View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Brigham Young University Law School

BYU Law Review

Volume 1998 | Issue 4 Article 3

11-1-1998 The hC aracter Evidence Rule Revisited Kenneth J. Melilli

Follow this and additional works at: https://digitalcommons.law.byu.edu/lawreview Part of the Evidence Commons

Recommended Citation Kenneth J. Melilli, The Character Evidence Rule Revisited, 1998 BYU L. Rev. 1547 (1998). Available at: https://digitalcommons.law.byu.edu/lawreview/vol1998/iss4/3

This Article is brought to you for free and open access by the Brigham Young University Law Review at BYU Law Digital Commons. It has been accepted for inclusion in BYU Law Review by an authorized editor of BYU Law Digital Commons. For more information, please contact [email protected]. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

The Character Evidence Rule Revisited

Kenneth J . Melilli*

I. INTRODUCTION A long-standing fundamen ta l tenet of the rules of evidence in American jurisdictions h as been a gen er al prohibition of character evidence for the purpose of proving an actor’s confor- mity with a character trait on a particular occasion.1 Although this character evidence r ule is n ot restrict ed to the circumstance in wh ich the prosecut ion seeks to intr oduce n ega- tive chara cter evidence concernin g a crim ina l defendan t, it is the desire t o pr otect the cr im in al defenda nt from the im pa ct of such evidence th at accounts for the firm vitality of the rule.2 Recently, the Federal Rules of Evidence have been amended so as t o permit the gover nmen t, in pr osecution s for sexu al as- sault or child molesta tion, to introdu ce eviden ce of the criminal defendant’s prior offenses of sexu al a ss ault or child moles tation , respectively.3 Th ese n ew add it ions t o the F eder al Ru les of Evi-

* Pr ofess or of Law, Albany Law School. The author grat efully acknowledges the contributions of Maya Hadjiyancheva an d Lori Fowler. 1. See Edwar d J . Imw ink elrie d, A S ma ll Contribution to the Debate Over the Prop osed Legislation Abolishing th e Charact er Evidence Prohibition in Sex Offense Prosecu tion s, 44 SYRACUSE L. REV. 1125, 1129 (1993). For example, the Federal Rules of Evidence, the formulation of the ru les of evidence in place in the substantial majority of America n ju risd ictions , id. at 112 7, p rovid e t ha t “[e]v ide nce of a person’s character or a trait of character is not adm iss ible for t he pu rp ose of p roving act ion in con for m it y t h er ew it h on a pa r t ic u la r occ a s io n , e xc ep t . . . . ” F ED. R. EVID. 404(a ). 2. S ee, e.g., Thom as J . Reed, Trial by Prop ensity: Admission of Other Criminal Acts Evidenced in Federal Criminal T rials, 50 U. CIN. L. REV. 713, 739 (1981) (“I t for[bids] the admission of other crimi na l a cts of th e a ccu se d a t t ri al becau se of th e substantial da nger of un du e pr eju dice t o th e a ccus ed a risin g from infe re nt ial pr oof of the a ccused’s disposition to commit the crime.”). In this regard, the Feder al Rules of Evidence restate the general chara cter e vidence r ule in th e context in wh ich it is of pa rt icul ar conce rn : “Evide nce of oth er crimes, wrongs, or acts is not admissible to prove th e ch ar act er of a per son in or der to s how action in conform ity t her ewith.” FED. R. EVID. 404 (b). 3. See FED. R. EVID. 413 and 414, which became effect ive on Ju ly 9, 1995. Rule 415 of the Federal Rules of Evidence, wh ich per mit s th e int rodu ction of simila r evid en ce in civil ca ses concerning sexual assa ult or child molestation, was ena cted simu ltaneou sly along with Rules 413 and 414 of the Feder al Rules of Evidence.

