witness’s previous statement to him); United States v. Rinke, 778 F.2d 581, 588 (10th Cir. 1985) (finding it proper to let a witness hold and refer to his notes while testifying to refresh his memory); supra note 59 (citing authority). LEUBSDORF_FINAL 1238 7/19/2006 7:36:19 AM 91 IOWA LAW REVIEW [2006] now that the Federal Rules of Evidence have diluted the authentication requirement by allowing self-authentication of some evidence,150 and, for the rest, authentication by mere “evidence sufficient to support a finding that the matter in question is what its proponent claims.”151 1. The first possible rationale—that documents, photographs, and evidence other than courtroom testimony are less trustworthy than testimony and therefore require backing by a witness—has the disadvantage of being based on an untruth. It is not true that witnesses are in general more reliable than documents or other evidence. It is not even true that the framers of modern evidence law believed in their greater reliability.152 Here is one more example of evidentiary rules based on vastly overbroad generalizations.153 2. A second rationale is that the authentication requirement equalizes evidence law by providing the same safeguard of live sworn evidence subject to cross-examination that the testimony of witnesses provides. But the ability of the opposing party to cross-examine an authenticating witness about whether someone actually wrote a document is not equivalent to the ability to examine the author about what the document says. Moreover, this rationale gives authentication a role already filled by the hearsay rule. If evidence is hearsay and does not fall within an exception, then crossexamination would indeed (under the theory of the hearsay rule) be helpful, but the rule itself protects the opportunity to cross-examine by excluding the evidence, whether or not authentication is provided. If evidence is nonhearsay or falls within an exception, that reflects a judgment that the evidence should be considered even in the absence of crossexamination or live testimony.154 In neither instance is the authentication requirement needed. 150. FED. R. EVID. 902. 151. FED. R. EVID. 901(a). For a pre-Rules critique, see generally John William Strong, Liberalizing the Authentication of Private Writings, 52 CORNELL L.Q. 284 (1967). 152. See supra notes 133–36 and accompanying text (debunking the myth that legal scholars have always preferred witness testimony). But see Edward J. Imwinkelried, The Worst Evidence Principle: The Best Hypothesis as to the Logical Structure of Evidence Law, 46 U. MIAMI L. REV. 1069, 1083–84 (1992) (attributing the authentication requirement to a fear of forgery). 153. See supra Part II.D (discussing evidence law’s reliance on generalizations about classes of evidence). 154. Dale Nance, among other helpful comments, has pointed out to me that the applicability of a hearsay exception warrants dispensing with cross-examination of the hearsay declarant, but does not establish that the declarant actually made the statement. The authentication rule requires evidence of the latter point. But it seems to me that such evidence normally overlaps the factual showing necessary to show that the hearsay exception applies in the first place, for example a showing that a document was written or adopted by an opposing party or constitutes a business record. To the extent this is true, the authentication rule adds nothing to what would in any event be required by the hearsay rule other than requiring the supporting evidence to be placed before the jury as well as the judge. LEUBSDORF_FINAL PRESUPPOSITIONS OF EVIDENCE LAW 7/19/2006 7:36:19 AM 1239 3. A third and more plausible rationale focuses on the narrow scope of the authentication requirement, which might be thought to ensure that evidence is properly related to the facts in dispute. Suppose, for example, that the plaintiff in a personal-injury suit against a tire company submits to the court a ruptured tire that was demonstrably defective. The tire is irrelevant unless the plaintiff shows that the defendant manufactured it and that it was on the car that injured the plaintiff. Authentication, until recently, required evidence of both these facts, although today manufacture can be established by the manufacturer’s imprint on the tire.155 On this view, “[a]uthentication is perhaps the purest example of a rule respecting relevance.”156 Even in situations like this one, this rationale is feeble though not worthless. It is feeble because the authentication rules require that the evidence be linked to the case, and this would in any event be accomplished by the requirement that evidence must be relevant. The example in the previous paragraph simply shows that evidence must sometimes be “connect[ed] up” to be admissible.157 The rationale is nevertheless not worthless because authentication rules could play a modest but helpful role by reminding lawyers and judges of the need to connect evidence, and specifying how in some circumstances this can be done. Unfortunately for this rationale, the application of the witness escort principle extends well beyond the need to ensure that evidence is relevant. Consider, for instance, a contract action in which the plaintiff offers in evidence a document purporting to be a contract between the parties. Even without live testimony, the document is relevant: it helps make the existence of the alleged contract more probable than would otherwise be the case.158 True, the document might be forged, but any evidence may be inaccurate or dishonest. In sum, ensuring that evidence is relevant is not a rationale 155. Compare Keegan v. Green Giant Co., 110 A.2d 599, 601 (Me. 1954) (finding that the Green Giant label did not authenticate a can as produced by Green Giant Co.), and Patel v. Comptroller of Customs, [1966] A.C. 356, 358 (P.C. 1965) (appeal taken from Fiji) (U.K.) (stating that because of the hearsay rule, the source of a bag’s contents cannot be shown by “produce of Morocco” label), with FED. R. EVID. 902(7) (holding that inscriptions and labels suffice to show origin). 156. United States v. Sliker, 751 F.2d 477, 499 (2d Cir. 1984) (Friendly, J.). 157. E.g., 1 STRONG ET AL., supra note 14, § 58; see also FED. R. EVID. 104(b) (considering evidence whose relevance is conditional on “the fulfillment of a condition of fact”); Dale A. Nance, Conditional Relevance Reinterpreted, 70 B.U. L. REV. 447, 449 (1990). 158. FED. R. EVID. 401 (defining relevant evidence). Authentication of a document may be based on its distinctive characteristics, but only “taken in conjunction with circumstances.” FED. R. EVID. 901(b)(4). But see People v. Ely, 503 N.E.2d 88, 89 (N.Y. 1986) (finding that admission of a tape requires clear and convincing evidence that it is genuine and unaltered); Robinson v. Commonwealth, 183 S.E.2d 179, 180 (Va. 1971) (holding that chain of custody of a victim’s clothing must be shown with reasonable certainty). LEUBSDORF_FINAL 1240 7/19/2006 7:36:19 AM 91 IOWA LAW REVIEW [2006] sufficiently broad enough to support the existing requirements that witness testimony accredit all evidence except for other witness testimony.159 4. A fourth rationale for the escort principle is that, in its present form, it excludes only evidence that would not be believed if admitted. Because authentication requires only “evidence sufficient to support a finding that the matter in question is what its proponent claims,”160 it is lacking only when no reasonable jury could accept, for example, that the parties signed a purported contract. Excluding evidence that is not credibly linked to a fact in dispute saves time and prevents jury irrationality. On the other hand, appraising the authenticity of proposed evidence itself takes time, as does evaluating the appraisal on appeal, and judges who reject evidence as well as jurors who believe it may fall short of rationality. Nor is it clear why a check on jury irrationality is especially needed each time an exhibit is admitted, given the standard method of appraising the sufficiency of the evidence on a motion for directed verdict.161 5. The fifth rationale for the witness-escort principle is less a justification than a psychological explanation: the sheer unthinkability, for common lawyers, that evidence could be admitted simply because a lawyer submits it to the court. Perhaps, that reflects our adherence to a view of adjudication as a live, human transaction162—though that view scarcely reflects a world in which trials have become rarities.163 In any event, that unthinkability is precisely the point. Our assumption that evidence not presented by live witnesses is anomalous and suspect shapes evidence law regardless of its basis in fact. No one suggests that a witness may testify only if there is some additional evidence supporting her honesty or personal knowledge. But just such a requirement, with rare-and-recent exceptions,164 applies to evidence other than witness testimony. What we find unthinkable is taken for granted in other legal systems. For hundreds of years, French lawyers in civil cases have relied on 159. But see Nance, supra note 157, at 484–88, 492–97 (defending a flexibly interpreted authentication requirement as a means of promoting the introduction of reasonably available supporting evidence that the proponent might not otherwise introduce). 