1547 D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1548 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 dence received substantial opposition prior to their enactment,4 as well as a great deal of criticism since that time.5 Not the least of these criticisms has been directed at the new rules’ unfounded premise t ha t sex offense cases merit ent irely differ- ent evidentiary treatment than do prosecutions for other crimes.6 But while there might be no just ifica tion for a crime-specific adjustment to the rules governing the admissibility of character evidence, it does not follow that the solution is a return to an offense-neutral proscript ion against the introduction of character evidence. I n fact, as will be dem onst ra ted shortly, it does not app ea r that any such compr eh en sive a dh er en ce t o the character evidence rule ha s ever really existed. Neverth eless, by their assa ult u pon the character evidence rule, the new federal rules have provoked a backlash in defense of this sacred cow of evidence la w. A close exa min ation of the r ation ales advanced for the preservation of the char acter evidence rule actually lead to the following conclusions. First, the character evidence rule is m ore r het oric t han subst ance; the practice in American courts belies any real commitment to excluding character evidence. An uninitiated observer of our adjudicatory system would be astounded to learn of the ch aracter evidence r ule, for cha ra cter evidence is adm itted here, there and everywhere. Secon d, when stripped of hyperbole, the standard arguments for the character evidence rule are lar gely speculative and unpersuasive. Third, th e only tru ly cogent justifications for the character evidence rule are t hose founded, to some degree or another, upon judicia l econ omy and courtroom efficiency. These concerns do not support a broad exclusionary rule, but rather can be accommodated by presumptively limiting relevant character evidence t o the for m of pr ior convict ion s. Fourth, when char acter evidence is admit ted under the current rules , it get s in under one or more of a variety of

4. S ee, e.g., R eport of th e J ud icia l Con feren ce on t he A dm iss ion of Ch ara cter Evidence in Cert ain S exu al M iscon du ct Ca ses, 159 F.R.D. 51, 53 (1995) [hereinafter Ju dicial Con feren ce]; Myrna S. Rae der , Am erica n B ar A ssoci ation C rim in al J us tice S ection R eport to t he H ous e of Del egat es, 22 FORD HAM URB. L.J . 343 (199 5). 5. See infra notes 150–163 a nd a ccompan ying te xt. 6. See infra notes 164–177 a nd a ccompan ying te xt. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1549

complex theories and excep tion s. As a consequ en ce, s uch evidence is usually accompan ied by jur y instructions intended to assist the jury in its appropriate use. H owever , be cause su ch instructions ar e often both hopelessly confusing and contrary to common sense, there is no realistic hope that they will be followed by ju rors. A simpler , m ore straightforward approach to the admissibility of chara cter evidence would alleviat e the above problems. This article will exam ine t he deficiencies in t he current app roach, a nd will recom men d s uch a simplified solution.

II. THE TRADITIONAL STATE OF AFFAIRS The essential character istic of character evidence is that it requires two steps of logical inferen ce. First, th e jur y mu st draw a conclusion from the evidence about an individual’s genera l propensity to behave in a certain way, an d second, the jury must measure the likelihood that the same individual behaved in accordance with this general propensity on the particular occasion in qu estion .7 Thus, the essential feature of character evidence is that it s r ele va nce depends upon a conclu sion abou t an individual’s gener al propensity; if t he evidence ha s relevan ce without n ecessar ily requirin g a conclu sion about general propensity, it is not cha racter evidence and is not barred by the character evidence rule.8

7. See Edwar d J . Imw ink elrie d, The Use of Evidence of an Accu sed ’s Unch arg ed Mi scon du ct to P rov e Mens Rea: The Doctr in es W hi ch T hr eat en t o En gu lf t he C ha ract er Evidence Proh ibit ion , 51 OHIO ST. L.J . 575 , 58 0–8 1 (19 90). 8. In som e ca ses, t he firs t st ep in t his pr ocess simp ly involves cred iting t he testimony of the cha ra cte r w itn ess . Th is occu rs wh en th e wit ne ss t est ifies in t he form of opin ion (th e p er son al opin ion of th e witness, based upon the witness’s personal knowledge of the in dividua l in question) or rep ut at ion (a consen sus or collective opin ion about th e individu al in q uestion, a s relayed to the jurors by the character wit ne ss ). Examples of character evidence in t hese forms w ould be t estim ony by a character witness that the relevant individual is a violent p erson (opinion) or is reput ed by members of that individual’s social circle to be a violen t p er son (re pu ta ti on). In other cas es, the first step in t he pr ocess requires an actu al inference to be made by th e jur y. This occur s whe n t he cha ra cter w itn ess t estifies in the form of one or more specific insta nces of beha vior by th e rele van t in dividua l. For exa mple, if t he character witness were to testify that the relevant individual had physically beaten his spouse on a pa rt icular occasion or occasions, t he ju ry m ight infer t he in dividua l’s general pr open sit y tow ar d viole nce ba sed up on s uch spe cific beh avi or. Ess ent ially, th e a na lyt ical dis tin ction bet wee n ch ar act er evid en ce in th e for m of D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1550 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998