160. FED. R. EVID. 901(a); CAL. EVID. CODE § 1400 (West 2004) (using similar language). 161. FED. R. CRIM. P. 29 (motion for judgment of acquittal); FED. R. CIV. P. 50 (motion for judgment as a matter of law). There are, however, some checks on jury irrationality that, like authentication, apply to limited classes of evidence. FED. R. EVID. 602 (stating that evidence “sufficient to support a finding” of personal knowledge is required for nonexpert witness, but that evidence may be the testimony of the witness herself); FED. R. EVID. 702 (listing requirements for expert witnesses). 162. See generally Crawford v. Washington, 541 U.S. 36 (2004) (discussing a criminal defendant’s right to confront witnesses); supra notes 93–95 and accompanying text (discussing Old Chief v. United States, 519 U.S. 172 (1997)). 163. See generally Symposium, The Vanishing Trial, 1 J. EMPIRICAL LEGAL STUDS. 459 (2004). 164. See supra notes 140–41 and accompanying text (explaining the implications of selfauthenticating evidence). LEUBSDORF_FINAL 7/19/2006 7:36:19 AM PRESUPPOSITIONS OF EVIDENCE LAW 1241 documents for proof. These documents are exchanged among counsel and then submitted to the court when the case is heard.165 Even when witnesses are used, as now sometimes happens, their testimony is submitted to the court in written form.166 Some other nations follow similar procedures.167 So, indeed, do our own administrative agencies in some situations.168 Such systems do not disregard the possibility that evidence may be forged or otherwise deceptive, but simply excuse its proponent from having to demonstrate the contrary in order to introduce the evidence. The parties remain free to introduce evidence or argument as to the evidence’s authenticity or inauthenticity. C. HISTORY SPEAKS It is tempting to regard the escort principle as a hangover from a preliterate age. Indeed, this historical explanation has its own history. Thomas Peake wrote long ago that: At the time when writing was but little practised among men, and when contracts were authenticated by seals only, it might be proper to insist on having some person who was present at the execution . . . ; but the characters of hand writing are in general so distinguishable from each other, that they cannot easily be mistaken.169 165. JEAN VINCENT & SERGE GUINCHARD, PROCÉDURE CIVILE 643–44, 792–94 (27th ed. 2003); Boucher d’Argis, Histoire Abrégée de l’Ordre des Avocats (1778), in 1 ANDRÉ DUPIN, AÎNÉ, PROFESSION D’AVOCAT 19, 92–94 (1832); James Beardsley, Proof of Fact in French Civil Procedure, 34 AM. J. COMP. L. 459, 470, 475 (1986). 166. VINCENT & GUINCHARD, supra note 165, at 826–57; Beardsley, supra note 165, at 478– 79. 167. See MAURO CAPPELLETTI ET AL., THE ITALIAN LEGAL SYSTEM: AN INTRODUCTION 130–38 (1967) (describing limits on oral testimony); BERNARDO M. CREMADES & EDUARDO G. CABIEDES, LITIGATING IN SPAIN § 12(2)(c) (1989) (explaining that in Spain, documents are admitted if properly filed with the court, unless the opponent challenges their authenticity); ELENA MERINO-BLANCO, THE SPANISH LEGAL SYSTEM 124–25 (1996) (explaining that most proceedings are written); see also RADHIKA SINGHA, A DESPOTISM OF LAW: CRIME AND JUSTICE IN EARLY COLONIAL INDIA 46 (1998) (describing resistance to the English introduction of sworn oral testimony by Indians who were accustomed to the traditional method under which witnesses and supporters signed a party’s pleadings). But see MURRAY & STÜRNER, supra note 41, at 278–80, 599 (explaining that in German civil procedure, authenticity of documents must be shown when not admitted; but that documents and other tangible evidence are usually placed before the court without a live witness). 168. See generally Vt. Yankee Nuclear Power Corp. v. NRDC, 435 U.S. 519 (1978) (demonstrating the submission of written witness testimony in a licensing proceeding); United States v. Fla. E. Coast Ry., 410 U.S. 224 (1973) (demonstrating the submission of written witness testimony in a rulemaking after hearing); Richardson v. Perales, 402 U.S. 389 (1971) (demonstrating the submission of written witness testimony in a disability-benefits denial). 169. THOMAS PEAKE, A COMPENDIUM OF THE LAW OF EVIDENCE 102 (3d ed. 1808). On the asserted connection between character and handwriting, see generally Randall McGowen, LEUBSDORF_FINAL 1242 7/19/2006 7:36:19 AM 91 IOWA LAW REVIEW [2006] Although there is some basis for this explanation, ultimately it is our own presuppositions rather than those of antiquity that must bear the responsibility for the principle’s continuing vitality. One difficulty with the historical explanation is that the transition from orality to writing—somewhat like the rise of the middle class—has been ascribed to widely differing dates. The rise of writing in legal and other matters in England is most frequently ascribed to the twelfth through fourteenth centuries.170 But even in the tenth and eleventh centuries, before there was a common law of evidence, documents were used as evidence in Anglo-Saxon and early Norman England.171 For that matter, historians have described the transition from orality to literacy among the ancient Egyptians, Jews, and Greeks.172 Certainly, the English judges who created evidentiary rules were literate and lived in a society that revered written authority; indeed, the inaccessibility of the written word may actually have enhanced its prestige. There is no reason to suppose that they were suspicious of evidentiary documents, which were after all written—though also, sometimes, forged—by members of their own social group. On the other hand, if we are looking for suspicion of writing, Derrida will inform us that all of Western philosophy has been marked by an ambivalent preference for speech over writing as being more spontaneous, interrogable, and “present”—precisely the view that the escort principle embodies.173 Knowing the Hand: Forgery and the Proof of Writing in Eighteenth-Century England, 24 HIST. REFLECTIONS 385 (1998). 170. M.T. CLANCHY, FROM MEMORY TO WRITTEN RECORD: ENGLAND, 1066–1307, at 260–65 (1979); see RICHARD FIRTH GREEN, A CRISIS OF TRUTH: LITERATURE AND LAW IN RICARDIAN ENGLAND, at xiv (1999); EMILY STEINER, DOCUMENTARY CULTURE AND THE MAKING OF MEDIEVAL LITERATURE 4–7 (2003). 171. Patrick Wormald, Charters, Law, and the Settlement of Disputes in Anglo-Saxon England, in the Settlement of Disputes, in EARLY MEDIEVAL EUROPE 149 (Wendy Davies & Paul Fouracre eds., 1986); see also Alan Kennedy, Law and Literacy in the Libellus Aethelwoldi Episcopi, 24 ANGLOSAXON ENG. 131, 161–73 (1995). For Continental equivalents, see JEFFREY A. BOWMAN, SHIFTING LANDMARKS: PROPERTY, PROOF, AND DISPUTE IN CATALONIA AROUND THE YEAR 1000, at 141–64 (2004); CHRIS WICKHAM, COURTS AND CONFLICT IN TWELFTH-CENTURY TUSCANY 28, 41, 73–76, 131, 296 (2005). Documentary evidence of title was used to some extent in the Domesday Inquest of 1086, and the Domesday Book itself was considered unimpeachable authority by the twelfth century. R. WELLDON FINN, THE DOMESDAY INQUEST AND THE MAKING OF DOMESDAY BOOK 102–13 (1961); ROBIN FLEMING, DOMESDAY BOOK AND THE LAW: SOCIETY AND LEGAL CUSTOM IN EARLY MEDIEVAL ENGLAND 56–67 (1998); DAVID ROFFE, DOMESDAY: THE INQUEST AND THE BOOK 4–5 (2000). 172. WILLIAM M. SCHNIEDEWIND, HOW THE BIBLE BECAME A BOOK: THE TEXTUALIZATION OF ANCIENT ISRAEL 91–117 (2004); see also John Baines, Literacy and Ancient Egyptian Society, 18 MAN: J. ROYAL ANTHROPOLOGICAL INST. 572, 589 (1983) (describing written legal documents). See generally ERIC A. HAVELOCK, THE MUSE LEARNS TO WRITE: REFLECTIONS ON ORALITY AND LITERACY FROM ANTIQUITY TO THE PRESENT (1986). 