As noted earlier,9 th e pr ima ry concer ns un der lying t he character evidence rule are the consequ ences of allowing th e jury to draw conclusions about the general propensity of a criminal defendant. If t he in trodu ction of character evidence against the accused in a criminal prosecution can be defended, then th e use of chara cter evidence in all other situ at ions is appropriate as well. Consequent ly, the balan ce of this art icle will focus on the most common and most dan gerous scenar io: the prosecution’s introduction of evidence of the defendant’s character. Prior to the recent addit ions of Rules 413 a nd 414 to the Federal Rules of Evid en ce, t her e a lr ea dy exist ed a va riet y of circumstances in which the government could introduce evidence of the defendant’s character. These circumstances are embodied at five differ en t points in the F eder al Ru les of Evidence. Each of these will be addr essed in tu rn .

A. Rules 404(a) and 405(a) One of the two exceptions to Rule 404(a)’s general proscript ion against the admission of character evidence is a specific allowance for evidence of the character of the accused.10 There ar e, of course, two importa nt restr ictions u pon t his exception to the character evidence rule. The first, and most important, of these limitations is that the defendant must go first; i.e., the government may not introduce character evidence regarding the defendant unless and until the defen da nt in trodu ces character evidence a bou t him self or her self.11 Wit hout qu estion , t he goa l of protecting the defendant from t he governm ent ’s negat ive chara cter evidence is

opin ion or reputation t estim ony an d char acter evidence in th e form of specific instan ces is t ha t, in th e la tt er cir cum st an ce, t he ju ry mu st ta ke th e first ste p fr om beh avi or to gene ra l pr opensity, while in the former situation, the witness (in the case of opinion t estim ony) or even pers ons n ever h ear d from dir ectly in th e court room (in the case of reputation test imony) have taken this firs t s te p for th e ju ry. In bot h scenarios, it is for t he ju ry t o dete rm ine if, and to what degr ee, the second step (beh avi or on t he occas ion i n q ue st ion i n con formity with the individual’s general propen sity) should be taken. 9. See supra note 2 and accompanying text. 10. See FED. R. EVID. 404 (a)(1 ). 11. This sequ ence is a ppar ent from t he wor ding of Rule 404 (a)(1), which perm its “[e]vide nce of a pert inen t t ra it of char acter offered by an accused , or by th e pr osecu tion to rebut t he s am e.” Id. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1551 fully advanced by providing t he defendant with t he only key to the locked door, behind which lies evidence of the defendan t’s character. But the price that the defendan t must pay to keep that door locked is the fir st demonst ration of the la w’s la ck of total commitment to the character evidence rule. In response to a false accusation, a defenda nt’s natural response would be to state, “I would not do such a thing, and the people who kn ow me will tell you t ha t.” A defendant who chooses t o call witnesses to so t estify ess en tia lly w aives t he protect ion of th e character evidence rule. There is, to be sure, not the slightest un fairness in th is arrangement. If there indeed exist s n ega tive character evidence behind the now unlocked doors, then perhaps the defendant’s own chara cter witn esses are n ot to be believed, and surely the jury is ent itled to more th an just h alf the story. Never th eless, if character evidence, and in particular, negative character evidence, is so prejudicial, why allow it even under these circumstances? It wou ld seem that eit her su ch character evidence has s ome r ea l va lu e or the rule is a less t ha n gen uin e attempt to inhibit defendan ts from in tr oducing positive character evidence concerning themselves. The second lim it ation on the in trodu ction of character evidence regarding the defendant is that such evidence must be presented in the form of opinion or reputation testimony only and not as evid en ce of specific inst ances of conduct.12 While t his

12. See FED. R. EVID. 405(a). The d istin ction bet ween opinion an d r epu ta tion testimony as contrasted with testimony as to specific instances of conduct is discussed supra at note 8. The choice made in Rule 405(a) might very well be regarded as an odd one. It requ ires th e jury to tr ust the ju dgment of the witness (or those who gossip with the witness) rather than being presented with the actu al beh av ior of th e defendant. Th e pr efer en ce for op ini on a nd re pu ta tion evid en ce is cont ra ry to t he conve nt iona l wisdom as to t he re lat ive p rob at ive s up er iority of evidence of specific instan ces of condu ct. S ee, e.g., David P. Le onard, The Use of Character to Prove Conduct: Rationalit y and Catharsis in the Law of Evid ence, 58 U. COLO. L. REV. 1, 16 & n.94 (1986–87). Moreover, reputation evidence is thought to be the least reliable and potentially the m ost dangerous form of character eviden ce. S ee H. Richard Uviller, Evidence of Character to Prove Conduct: Illusion, Illogic, and Injustice in the Cou rtr oom , 130 U. PA. L. REV. 845, 885 (1982 ). Never th eless , opting to exclude character evidence in the form of specific instances of conduct does avoid potentially time-consuming cour t b at tl es over th e a ctu al pr ior cond uct of the defendant. It is, ther efore, one of severa l specifics in th e law gover nin g char acter evidence, suggest ing that su ch e vide nt iar y rule s a re dr iven in s ub st an tia l pa rt by con cer ns for cou rt room econ omy an d efficiency. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1552 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 migh t appear t o be only a matter of form rather than substance, ther e is one fu rther asp ect of the r ule t hat deserves atten tion . On cross-examination of a character witness, questions are perm itted as to specific instances of the defendant’s conduct.13 Thus, for example, the character witness who testifies that the defendant is, or is reputed to be, a peaceful person may be cross-examined as to specific instan ces of violent or aggr essive behavior of th e defendant. The theor y u nderlying t he a llowance of su ch in qu ir ies in to sp ecific in st ances of condu ct is not that such questions and answers are admitted to prove the defendant’s character, but rather are admitted to enable the jur y to deter min e what weigh t, if a ny, s hould be give n to the testimony of the character witness.14 If the positive character witness does n ot know or ha s not heard of the nega tive behavior of the defendant, then the character witness arguably does not truly know of t he defenda nt’s character or reputation ; if the pos it ive cha racter wit ness d oes kn ow or has h ea rd of t he negative behavior of the defendant, then the positive opinion or reputation testimony of the character witness is pr esu ma bly of lesser or no value.15 Such inquiries on cross-examination are typically accompa nied by a limit ing in stru ction directing th e jury to consider such questions and answers, not as nega tive evidence of th e defenda nt’s character , bu t rather as evid en ce which, in the judgmen t of the jury, can only be used in determining wh at weigh t, if a ny, t o ass ign to the character witness’s testimony of the defendant’s positive character.16 Here we have the first of several examples of the law’s theoretical observance of the character evidence rule by the device of a likely ineffect ual jury instruction. The actual evidence presented to the jury on cross-examina tion—specific instances of ba d con du ct in the context of testimony as to the defendant’s good character—has all th e earmarks of character evidence. The defendant is said to be a peaceful, nonviolent person, and it comes to light that he or she has physically assaulted several individuals on several prior occasions. That

13. See FED. R. EVID. 405 (a). 14. See 2 J ACK B. WEIN STE IN E T AL., WEINSTE IN’S EVIDENCE ¶ 405[04], at 405–50 (199 6). 15. See EDWARD W. CLEAR Y ET AL ., MCCORMICK ON EVIDENCE § 191, at 569 & n.28 (3d e d. 1 984 ). 16. See 2 WEIN STE IN E T AL., supra note 14, ¶ 405[04], at 405–50 to 405–51. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1553 this is affirmative evidence of a viole nt pr opensity ca nnot seriously be doubted, and, more to the point , it is im probable that the ju ry will con fin e it s con sideration of this evidence to its nominal noncharacter perimeters.17 In fact, one of t he t wo t heories on which such cross- examination is permit ted—that the value of the character witness’s testimony is diminished because the character witness is unaware of the negative behavior—depends necessarily upon the actual existence of such behavior on the part of the defendant. It is unremarkable tha t a character wit ness is u nawa re of an event that did not actually take place. Therefore, when the character witness who has testified as to the defendant’s nonviolent character admits on cross- examination to ignorance concerning the defendant’s prior assaults, there can be no diminution of the character witness’s testimony unless the jury accepts t hat ther e r ea lly were such incidents of prior assaultive behavior on the part of the defendant. This ren ders it all the more improbable that the applicable limiting instruction will accomplish its stated purpose. The jury, having concluded that the defendant has engaged in assa ult ive conduct, is now instructed not to consider this as affirm ative evidence of the defendan t’s character even as it resolves the very qu estion of the defenda nt’s character for violence or peacefulness. The unlikelihood that the jury will both comprehend and obey the limit in g in st ruction atten da nt to this evidence is not prem ised upon a pejorative assessment of either the intelligence or the good faith of jurors. The problem lies not with the jurors, but with the instruction itself. Lawyers are prone to create theoretical distin ction s t hat are eit her illusor y or im pos sible to app ly p ractically, especia lly by ju rors w ho are not participants in the intellectual hair-splitting activities that spawn such distinctions.18 Children can pretend th at th e family dog is a lion for the purposes of a party game, but the lat e- arriving child not involved in the gam e can hardly be blamed for gazing at the animal and pronouncing, “It’s just a dog.” Although even a thorough indoctrination in the reasons