173. JACQUES DERRIDA, OF GRAMMATOLOGY ch. 1 (Gayatri Chakravorty Spivak trans., 1976) (1974); JACQUES DERRIDA, WRITING AND DIFFFERENCE 196–97 (Alan Bass trans., 1978) (1967). LEUBSDORF_FINAL PRESUPPOSITIONS OF EVIDENCE LAW 7/19/2006 7:36:19 AM 1243 Even if we stick to late medieval England, it is far from clear what consequences for the law of evidence flowed from the growing familiarity of documents. Bracton did indeed call for witnesses to verify the making or approval of a charter, but this may simply reflect the substantive principle that a charter by itself did not convey land without some sort of delivery ritual.174 And “Glanvill,” writing before Bracton and speaking of debts rather than real property, recognized proof of a charter by comparison with other documents bearing the same seal whose validity had been admitted.175 In addition, the civil-law practice under which a notary’s certificate established a document’s authenticity made at least tiny inroads in medieval England.176 At the other end of the time line, we can see judges continuing to require live witnesses even when that can hardly be attributed to distrust of writing. The authentication requirement reached its full blossoming during the nineteenth century, a period more pervaded by writing and printing than any before or perhaps since.177 It was during that century that judges applied the requirement of an escorting witness to photographs, of which they were suspicious for reasons entirely unrelated to the lingering effects of medieval attitudes to writing.178 Nor could such attitudes be responsible when the escort requirement is applied today to non-documentary exhibits or to telephone conversations, or when it is said that views are not evidence.179 In such instances, the escort principle, whatever its origins, has 174. 2 BRACTON ON THE LAWS AND CUSTOMS OF ENGLAND 119–21, 124 (George E. Woodbine ed., Samuel E. Thorne trans., 1968) (written during the thirteenth century). 175. RANULF DE GLANVILL, THE TREATISE ON THE LAWS AND CUSTOMS OF THE REALM OF ENGLAND COMMONLY CALLED GLANVILL 127 (G.D.G. Hall ed. & trans., 1965) (written 1187–89). Although comparison with other documents was allowed by Glanvill and is allowed today, FED. R. EVID. 901(b)(3), it has not always been allowed. Doe v. Suckermore, (1836) 111 Eng. Rep. 1331, 1333 (K.B.); Evans, supra note 26, at 141–44 (citing conflicting authority); Jennifer L. Mnookin, Scripting Expertise: The History of Handwriting Identification Evidence and the Judicial Construction of Reliability, 87 VA. L. REV. 1723, 1764–74 (2001). 176. C.R. CHENEY, NOTARIES PUBLIC IN ENGLAND IN THE THIRTEENTH AND FOURTEENTH CENTURIES 19–23 (1972). For later history, see generally C.W. BROOKS ET AL., NOTARIES PUBLIC IN ENGLAND SINCE THE REFORMATION (1991). In parts of the Continent, the notarial tradition was of course much stronger than in England. See generally ALAIN MOREAU, LES METAMORPHOSES DU SCRIBE: HISTOIRE DU NOTARIAT FRANÇAIS (1989). For a description of the intensive use of documentary evidence, both notarial and not, in late medieval civil litigation, see DANIEL LORD SMAIL, THE CONSUMPTION OF JUSTICE: EMOTIONS, PUBLICITY AND LEGAL CULTURE IN MARSEILLE, 1264–1423, at 52–55 (2003). 177. See generally Mudd v. Suckermore, (1836) 111 Eng. Rep. 1331 (K.B.) (agonizing over whether a document may be authenticated by a witness who acquired familiarity with the alleged author’s handwriting in order to testify); 7 JOHN WIGMORE, WIGMORE ON EVIDENCE 693– 815 (Chadbourn rev. ed. 1978) (discussing authentication and its development with reference to virtually no authority before the nineteenth century). 178. Mnookin, supra note 142, at 43. 179. FED. R. EVID. 901(b)(5)(6); MUELLER & KIRKPATRICK, supra note 126, at 1049–63; supra notes 144–45 and accompanying text. The notion that views are not evidence is another nineteenth-century development. 4 WIGMORE, supra note 177, § 1168. LEUBSDORF_FINAL 1244 7/19/2006 7:36:19 AM 91 IOWA LAW REVIEW [2006] acquired an independent vitality that makes it one of the presuppositions with which lawyers and judges approach evidentiary issues. The historical trends that lawyers and judges now have to face concern less the relative primacy of speakers and writers than the rise of new kinds of evidence whose force rests only in part on one individual’s authority. One such kind of evidence is scientific evidence, appraised by the courts under rules that put to one side some of the traditional law of evidence.180 Another is information emerging more or less anonymously from a computer system, database, or reference work, and often fitting within the steadily growing confines of the business-record rule and various parallel rules,181 or prepared especially for use at trial.182 Neither kind of evidence is infallible, even when it has long been accepted,183 and a tribunal using either of them should have ways of appraising its reliability. Often, but not always, those ways will rely on live testimony. But the assumption that a live witness is a necessary and sufficient condition for admissibility will sometimes require too little to assure reliability and sometimes more than is needed. IV. AMBIVALENCE AND TRIAL STRUCTURE Below the explicit rules and policies of evidence law, and below even the methods and assumptions so far considered, lies what might be called its unconscious—a dark realm of drives in conflict with each other and with themselves. Two of these will be discussed here. The first is the ambivalent polarity of reason and emotion, and the second is the ambivalent polarity between jurors and judges. Each has often been discussed, but the ambivalence of each and its interplay with the other have not yet been fully explored. This Article charts the paradox by which fact finding, which the law usually claims should be dispassionate and objective, is assigned to jurors whom the same law usually sees as emotional and unreliable. This paradox goes far to explain, if not to justify, some of the oddities of evidence law, which is compelled to disregard its usual beliefs when they conflict with the role assigned to jurors by the structure of trials. More broadly, one might view the presuppositions already discussed and evidence law in general as a 180. See Simon Schaffer, Self Evidence, in QUESTIONS OF EVIDENCE 56 (James Chandler et al. eds., 1994) (relating the rise of science to a decline in reliance on authority as a basis for belief). See generally FED. R. EVID. 702–705; Daubert v. Merrell Dow Pharm., 509 U.S. 579 (1993); DAVID L. FAIGMAN ET AL., MODERN SCIENTIFIC EVIDENCE (2d ed. 2002). 181. FED. R. EVID. 803(6)–(11), 803(17), 803(18), 902(4)–(7), 902(11), 902(12). 182. See generally Verizon Directories Corp. v. Yellow Book USA, Inc., 331 F. Supp. 2d 136 (E.D.N.Y. 2004) (computer-generated exhibits); State v. Swinton, 847 A.2d 921 (Conn. 2004) (digitally enhanced photographs). 183. See generally SIMON A. COLE, SUSPECT IDENTITIES: A HISTORY OF FINGERPRINTING AND CRIMINAL IDENTIFICATION (2001); Mnookin, supra note 175; D. Michael Risinger et al., Exorcism of Ignorance as a Proxy for Rational Knowledge: The Lesson of Handwriting Identification “Expertise,” 137 U. PA. L. REV. 731 (1989). LEUBSDORF_FINAL 7/19/2006 7:36:19 AM PRESUPPOSITIONS OF EVIDENCE LAW 1245 giant system of rationalizations, masking ambivalences and contradictions with an appearance of precision and technicality. A. REASON AND EMOTION 1. Bad Emotion It is a familiar proposition that evidence law, and procedural law in general, seeks trials governed by reason rather than emotion. As Rosemary Hunter tersely explains: [T]he rules of evidence clearly embody Enlightenment epistemology. They privilege fact over value, reason over emotion, presence over absence, physical over psychological, perception over intuition. They are part of the same discursive regime of hierarchized dualisms that imparts greater cultural value to the masculine than to the feminine, partly through the association of “masculine” with attributes such as reason, presence, and perception and of “feminine” with emotion, absence, and intuition.184 Mark Cammack relates a similar dichotomy to the distinction between fact and law: the jury is to engage in “factfinding as a value-free process of disinterested evaluation of evidence,” while judges shape values into law.185 Putting these two dichotomies together produces the notion that the jury must remain in the realm of fact, without rising to the realm of law reserved for judges or falling into the realm of emotion. My concern here is with the latter danger. It would be easy to accumulate authorities supporting the view that “[i]n judicial inquiry the cold clear truth is to be sought and dispassionately analyzed under the colorless lenses of the law.”186 The rules on prejudicial evidence, prior crimes, and impeachment are only some of those traditionally based in large part on the danger of evidence appealing to jury emotions.187 More recently, many inside and outside the law have sought to 184. Rosemary C. Hunter, Gender in Evidence: Masculine Norms vs. Feminist Reforms, 19 HARV. WOMEN’S L.J. 127, 129–30 (1996) (citations omitted). The privileging of “masculine” values in judicial procedure is by now a familiar theme. See Elizabeth G. Thornburg, Metaphors Matter: How Images of Battle, Sports and Sex Shape the Adversary System, 10 WIS. WOMEN’S L.J. 225, 226 (1995) (analyzing the masculine metaphor of a lawyer as a valiant hero in a trial that is analogized to a battle or sporting event). 185. Mark Cammack, Evidence Rules and the Ritual Functions of Trials: “Saying Something of Something,” 25 LOY. L.A. L. REV. 783, 793 (1992). On the construction of the law-fact distinction in different cultures, see CLIFFORD GEERTZ, LOCAL KNOWLEDGE 167–234 (1983). 186. F.W. Woolworth Co. v. Wilson, 74 F.2d 439, 443 (5th Cir. 1934) (condemning an appeal to the jury’s sympathy and reference to defendant’s wealth). 187. FED. R. EVID. 403–404, 412, 608–610; see, e.g., United States v. Hernandez, 975 F.2d 1035, 1041 (4th Cir. 1992) (stating that Rule 403 is based on the fear of “‘jury emotionalism or irrationality’” (quoting United States v. Greenwood, 796 F.2d 49, 53 (4th Cir. 1986))); People v. LEUBSDORF_FINAL 7/19/2006 7:36:19 AM 1246 91 IOWA LAW REVIEW [2006] deconstruct this dichotomy of reason and emotion and to challenge the former’s privilege over the latter.188 2. Good Emotion Curiously enough, however, evidence law is more ambivalent than might be thought: sometimes, it prizes passion. The clearest example is the excited-utterance exception to the hearsay rule, under which it is precisely “the stress of excitement caused by the event or condition” that makes the utterance admissible.189 Although “overpowering emotion”190 is thought to make a declarant more credible, it would undermine the reliability of a juror’s decision.191 Similarly, the fear of imminent death is said to justify the admission of dying declarations.192 The lack of empirical support for this belief in emotion as a guarantor of honesty193 makes still more striking its divergence from the standard wisdom that procedural law is wedded to an exclusive faith in reason. Recognizing that relevant evidence may be admitted for its “legitimate moral force”194 is another way in which evidence law legitimates appeals to the emotions. Under this principle, evidence may properly be used to motivate jurors to do their duty, not just to ground rational inferences. The principle has often been invoked on behalf of the prosecution in criminal cases to Zackowitz, 172 N.E. 466, 466 (N.Y. 1930) (“There must be no blurring of the issues by evidence illegally admitted and carrying with it in its admission an appeal to prejudice and passion.”). 188. See generally MARTHA C. NUSSBAUM, UPHEAVALS OF THOUGHT: THE INTELLIGENCE OF EMOTIONS (2001); THE PASSIONS OF LAW (Susan A. Bandes ed., 1999); Lynne N. Henderson, Legality and Empathy, 85 MICH. L. REV. 1574 (1987); Symposium, Reason, Passion, and Justice Brennan, 10 CARDOZO L. REV. 1 (1988). 189. FED. R. EVID. 803(2). 190. State v. Bass, 12 P.3d 796, 804 (Ariz. 2000); People v. Washington, 270 N.W.2d 511, 513 n.1 (Mich. Ct. App. 1978) (citing Commonwealth v. Albrecht, 720 A.2d 693, 704 (Pa. 1998)). 191. E.g., Moore v. Norton, 255 F.3d 95, 117 (3d Cir. 2001) (citing cases); Malandris v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 703 F.2d 1152, 1167–68, 1178–80 (10th Cir. 1981); see California v. Brown, 479 U.S. 538, 543 (1987) (upholding an instruction that the jury should not be swayed by mere sentiment, passion, etc.). Although many (or most) of the cases can be explained by focusing on how a party appealed to a specific emotion, judicial language often implies a broader condemnation of all verdicts based on emotion. 192. FED. R. EVID. 804(b)(2); see, e.g., King v. Woodcock, (1789) 168 Eng. Rep. 352, 353–54 (K.B.); Manderson, supra note 20, at 35. Somewhat similar are rules that rely on a speaker’s interests, rather than his emotions, to warrant his reliability. FED. R. EVID. 803(4), 804(b)(3) (allowing statements for purposes of medical diagnosis; declarations against interest). See generally ALBERT O. HIRSCHMAN, THE PASSIONS AND THE INTERESTS: POLITICAL ARGUMENTS FOR CAPITALISM BEFORE ITS TRIUMPH (1977) (discussing the origins of the distinction between passions and interests). 193. Bryan A. Liang, Shortcuts to “Truth”: The Legal Mythology of Dying Declarations, 35 AM. CRIM. L. REV. 229, 259–60 (1998); Orenstein, supra note 76, at 168–83 (discussing excited utterances). The weakness of the rationale has long been recognized. See Evans, supra note 26, at 223. 194. 9 WIGMORE, supra note 23, § 2591, at 824–25. LEUBSDORF_FINAL PRESUPPOSITIONS OF EVIDENCE LAW 7/19/2006 7:36:19 AM 1247 justify the admission of varied evidence—for example, photographs of a corpse;195 evidence of what the victim was like when alive;196 and prior bad acts of the defendant197— notwithstanding the defendant’s offer to stipulate to the facts. The admission of victim-impact evidence in capital sentencing proceedings rests on similar grounds.198 And we have already noted Justice Souter’s broad endorsement in Old Chief of the use of detailed stories to “sustain the willingness of jurors to draw the inference[]” of guilt or innocence.199 In these instances, judges uphold the value of jurors’ reactions to the human impact of crime—that is, the value of their emotions—as a ground for the admission of evidence. Indeed, from a broader perspective, rather than privileging reason over emotion, evidence law appears to be engaged in a peripheral and symbolic attempt to exclude emotion from trials, which are permeated by emotion from beginning to end. Just as the finicky rules about when a court may take judicial notice of a fact scarcely limit the vast range of knowledge that jurors and judges routinely rely on,200 the condemnation of emotion is far less significant than one might think. Jurors beholding a criminal trial will feel pity and terror as they would at the performance of a tragedy. A tort case will not be experienced as a discussion of where social policy should place costs,201 but as a struggle to assign blame and make the blameworthy pay. A contract dispute calls on the trier of fact to decide whether the defendant broke his word and whether the plaintiff was harmed as a result. In each instance, those in the courtroom will feel sympathy, shock, admiration, fear, disgust, anger, or other emotions.202 These emotions do not intrude, as it were, from outside a legal system otherwise free of passion but are inherent in the way that the system frames trials. Substantive law defines the issues in terms that implicate moral 195. E.g., State v. Leisure, 772 S.W.2d 674, 681–82 (Mo. Ct. App. 1989); State v. House, 456 S.E.2d 292, 294 (N.C. 1995). 196. E.g., McQueen v. Commonwealth, 669 S.W.2d 519, 523 (Ky. 1984) (descriptive testimony); State v. Broberg, 677 A.2d 602, 610–12 (Md. 1996) (“in life” photograph). 197. United States v. Jemal, 26 F.3d 1267, 1275 (3d Cir. 1994). 198. See generally Payne v. Tennessee, 501 U.S. 808 (1991); Symposium, Victims and the Death Penalty: Inside and Outside the Courtroom, 88 CORNELL L. REV. 257 (2003). 199. Old Chief v. United States, 519 U.S. 172, 187 (1997); see supra notes 93–96 and accompanying text (discussing the Old Chief case). 200. FED. R. EVID. 201; see also Richard M. Fraher, Adjudicative Facts, Non-Evidentiary Evidence Facts, and Permissible Jury Background Information, 62 IND. L.J. 333, 333–34 (1987); John H. Mansfield, Jury Notice, 74 GEO. L.J. 395, 417–18 (1985). 201. See GUIDO CALABRESI, THE COSTS OF ACCIDENTS: A LEGAL AND ECONOMIC ANALYSIS 24– 33 (1970). 202. See MARTHA C. NUSSBAUM, POETIC JUSTICE: THE LITERARY IMAGINATION AND PUBLIC LIFE 79–121 (1995) (comparing judges to readers of fiction); Peter Brooks, Policing Stories, in LAW’S MADNESS 29 (Austin Sarat et al. eds., 2003) (contending that the law represses awareness of its own dependence on story telling). See generally ROBERT P. BURNS, A THEORY OF THE TRIAL (1999). LEUBSDORF_FINAL 7/19/2006 7:36:19 AM 1248 91 IOWA LAW REVIEW [2006] judgment and emotional involvement. Procedural law makes trials adversarial combats. Evidence law encourages those involved in the dispute to tell their stories in person. Lawyers, following a rhetorical tradition as old as trials themselves,203 shape not just their arguments, but their whole cases to touch the feelings of the triers of fact.204 As part of the legal system, evidence law thus promotes and discourages appeals to emotion. Needless to say, a juror can distinguish one emotional appeal from another.205 Seeking the conviction of a defendant because he has cruelly harmed the victim of the alleged crime is not the same as seeking his conviction because of his race, though each effort may involve appeals to emotion. Much of the case law can be viewed as drawing distinctions of this sort between proper and prejudicial appeals. Yet, it remains true that one man’s prejudice is another woman’s legitimate moral force.206 Ultimately, evidence law does not privilege reason over emotion but remains firmly and complexly ambivalent. B. JURORS Ambivalence also pervades evidence law where treatment of jurors is concerned. Here, the accepted view is that much of the law reflects judicial distrust of jurors. That is correct, but only part of the story. 