17. See 2 id. ¶ 405[04], at 405–51. 18. See Abra ha m P . Ord over, Balancing the Presumptions of Guilt and In nocen ce: Rules 404(b), 608(b) and 609(a), 38 EMORY L.J . 135 (198 9). D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1554 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 underlying the limit in g in st ruction migh t not advance t he goa l of the instruction, the absence of any explan at ion virt ua lly insures failur e. So why does the law continue to permit what appear s to be character evidence to reach the jury, shielding the defendan t with only a feeble jury cha rge r estricting considera tion to its noncharacter use? Sur ely it is not an unwarranted faith in such an in st ruction ; the ca se for the ch aracter evidence prohibition is lar gely prem ised upon the cla im ed in effectiveness of instructions confin in g ju ror consideration of such evidence t o a non-prejudicial scop e.19 Per ha ps t he explana tion is tha t, wh ile the limiting instruction allows u s t o wor sh ip at the a lt ar of th e character evidence rule, in realit y we are not truly persuaded of the rule’s worth. Perhaps we are not terribly both ered by t he admission of what has all the appea rance of what the character evidence rule categorizes as prejudicial evidence because we believe, or at least suspect, that such evidence, even used “improperly,” is really quit e relevant .

B. Rule 404(b) The circumstances governed by Rules 404(a ) and 405(a) are by no mea ns t he on ly s it uation in wh ich this phenomenon occurs. Wit hout a dou bt , t he m ost sign ifica nt area of the la w of evidence in which this occurs is the well-ensconced rule, embodied in the second senten ce of F eder al Ru le of E vid en ce 404(b), 20 allowin g the int roduction of specific instances of bad conduct, having all the indications of character eviden ce, for what the law deems a noncharacter purpose. To illustrate, suppose that the defendant is on tria l for a bank robber y in which the actor crea ted a diversion by placing an explosive in the automatic teller machine and then, posing as an employee of an indepen dently contracted clea nin g ser vice, u sed a va cuum cleaner to collect the cash from the abandoned st ations of t he distracted tellers. In the like ly even t that the id en tit y of the

19. S ee, e.g., Imw ink elrie d, supra note 1, at 1140–41, 1145–46. 20. After resta ting th e general pr oscr ipt ion em bodi ed i n t he cha ra cte r e viden ce rule, th e rele van t sect ion of Rule 404(b) provides t ha t “[e]vidence of other crimes, wrongs, or acts . . . may, however, be admissible for other purposes, such a s pr oof of motive, opportunity, intent, preparat ion, plan, knowledge, ident ity, or a bse nce of mistake or accident .” FED. R. EVID. 404 (b). D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1555 perpetrator is a contested issue at trial, the governmen t would almost su rely be permit ted to in trodu ce evidence that the particular defendant had used the same modus operandi to rob other banks on other occasions. This evidence would be allowed despite the fact that the defendant is not charged with any but the first-mentioned robbery, and despite the fact that the evidence itself has all the qualities of the supposedly dreaded character evidence, i.e., specific acts (prior bank robberies) indicating a general propensity (defendant robs banks) used as evidence of behavior on the occasion in question (robbed t his bank). The theory upon which the a dmissibilit y of such eviden ce is allowed is that it ha s a legitima te noncharacter relevance (in the example above, to prove the robber’s identification by th e dist inctive method used) that is distinct from the use of such evidence to demonstrate a general propensity. However, this dist inction exist s in pr in ciple on ly. On e ca n analyt ica lly separate how the defendant robs banks from the fact that the defendant robs banks, but in reality one cannot learn how the defendant robs banks without being informed that the defendant robs banks. The evidence of the defendant’s character is admitted, and it is only to the extent that we are successful in persu ading the jur y to view it otherwise that such evidence could be deemed to be nonchara cter evidence.