1. Bad Jurors The usual manifestation of mistrust is the assertion that jurors will give too much weight to a given kind of evidence.207 Judges assert that expert evidence must be carefully controlled because “scientific proof may in some instances assume a posture of mystic infallibility in the eyes of a jury of laymen,”208 a proposition contrary to substantial empirical evidence.209 203. See generally ARISTOTLE ON RHETORIC: A THEORY OF CIVIC DISCOURSE (George A. Kennedy trans., 1991). 204. See THOMAS A. MAUET, TRIAL TECHNIQUES ch. 2 (5th ed. 2000); NOELLE C. NELSON, A WINNING CASE: HOW TO USE PERSUASIVE COMMUNICATION TECHNIQUES FOR SUCCESSFUL TRIAL WORK 205–38 (1991); EDWARD T. WRIGHT, HOW TO USE COURTROOM DRAMA TO WIN CASES 143 (1987). 205. Susan Bandes, Empathy, Narrative, and Victim Impact Statements, 63 U. CHI. L. REV. 361, 370–71 (1996). 206. See Commonwealth v. DeJesus, 860 A.2d 102, 115 (Pa. 2004) (considering whether the prosecution’s argument that the jury should “send a message” is proper). 207. See generally Richard Friedman, Minimizing the Jury Over-Valuation Concern, 2003 MICH. ST. L. REV. 967 (criticizing this assertion). 208. United States v. Addison, 498 F.2d 741, 744 (D.C. Cir. 1974). For similar assertions, see Barefoot v. Estelle, 463 U.S. 880, 926–28 (1983) (Blackmun, J., dissenting); Allison v. McGhan Med. Corp., 184 F.3d 1300, 1310 (11th Cir. 1999) (finding a jury more likely to be “awestruck by the expert’s mystique”). 209. Joseph Sanders, The Merits of the Paternalistic Justification for Restrictions on the Admissibility of Expert Evidence, 33 SETON HALL L. REV. 881, 907–08 (2003); Neil Vidmar & Shari Seidman Diamond, Juries and Expert Evidence, 66 BROOK. L. REV. 1121, 1166–67 (2001). LEUBSDORF_FINAL PRESUPPOSITIONS OF EVIDENCE LAW 7/19/2006 7:36:19 AM 1249 Courts sometimes express the equally-unfounded belief that jurors will be unduly impressed by mathematical evidence.210 The hearsay rule has likewise often been attributed to the likelihood that jurors will give undue weight to hearsay,211 although research on whether this is correct has been at most inconclusive.212 Character evidence as well “is said to weigh too much with the jury . . . .”213 Indeed, Wigmore ascribes even the authentication requirement to the tendency of credulous jurors to accept any document placed before them at its full apparent value.214 The belief in jury overvaluation is thus a handy rationale that can be invoked to justify almost any exclusionary-evidence rule. Curiously, there are few instances in which evidence law proceeds on the contrary assumption that jurors will undervalue evidence.215 Such an assumption, whether or not correct, has no place in evidence law because it would not support the exclusionary rules. It could justify instructions on the weight juries should attribute to evidence, but courts have shown little interest in requiring such instructions. 210. E.g., United States ex rel. DiGiacomo v. Franzen, 680 F.2d 515, 518 (7th Cir. 1982); Ex parte Perry, 586 So. 2d 242, 254 (Ala. 1991); People v. Collins, 438 P.2d 33, 41 (Cal. 1968). For empirical evidence to the contrary, see Margaret A. Berger, Laboratory Error Seen Through the Lens of Science and Policy, 30 U.C. DAVIS L. REV. 1081, 1094–95 (1997); Dale A. Nance & Scott B. Morris, Juror Understanding of DNA Evidence: An Empirical Assessment of Presentation Formats for Trace Evidence with a Relatively Small Random-Match Probability, 34 J. LEGAL STUD. 395, 396–97 (2005). 211. Berkeley Peerage Case, (1811) 171 Eng. Rep. 128, 135 (H.L.) (appeal taken from Eng.); Wright v. Tatham, (1837) 112 Eng. Rep. 488, 512 (Exch.), aff’d, (1838) 7 Eng. Rep. 559, 567 (appeal taken from Eng.); DAMAŠKA, supra note 41, at 28 (describing this as the most widely accepted rationale for the hearsay rule); 1 THOMAS STARKIE, A PRACTICAL TREATISE OF THE LAW OF EVIDENCE 45 (1860); 1 WIGMORE, supra note 23, § 4b. See Note, The Theoretical Foundation of the Hearsay Rules, 93 HARV. L. REV. 1786, 1788 (1980). But cf. Dale A. Nance, Rethinking Confrontation After Crawford, INT’L COMMENT. ON EVIDENCE, Oct. 2004, at 1, 13, http://www.bepress.com/cgi/viewcontent.cgi? article=1017&context=ice (arguing that Crawford v. Washington, 541 U.S. 36 (2004), shifts the rationale of Confrontation Clause case law from distrust of jurors to distrust of government). 212. See generally Roger C. Park, Visions of Applying the Scientific Method to the Hearsay Rule, 2003 MICH. ST. L. REV. 1149 (evaluating various attempts to measure the effects of hearsay on jurors and finding these attempts lacking). 213. Michelson v. United States, 335 U.S. 469, 475–76 (1948); 1A JOHN HENRY WIGMORE, EVIDENCE IN TRIAL AT COMMON LAW § 58.2, at 1212 (Tillers ed., 1983); Friedman, supra note 207, at 979 (urging that the character-evidence rules should be based not on the claim that the jury will fail to appraise the evidence rationally, but on the likelihood that it will cause the jury (or the judge) to lower the burden of persuasion because it believes the defendant to be a bad person). 214. 4 WIGMORE, supra note 23, § 1157, at 254. 215. One example is the fresh-complaint rule. This rule permits the admission of an alleged rape victim’s fresh complaint on the theory that jurors might otherwise disbelieve her testimony because of an erroneous belief that rape victims usually complain. E.g., People v. Brown, 883 P.2d 949, 956–57 (Cal. 1994) (citing State v. Hill, 578 A.2d 370, 377–78 (N.J. 1990)). LEUBSDORF_FINAL 1250 7/19/2006 7:36:19 AM 91 IOWA LAW REVIEW [2006] Overvaluation is far from the only failing ascribed to the jury. Rules excluding evidence of subsequent remedial measures, settlement offers and discussions,216 the purchase of insurance,217 and a party’s consultation with counsel218 are based, at least in part, on the fear that jurors would draw unwarranted inferences from such evidence or would fail to recognize the social desirability of protecting persons who engage in the inadmissible conduct from adverse consequences in court. The courts’ discretion to exclude unduly prejudicial evidence guards against a variety of improper influences that judges and lawyers do not trust jurors to resist.219 Procedural law also resists such influences by means such as limits on arguments to the jury,220 summary judgments and directed verdicts,221 and new trials.222 Recent studies of the history of the law of evidence do not require us to abandon the view that suspicion of jurors underlies much of that law. True, not many historians would now accept without qualification Thayer’s version of the development of evidence law as a way to control juries after earlier means of doing so, such as the attaint, became obsolete in the seventeenth century.223 Mirjan Damaška has pointed out how characteristics of commonlaw procedure, such as a bifurcated tribunal composed of judge and jury, made complex evidence law possible.224 John Langbein has demonstrated how the appearance, in the eighteenth century, of lawyers at English felony trials led to evidentiary challenges, out of which evidence law arose.225 216. FED. R. EVID. 408–410; see 2 JACK B. WEINSTEIN & MARGARET A. BERGER, WEINSTEIN’S EVIDENCE ¶ 408[012], at 408–31 (cumm. supp. 1996) (stating rationale). 217. FED. R. EVID. 411 & advisory commitee’s note (stating rationale). This rule is just one example of the quixotic attempt to hide from jurors the fact that the real struggle in tort cases is typically between the plaintiff and the defendant’s insurer rather than between the plaintiff and the nominal defendant. See FLEMING JAMES ET AL., CIVIL PROCEDURE § 10.9 (5th ed. 2001). 218. Henderson v. United States, 632 A.2d 419, 432–34 (D.C. 1993); Casey v. State, 722 A.2d 385, 389 (Md. Ct. Spec. App. 1999); see Knorr-Bremse Systeme Fuer Nutzfahrzeuge GmbH v. Dana Corp., 383 F.3d 1337, 1344–45 (Fed. Cir. 2004) (barring adverse inferences from invocation of lawyer-client privilege or failure to consult counsel). 219. FED. R. EVID. 403; see WEINSTEIN, supra note 39, § 403.04. 220. E.g., N.Y.C. Cent. R.R. Co. v. Johnson, 279 U.S. 310, 316–19 (1929); MODEL RULES OF PROF’L CONDUCT R. 3.4(e) (1989) (noting that a lawyer may not state a personal opinion to the jury); Debra T. Landis, Annotation, Prosecutor’s Appeal in Criminal Case to Racial, National, or Religious Prejudice as Grounds for Mistrial, New Trial, Reversal, or Vacation of Sentence—Modern Cases, 70 A.L.R. 4th 664 (1989) (discussing prosecutor’s appeals to racial, national, or religious prejudice). 221. FED. R. CIV. P. 50, 56. 222. FED. R. CIV. P. 59; 12 MOORE’S FEDERAL PRACTICE § 59.13[2][e]–[g] (3d ed. 1998). See generally State Farm Mut. Auto. Ins. Co. v. Campbell, 538 U.S. 408 (2003) (discussing judicial control of punitive damage awards). 223. JAMES BRADLEY THAYER, A PRELIMINARY TREATISE ON EVIDENCE AT THE COMMON LAW 2, 180–81, 534–35 (1898). 224. DAMAŠKA, supra note 41, at 46–52. 225. See supra note 24. George Fisher describes some evidence rules as meant to avoid the necessity of deciding between conflicting sworn testimony. Fisher, supra note 25, at 581. LEUBSDORF_FINAL 7/19/2006 7:36:19 AM PRESUPPOSITIONS OF EVIDENCE LAW 1251 But our greater understanding of why evidence law developed in England during the eighteenth century does not explain the content or rationale of that law, and therefore, does not destroy the significance of distrust of jurors. And even if that distrust may not fully explain the origin of many evidence rules, it has regularly been invoked to warrant their maintenance and expansion for almost two-hundred years.226 Whatever evidence law may have been based on long ago, it is now based in very large part on distrust of juries. 2. Good Jurors Even while it regards jurors with sometimes-exaggerated mistrust, the law of evidence relies on them with a trust that may also be exaggerated. The belief that jurors will obey instructions to disregard evidence they have heard227 is only one of the implausible assumptions about jury instructions on which the law of evidence is built. Another is that jurors will follow instructions to use evidence only for specified purposes—an assumption contrasting oddly with the fact that the rules excluding other uses are often based on the belief that jurors cannot be trusted to evaluate certain evidence properly. Another assumption is that jurors will understand instructions traditionally composed of unfamiliar words in complex sentences,228 and they will remember them even when the judge does not exercise her discretion to let the jurors take written instructions to the jury room.229 Yet another is that they will be able to overcome misleading hints and omissions in instructions.230 Still another is that, after learning at the end of the trial what standards they are to apply, they will be able to apply them accurately to evidence heard long before.231 226. See EDMUND MORRIS MORGAN, SOME PROBLEMS OF PROOF UNDER THE ANGLOAMERICAN SYSTEM OF LITIGATION 106–17 (1956) (arguing that distrust of juries was first invoked to justify the hearsay rule in the 1830s); see also supra note 207 (citing authority). 227. See supra notes 4, 65–67 and accompanying text. 228. Alan Reifman et al., Real Jurors’ Understanding of the Law in Real Cases, 16 LAW & HUM. BEHAV. 539, 539 (1992) (finding that jurors understand fewer than half of the instructions given); Michael J. Saks, What Do Jury Experiments Tell Us About How Juries (Should) Make Decisions?, 6 S. CAL. INTERDISC. L.J. 1, 34–37 (1997). On attempts to rewrite pattern instructions to make them more comprehensible, see Leonard Post, Spelling It Out in Plain English: California Simplifies Its Jury Instructions, NAT’L L.J., Nov. 8, 2004, at P1. 229. See generally United States v. Holman, 680 F.2d 1340 (11th Cir. 1982); Matire v. State, 232 So. 2d 209 (Fla. Dist. Ct. App. 1970). 230. See generally Stephen P. Garvey et al., Correcting Deadly Confusion: Responding to Jury Inquiries in Capital Cases, 85 CORNELL L. REV. 627 (2000); Lawrence M. Solan, Refocusing the Burden of Proof in Criminal Cases: Some Doubt About Reasonable Doubt, 78 TEX. L. REV. 105 (1999). 231. See generally Neil P. Cohen, The Timing of Jury Instructions, 67 TENN. L. REV. 681 (2000); Paula L. Hannaford et al., The Timing of Opinion Formation by Jurors in Civil Cases: An Empirical Examination, 67 TENN. L. REV. 627 (2000). For advice on how to avoid some of these problems, see MANUAL FOR COMPLEX LITIGATION (FOURTH) § 12.43 (2004). LEUBSDORF_FINAL 1252 7/19/2006 7:36:19 AM 91 IOWA LAW REVIEW [2006] Faith in jurors’ ability to discern which of the conflicting witnesses is telling the truth is another foundation of the evidentiary system.232 This faith is not merely built into existing law but is routinely invoked by reformers as a reason to admit classes of previously excluded evidence.233 The law’s faith is more striking because of its dubious basis in reality. Empirical research shows that jurors are not able to tell by observing witnesses which of them is telling the truth.234 When expert witnesses conflict, jurors make a serious effort to decide which is correct, but they are often hampered by confusion—though not necessarily more so than judges or lawyers would be.235 The law also assumes that jurors will be able to remember the evidence well enough to appraise it correctly. This assumption made more sense when all trials began and ended the same day236 than it does today, when some trials may last for weeks or months.237 Yet, even today, the jury may take notes238 or read part of the trial transcript239 only if the judge chooses to permit them to do so. Perhaps, this will change if and when all evidence is videorecorded and only an edited recording of the admissible portions is played to the jury. In the meantime, it is rather amazing that important 232. On the rise of this faith, see Fisher, supra note 25, at 581. See generally Chris William Sanchirico, Evidence, Procedure, and the Upside of Cognitive Error, 57 STAN. L. REV. 291 (2004) (explaining how evidentiary and procedural rules promote lie detection). 233. See generally MODEL CODE OF EVIDENCE 48 (1942); Gordon Van Kessel, Hearsay Hazards in the American Criminal Trial: An Adversary-Oriented Approach, 49 HASTINGS L.J. 477 (1998) (citing authority). 234. ALDERT VRIJ, DETECTING LIES AND DECEIT 67–69 (2000); Wellborn, supra note 10, at 1075. 235. Vidmar & Diamond, supra note 209, at 1176–78. 236. Not until 1794 was it established that the court could adjourn a jury trial overnight, rather than proceeding continuously until the trial ended. 24 T.B. HOWELL, A COMPLETE COLLECTION OF STATE TRIALS 414–18 (1818) (R. v. Hardy (1794)); 25 HOWELL, supra, 128–32 (R. v. Tooke (1794)). On the extreme brevity of eighteenth-century trials, see Daniel D. Blinka, Trial by Jury on the Eve of Revolution: The Virginia Experience, 71 UMKC L. REV. 529 (2003), which discusses trials in Virginia; John H. Langbein, Shaping the Eighteenth-Century Criminal Trial: A View from the Ryder Sources, 50 U. CHI. L. REV. 1, 115–23 (1983), which discusses English criminal trials; and James Oldham, Law-Making at Nisi Prius in the Early 1800s, 25 J. LEGAL HIST. 221, 226–28 (2004), which discusses English civil trials. 237. See Marc Galanter, The Vanishing Trial: An Examination of Trials and Related Matters in Federal and State Courts, 1 J. EMPIRICAL LEGAL STUD. 459, 477–79, 497–98 (2004). 238. People v. Hues, 704 N.E.2d 546, 549 (N.Y. 1998); Steven D. Penrod & Larry Heuer, Tweaking Commonsense: Assessing Aids to Jury Decision Making, 3 PSYCHOL. PUB. POL’Y & L. 259, 262–67 (1997). 