1. The 404(b) prerequisites To be sure, a dm ission of su ch evidence is not automatic. First, the proponent must give notice t o the a ccused u pon request.21 Second, the trial court must be satisfied that a rational jury could find that the uncha rged misconduct did occur.22 Third, the trial court must be persuaded that there is a legitimate, noncharacter purpose to such evidence.23 Fourth, the trial court must balance the usefulness of the evid en ce against th e potentia l prejudice of the ju ry u sing such evidence to determine the gener al propensity, or cha racter , of the

21. See id. 22. S ee, e.g., H ud dle st on v. U ni te d S ta te s, 4 85 U .S. 681 , 68 9–9 0 (19 88). 23. See CLEARY ET AL ., supra not e 15 , § 190 , at 558; 2 WEIN STE IN E T AL., supra note 14, ¶ 404[08], at 404–44 to 404–45. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1556 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 accused.24 In making this final determina tion, the court must consider that such evidence is, seemin gly wit hout exce pt ion , accompanied by a limiting instruction juxtaposing the permissible noncharacter use and the forbidden character treatment of the evidence.25 In practice, these requirements are often satisfied. The first is solely procedural a nd limits only those who are simply unprepared for trial. The second r equirem ent is a nomina l requirement that ther e sim ply exist su fficient evidence t o raise a jury question as to the occurrence of the prior bad acts.26 The third prerequisite to admissibility u nder Rule 404(b) migh t have been a significan t lim ita tion if the permissible noncharacter uses of such evidence had been na rr owly circumscribed. But th is has not been th e case. The decision s resolving Rule 404(b) issues have been quite liberal in sustaining the theories of admissibility advanced by prosecutors.27 In deed, t he fr equen cy with wh ich the excep tion s to Rule 404(b)’s general ru le of exclusion ar e successfully invoked has pr ompt ed some comment at ors t o suggest t ha t it is the exclusion of such evidence that is truly the exception rather than the rule.28 The practical impact of Rule 404(b) m ust be underst ood n ot only in th e proportion of cases in which these issu es are resolved, but also in the quantity of such cases in which these issues materialize. Rule 404(b) accounts for a greater number of published judicial opinions than any other provision in the

24. See 2 WEIN STE IN E T AL., supra note 14, ¶ 404[08], at 404–45. 25. See Imw ink elrie d, supra note 1, at 1126. 26. In fact, even in a crimin al case in wh ich th e eviden ce is offered aga inst th e accused, th e gove rn me nt ne ed on ly sa tis fy th e cou rt th at th e ju ry could find that the pr ior bad acts occurred by a preponderance of the evidence. S ee Hu dd lest on, 485 U.S. at 689–90. Cons equent ly, even condu ct which ha s been th e sub ject of a prior acquit ta l (representing only the failure of the governm ent’s evidence to reach the higher th reshold of pr oof beyon d a re as ona ble doubt) may be intr oduced in a subsequen t pr osecu tion un der Rule 4 04(b). S ee Dow lin g v. U ni te d S ta te s, 4 93 U .S. 342 (199 0). 27. See Raed er, supra not e 4, a t 3 48–4 9; cf. Edward J . Imw ink elrie d, Some Comments About Mr. David Karp’s Remarks on Propensity Evid ence, 70 CHI.-KENT L. REV. 37, 39 (1994) (reporting that both sides of the debate feel cour ts a re “un duly solicitous” to t he oth er sid e). 28. See 2 DAVID W. LOUISELL & CHRISTOPHE R B. MUELLER, FEDERAL EVIDENCE § 136, at 132 (1985); Glen We issen berger, Making Sense of Extrins ic A ct E vid ence: Federal Ru le of Evidence 404(b), 70 IOWA L. REV. 579 , 57 9-80 (198 5). D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1557

Federal Rules of Evidence,29 and the in trodu ction of evidence of uncharged crim ina l conduct under Rule 404(b) has appa ren tly increased subst antially since 1975, when the F eder al Ru les of Evidence were enacted.30 Th e volume of ca ses in wh ich courts are both asked t o admit evidence under Ru le 4 04(b) a nd in fact grant such requests has led to the popu lar conception that crea tive prosecutors will usually be successful in genera tin g a theory for introdu cing evidence of the defenda nt’s prior , un charged misbeh avior before t he jury.31 That part icular piece of plebeian wisdom is undoubtedly overstated, but the fact remains that, in practice, Ru le 404(b)’s gate swings open wid er and m ore oft en than one m igh t exp ect , given the constant and vociferous incantation of the character evidence rule. Some have suggested that the problem, if it is a problem, can be tra ced to the nonexclusive lan guage of Rule 404(b)’s second sen ten ce. S pecifica lly, t he r ule’s provis ion that evidence otherwise inadmissible under the ch aracter evidence rule “may . . . be admissible for other purposes, such as . . .” followed by a list of noncharacter theories of admissibility, makes clea r that this lis t is in ten ded t o be illust rative, a nd n ot exhaustive.32 Ra ther than viewing t he Rule 4 04(b) list of theories of admissibilit y as a closed universe of possibilities, the prevailing interpretation of Rule 404(b) is that, subject to the other requirem ents specified above,33 evidence of prior misconduct is admissible un less it s relevan ce depen ds solely upon the defenda nt’s gen er al propensity t o commit the charged crime.34