239. E.g., United States v. Hernandez, 27 F.3d 1403, 1408 (9th Cir. 1994); United States v. Edwards, 968 F.2d 1148, 1152 (11th Cir. 1991); see FED. R. EVID. 803(18) (stating that even when learned treatises are admitted in evidence, they are not exhibits and hence, do not go to the jury room). But see ABA, PRINCIPLES FOR JURIES AND JURY TRIALS 17–21 (2005), available at http://www.abanet.org/juryprojectstandards/principles.pdf (recommending methods to improve jury comprehension). LEUBSDORF_FINAL 7/19/2006 7:36:19 AM PRESUPPOSITIONS OF EVIDENCE LAW 1253 disputes are still resolved by jurors who are expected to remember complicated testimony given during many days. In short, evidence law treats jurors with great trust as well as with great suspicion. In itself, it would not be surprising to proceed on the assumption that jurors have strengths as well as weaknesses. Who does not? The surprising thing is that both strengths and weaknesses seem to have been designated so arbitrarily, neither of them consistent with what we know about how jurors actually behave.240 This suggests that some other logic underlies the designation. C. JURORS, EMOTIONS, AND THE STRUCTURE OF THE TRIAL The ambivalences we have seen can most plausibly be explained as resulting from the structure of the Anglo-American trial. That structure gives certain roles to jurors and others to judges and then assigns them strengths and weaknesses appropriate to their roles. Like a playwright dramatizing a myth, the law imputes to its personages characters appropriate to what it has already decided they must do. That may not be the whole explanation, but it is at least illuminating. At least since Coke’s time, jurors have been expected to decide factual but not legal issues.241 If jurors are to decide factual issues, they will have to be able to determine which witnesses are credible. If they are not to resolve issues of law, they will have to be able to understand and follow a judge’s instructions on what the law requires. And if trials are to be live, oral events, then—at least until very recent times—jurors will have to be able to remember testimony. So it is necessary to attribute these abilities to jurors regardless of what jurors are actually like or what law elsewhere assumes them to be like. The almost unthinkable alternative would be a radical change in the structure of trials. On the other hand, rules preventing juries from hearing large classes of evidence can best be justified by hypothesizing that their weaknesses will prevent them from appraising that evidence properly. Of course, one could simply do without exclusionary rules. As a historical matter, those rules may have originated in a belief that only the best evidence should be admitted, or in the belief that jurors should not be asked to choose between conflicting sworn testimony.242 But once lawyers realized that a competent factfinder would want to consider even defective evidence for what it is 240. Cf. Lempert, supra note 10, at 343 (arguing that many of the factual underpinnings of evidence law are false). 241. EDWARD COKE, THE FIRST PART OF THE INSTITUTES OF THE LAWS OF ENGLAND; OR, A COMMENTARY UPON LITTLETON 1556 (19th ed. 1823). But see generally Matthew P. Harrington, The Law-Finding Function of the American Jury, 1999 WIS. L. REV. 377. 242. See Fisher, supra note 25, at 599; Gallanis, supra note 24, at 521; see also ALLEN, supra note 25, at 20–24 (discussing the nineteenth-century belief in fixed-evidence rules). LEUBSDORF_FINAL 1254 7/19/2006 7:36:19 AM 91 IOWA LAW REVIEW [2006] worth,243 the rules could be justified only by attributing some lack of competence to the jury. The structure of trials also supports in another way the attribution of weaknesses to jurors. Because the Anglo-American trial involves both a judge and a jury, judges are in a position to institute and maintain rules that would prevent the trier of fact from receiving evidence244 and to justify those rules by ascribing to jurors poor reasoning and susceptibility to emotion. Recognizing that jurors do not differ much from judges in the way they decide cases245 would not necessarily bar such a justification. Judges could simply claim to be using their powers to protect jurors from frailties common to all. But it is natural for judges and lawyers to justify deciding what kinds of evidence should be kept from jurors by assuming that they know better than jurors both the true value of evidence and the ways jurors use or misuse evidence. “The jurors, if they are aware of any insult at all, are not part of the club of legal professionals and will soon be gone from the courtroom.”246 So, the procedural system in which judges rule on what the jury will hear implies a judicial posture of superior cognitive ability and greater freedom from bias. Trial structure also helps explain the law’s ambivalence to emotion. The judgment a court enters, although based on reasoning, is also an act of power.247 Acts require motivation, and motivating jurors and judges necessarily involves appeal to values and emotions. Some such appeals thus inhere in any trial. Judges may not always notice them but will not condemn them. In other situations, trial structure encourages judges to condemn evidence or arguments as appealing to emotion. Like the court’s judgments, a jury’s asserted errors are acts grounded in values and emotions as well as in reasoning—and the same is true of a judge’s rulings that seek to prevent or remedy those errors. Judges must motivate and justify rulings that might be seen as interfering with the jury, and judges will naturally see the errors they seek to forestall or remedy as grounded in the jury’s inappropriate values or emotions. Granted a cultural belief that judicial proceedings should be rational, viewing a jury’s actual or potential error as emotional provides a powerful justification for a judge’s counteracting rulings. As that justification 243. E.g., 3 JEREMY BENTHAM, RATIONALE OF JUDICIAL EVIDENCE 536–42 (J.S. Mill ed., 1827). 244. See DAMAŠKA, supra note 41, at 26–57. 245. Kevin M. Clermont & Theodore Eisenberg, Trial by Jury or Judge: Transcending Empiricism, 77 CORNELL L. REV. 1124, 1152 (1992); Chris Guthrie et al., Inside the Judicial Mind, 86 CORNELL L. REV. 777, 816 (2001). See generally Roselle L. Wissler et al., Decisionmaking About General Damages: A Comparison of Jurors, Judges, and Lawyers, 98 MICH. L. REV. 751 (1999) (discussing the decision-making process of jurors, judges, and lawyers). 246. Nance, supra note 211, at 13 n.35 (comparing a rationale based on distrust of prosecutors to one based on distrust of jurors). 247. See Robert M. Cover, Violence and the Word, 95 YALE L.J. 1601, 1621 n.48 (1986). LEUBSDORF_FINAL PRESUPPOSITIONS OF EVIDENCE LAW 7/19/2006 7:36:19 AM 1255 is asserted again and again, judges and lawyers will increasingly see jurors as dangerously open to passion and prejudice. A strong dosage of class and professional chauvinism accentuates the tendency of judges and lawyers to attribute error and emotion to jurors. We who have suffered the benefits of a legal education like to believe that it has vested us with greater ability and rationality in untangling legal disputes than others possess. It is always easier to see irrationality in others than in ourselves. As racial minorities, women, and poor people have gained greater access to jury service, it has become even easier for the powerful to project on jurors the stereotyped incapacity and emotionality that they often attribute to the weak.248 Of course one can also find idealizations of juries.249 My claim is of ambivalence, not of unremitting denigration. By now it should be clear that ambivalence about emotion and ambivalence about jurors are related. The premise that trials should be rational joins with the premise that jurors are irrational to support judicial limits on what jurors may hear. Likewise, the premise that those who decide must be motivated to act in matters of broad impact (that is, must be emotional) joins with the premise that jurors are responsible (that is, dispassionate) voices of the community to support entrusting decisions to juries. The trial’s double aspect (rational and emotional) can be combined with the jury’s double aspect (trustworthy and untrustworthy) to justify both a broad role for the jury and a broad role for judicial control over the jury. Judges and lawyers invoke conflicting sets of premises ad hoc, inconsistently, and without much relation to reality to justify rules that in fact arise from the structure and traditions of the trial.250 The epitome of these conflicts and ambivalences emerges when verdicts are challenged as resulting from passion, prejudice, or incomprehension. Here, the jury’s role as decisionmaker clashes directly with the invocation of jury weaknesses to support evidentiary and procedural doctrines. If judges are to parse and revise verdicts, why should there be trial by jury? But if jurors are prone to emotion and irrationality, why should their verdicts be respected? Evidence law responds to this challenge by closing its eyes, or rather those of the judge. Although passion and prejudice are proclaimed as grounds for vacating verdicts, they may only be found based on the verdict 248. See generally Laura Gaston Dooley, Our Juries, Our Selves: The Power, Perception, and Politics of the Civil Jury, 80 CORNELL L. REV. 325 (1995); see supra note 186 and accompanying text. Academics enjoy the further privilege of seeing as irrational not only jurors but also judges, lawyers, and other academics, as witness the present Article. 249. E.g., R.R. Co. v. Stout, 84 U.S. 657, 663 (1873). 