29. See Imw ink elrie d, supra note 7, at 577. Professor Imwinkelried has characterized “[t]he ad mi ssi bilit y of un char ged mis conduct evidence [as] the single most im po r t a n t is s u e i n con t em po r a r y c r im in a l e vi de n ce la w. ” Id. at 576 (citing to Edward J. Imw ink elrie d, Un cha rged Miscon du ct: On e of th e Mos t M isu nd erst ood Iss ues in Cri m in al E vid ence, CRIM. JUST., Su mm er 198 6, a t 6 , 7). 30. See Ord over, supra note 18, at 142. 31. See Weisen berger, supra note 28, at 579. 32. See Imw ink elrie d, supra note 1, at 1136. 33. See supra notes 21–25 a nd a ccompan ying te xt. 34. See 1A J OHN HEN RY WIGMORE, EVIDENCE IN TRIALS AT COMMON LAW § 54.1, at 1151–52 (1983); Office of Legal Policy, Rep ort t o the A tt orney General on the Ad m iss ion of Criminal Histories at Trial, 22 U. MICH. J.L. REFORM 707, 719 (1989) [hereinafter Ad m iss ion of Cr im in al H ist ories ]; Thomas J . Reed, Admiss ion of Ot her Criminal Act Evidence After Adoption of the Federal Rules of Evidence, 53 U. CIN. L. REV. 113, 115, 158, 161 (198 4). D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1558 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998

This inclusionary approach stands in contrast to the “traditional,” exclusionary rule that would only admit what would otherwise be char acter evidence if its logical rele va nce fell wit hin a lis t of well-d efin ed exce pt ion s. 35 It has been argued that the exclusionary for mula tion of th e chara cter evidence rule is the only approach that is faith ful to the origins of the rule, bot h in the com mon law of t he United States 36 and, even earlier, in the com mon law of E ngland.37 The arrival of the in clu sionary approach of Rule 404(b) in 1975 is supposed to have reversed the then-pr eva iling exclu sionary a pp roach,38 resulting in an objection able exp ansion of other crimes evidence introduced against criminal defendants.39 Presumably, then, the lar ge volume of other crimes evidence that ma kes its way into American courts pursuant to Rule 404(b) and its progeny is simply a recent mistake, represents no failure of the character evidence rule in principle, and can be corrected simply by a return to the traditional, exclusionary character evidence rule. However, neit her history n or rea son allows t he liberal u se of Rule 404(b) by pr osecut ors t o be expla ined away so easily.

2. Historical absence of an exclusionary rule First, any claim that the character evidence ru le developed as a gen era l ru le of exclusion is sim ply short sighted. The per vasive belief that the common law ordinarily disallowed evidence of simila r crim es is actually r efu ted by t he h istorica l recor d. 40 The chara cter evidence rule in this country has actually undergone several t ra nsforma tions du rin g its growth.41 The rule in the U nit ed St ates wa s or iginally ba sed u pon the then-existing practice in England.42 Neither the early English cases43 nor the early English text wr iters 44 specified any general