250. Cf. Lempert, supra note 10, at 353 (arguing that evidence law routinely relies on inconsistent and incorrect factual premises). LEUBSDORF_FINAL 1256 7/19/2006 7:36:19 AM 91 IOWA LAW REVIEW [2006] itself and the events of the trial.251 Federal Rule of Evidence 606(b) bars questioning a jury about jury deliberations, the effect of evidence or argument on a “juror’s mind or emotions,” or a “juror’s mental processes,” all subject to an exception for the receipt of extraneous prejudicial information or improper outside influences.252 Jurors may not be asked, for example, whether they drank alcohol or used drugs,253 disregarded the judge’s instructions,254 used bigoted language,255 or were incapacitated.256 The suppression of this evidence is necessary to protect the trial system. Judging from evidence law’s expressions of concern with insulating juries from evidence likely to induce error, one would expect that any proof that a jury had in fact violated instructions or succumbed to prejudice would be considered of vital importance. But letting this proof in would expose judges to “embarrassing choices.”257 If they set aside many verdicts for jury bias or incompetence, they would discredit trial by jury. If they were too willing to reject bias and incompetence as nothing to fuss about, they would discredit trial by jury almost as much, as well as undermining the justifications advanced for the rest of evidence law. The best way out, therefore, is to 251. E.g., Honda Motor Co. v. Oberg, 512 U.S. 415, 425–26 (1994) (discussing judicial review of damage awards); Dossett v. First State Bank, 399 F.3d 940, 945–46 (8th Cir. 2005) (relying on the size of the verdict); Whitehead v. Food Max, Inc., 163 F.3d 265, 267–68 (5th Cir. 1998) (relying on the size of the verdict and the counsel’s improper argument). 252. FED. R. EVID. 606(b). A few states admit evidence more liberally. E.g., CAL. EVID. CODE § 1150 (allowing evidence of statements and conduct in jury room, but not to show impact on jurors’ mental processes). See generally Gardner v. Malone, 376 P.2d 651 (Wash. 1962) (allowing evidence that jurors discussed improper factor). Washington Rule of Evidence 606 leaves this precedent in effect. See generally People v. Harlan, 119 P.3d 616 (Colo. 2005) (allowing evidence that juror brought Bible and notes on relevant passages into jury room and discussed them; citing conflicting authority). 253. See generally Tanner v. United States, 483 U.S. 107 (1987). 254. E.g., Multiflex, Inc. v. Samuel Moore & Co., 709 F.2d 980, 998 (5th Cir. 1983); Andrews v. O’Hearn, 387 N.W.2d 716, 718–19 (N.D. 1986). 255. See generally United States v. Gonzales, 227 F.3d 520 (6th Cir. 2000) (refusing to allow testimony about juror references to defendant’s wealth); United States v. Abcasis, 811 F. Supp. 828 (E.D.N.Y. 1992) (refusing to allow testimony of a juror’s anti-Semitic comments); Rios v. Danuser Mach. Co., 792 P.2d 419 (N.M. Ct. App. 1990) (refusing to allow testimony of a juror’s anti-Hispanic statements). Other common law jurisdictions have also excluded evidence of racist remarks by jurors. See generally Regina v. Mirza [2004] UKHL 2, [2004] 1 A.C. 1118 (H.L.) (appeal taken from Eng.) (U.K.) (refusing to allow testimony of a juror’s racist remarks); Pan v. The Queen, [2001] 2 S.C.R. 344 (same). But see United States v. Henley, 238 F.3d 1111 (9th Cir. 2000) (allowing a possible exception for racial statements); State v. Athora, 216 A.2d 369, 371 (N.J. 1966) (allowing evidence of racial or religious bigotry). 256. See United States v. Sherrill, 388 F.3d 535, 538 (6th Cir. 2004) (refusing a new trial after a juror slept through trial); Gov’t of V.I. v. Nicholas, 759 F.2d 1073, 1080 (3d Cir. 1985) (refusing a new trial after a juror with hearing problem gave contradictory affidavits as to whether he heard testimony); United States v. Dioguardi, 492 F.2d 70, 78–79 (2d Cir. 1974) (refusing a new trial after a juror’s post-trial letter evinced paranoia). But see Sullivan v. Fogg, 613 F.2d 465, 467 (2d Cir. 1980) (stating that evidence of juror’s mental incompetence required investigation). 257. Jorgensen v. York Ice Mach. Corp., 160 F.2d 432, 435 (2d Cir. 1947) (L. Hand, J.). LEUBSDORF_FINAL 7/19/2006 7:36:19 AM PRESUPPOSITIONS OF EVIDENCE LAW 1257 reject evidence of jury bias and incompetence on other grounds, such as concern about intruding on jury privacy and multiplying post-verdict proceedings.258 Once again, therefore, concerns about jury bias and error do not drive the rules, but are invoked only when necessary to justify practices resting on other grounds. V. CONCLUSION This Article’s goal has been to understand the presuppositions of existing law. I have tried to read those presuppositions the way an anthropologist tries to read the discourse and institutions of a culture,259 or a psychoanalyst tries to read the dynamics underlying the emotions and behavior of a patient. In all of these instances, the inquirer seeks to bring to light patterns, of which those studied are only partly aware.260 In this case, of course, as in almost all legal scholarship, the author is himself one of the natives and neurotics and does not claim to have freed himself from the presuppositions that he analyzes or to be able to propose a complete new evidentiary scheme. Examining the presuppositions of evidence law may nevertheless be useful by helping to explain the resistance that has blocked most proposed innovations. The causes of this resistance would otherwise be hard to understand. At no time has evidence law been so near perfection as to preclude all sorts of improvement. Empirical research has demonstrated time and again that the traditional rules are based on false generalizations about human psychology.261 Although organized political or economic groups have struggled about such evidentiary matters as rape-shield laws and professional privileges, it would be hard to find a group whose interests would be menaced by, say, sweeping reform of the hearsay rule. Most of the resistance to change has come from lawyers, but it is hard to see how even lawyers would be significantly affected by reform, except to the limited extent that we would all have to learn the new rules.262 In some instances, 258. E.g., Munafo v. Metro. Transp. Auth., 381 F.3d 99, 107 (2d Cir. 2004) (citing McDonald v. Pless, 238 U.S. 264, 267 (1915)). The older rationale for the exclusionary rule was that jurors should not be allowed to contravene their own sworn verdicts. Vaise v. Delaval, (1785) 99 Eng. Rep. 944 (K.B.). 259. For two classic applications of this approach to the anthropology of law, see generally CLIFFORD GEERTZ, LOCAL KNOWLEDGE (1983); BRUNO LATOUR, LA FABRIQUE DU DROIT: UNE ETHNOGRAPHIE DU CONSEIL D’ÉTAT (2004). 260. For vaguely similar essays in legal scholarship, see generally Robert C. Clark, The Morphogenesis of Subchapter C: An Essay in Statutory Evolution and Reform, 87 YALE L.J. 90 (1977); Duncan Kennedy, The Structure of Blackstone’s Commentaries, 28 BUFF. L. REV. 205 (1979). 261. See, e.g., supra notes 10, 76, 193, 209, 212 and accompanying text. 262. For some questions about assertions that lawyer self-interest delayed civil-procedure reform, see John Leubsdorf, The Myth of Civil Procedure Reform, in CIVIL JUSTICE IN CRISIS: COMPARATIVE PERSPECTIVES OF CIVIL PROCEDURE 53, 59–63 (Adrian A.S. Zuckerman ed., 1999). LEUBSDORF_FINAL 1258 7/19/2006 7:36:19 AM 91 IOWA LAW REVIEW [2006] lawyers may have stood in for otherwise unorganized client groups such as criminal defendants,263 but this could not be a complete explanation. One source of resistance to change may have been unexamined assumptions such as those discussed here. If one takes it for granted that evidentiary rules will normally be applied inference by inference and will determine admissibility by assigning inferences to classes of evidence,264 any proposals for reform will probably be couched and discussed in the same terms and will therefore not be much different from existing law. If one assumes that trials will be oral events in which virtually all evidence that is not oral must be escorted into court by live witnesses,265 then, for the most part, only changes adhering to that assumption will be thinkable. And if one views evidentiary issues through a web of beliefs about the strengths, weaknesses, rationality, and emotions of jurors that derives less from reality than from the structure of the trial system,266 that same web will probably thwart and deflect one’s vision of proposed innovations. Becoming aware of these and other presuppositions of evidence law might ultimately contribute to reform, even if its more immediate fruit is increased insight into the current system. 263. See 26 CHARLES ALAN WRIGHT & KENNETH W. GRAHAM, JR., FEDERAL PRACTICE AND PROCEDURE § 5661 (1992) (describing the controversy over governmental privilege and mentioning bar-association involvement). 264. See supra Part II. 265. See supra Part III. 266. See supra Part IV.