35. See Reed, supra note 2, at 713. 36. See Thom as J . Reed, The D evelop m ent of th e Propensity Rule in Federal Criminal Causes 1840–1975, 51 U. CIN. L. REV. 299 , 30 2, 3 03, 322 , 32 5 (19 82). 37. See Reed, supra note 2, at 717. 38. See Reed, supra note 34, at 113. 39. See Reed, supra note 2, at 713–14. 40. See J uliu s St one, The Rule of Exclusion of S im ila r A ct E vid ence: A m erica , 51 HARV. L. REV. 988 , 98 9–9 1 (19 38). 41. See J uliu s St one, The R ul e of E xclu sion of S im ila r A ct E vid ence: E ngland, 46 HARV. L. REV. 954 (193 3). 42. See Ston e, supra note 40, a t 991 ; Ston e, supra note 41, at 954. 43. See id. at 960–65; 1A WIGMORE, supra note 34, at 1213 (“In early English D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1559 rule of exclusion of similar-act evidence. The a dm issibilit y of specific acts of conduct was simply det erm ined by th e ordin ar y test of relevan ce, and a gener al exclusionary r ule would, therefore, actually be u nfait hful to the h istorical roots of th e law’s approach to evidence of similar bad acts.45 By th e nin eteenth cent ur y, the rule h ad d eve loped to the extent that it essentia lly pr eviewed the cu rren t in clusionary formulation of Rule 404(b). Specifically, if the eviden ce in question wa s r ele va nt for any purpos e ot her than, or in add it ion to, a suggestion of a genera l propensity to commit th e crimes, it was adm issible ; exclusion of su ch evidence was t he excep tion al result, warranted only wh en the evid en ce h ad n o rele va nce other th an to dem onst ra te a gener al crimin al propen sity.46 It was in this form that the r ule r ega rding eviden ce of prior crimes made its way into the American courts. The early Amer ica n cases followed the British inclusionary ru le, admitting su ch evidence excep t in circumstances in which it would sustain no relevant conclusion except to establish a genera l propensity.47 The purposes for which such evidence was admitted under this in clu sionary a pp roach in clu ded, among other things, proof of knowledge, intent, an d motive.48 This list is manifes tly t he h istorica l or igin of the cu rren t catalogue of in clu sionary purposes for which evidence may be admitted under Ru le 404(b). In th e lat ter ha lf of the nin et eenth cen tury, cou rts in bot h England and the United States shifted to an exclusionary approach.49 Exclusion of character evidence to show general propensity became the stated ru le, and the categories of admissibility which had developed under the earlier,

practice, [evidence of bad acts to show the character of the accused] was resorted to without limit at ion.”). 44. See Ston e, supra note 41, at 958–60. 45. See T.R.S. Allan , S im ila r Fa ct E vid ence a nd Disposition: Law, Discretion and Admissibility, 48 MOD. L. REV. 253 , 27 4 (19 85). 46. See Ja son M. Brauser , Intri ns ic or E xtr in sic?: Th e Confu sin g Dis tin ction Between Inextricably Intertwined Evidence and Other Crimes Evidence Under R ule 404(b), 88 NW. U. L. REV. 1582, 1592 (1994); Ston e, supra note 41, at 965–66. Ess ent ially, the r ule raised t he bar for character evidence by requir ing tha t it be more tha n mer ely relevant; i.e., that, because of the danger of prejudice, it be especially rele van t t o some p ar ticula r poin t. Id. at 983–84. 47. See Bra use r, supra note 46, a t 159 3; Ston e, supra note 40, at 991–1000. 48. See Ston e, supra note 41, at 966. 49. See Bra use r, supra note 46, a t 159 3; Ston e, supra note 41, at 966. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1560 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 in clu sionary formulat ion became the stated exceptions.50 Despite this reversal in formulation, it is doubtful that it had any practical consequences.51 Ca tegor ies for wh ich su ch evidence could be relevant (other than to show general propensity) continued to develop; the only distinction was that these categories were added to a list of exceptions to what was th eoret ically a general rule of exclusion, instea d of simply being included among an illust rative list demonstrating a general rule of admissibility. Notwithst anding a change in how the “rule” was articulated, the list of situation s in wh ich evidence of uncharged misconduct was found to be relevant continued to expand and develop.52 By the t im e of t he a dop tion of the F eder al Ru les of Evidence, most court s wer e art iculatin g th e ch aracter evidence rule using t he exclu sionary for mula tion .53 But, semantic distinctions aside, the r e-adopt ion of inclusiona ry language in Rule 404(b), like the nineteenth century switch from the in clu sionary to the exclusionary approach, did not give rise to any significant change in t he admissibilit y of such evidence. Even under the prerules common law formulation, the exclusionary rule was avoided easily and often.54 Shifting back

50. See Ston e, supra note 41, at 975–76. 51. See id. at 976. 52. See id. Any other result would have been remarkable. Why would, or should, the la w s us ta in in pe rp et ui ty th eor ies of relevance th at ha d a risen in ca ses up to a particular mom en t i n t im e, b ut per ma ne nt ly s hu t t he door to t he ide nt ica l type of development of theorie s of relevan ce in un precedented circumstances in future cases? 53. See 22 CHARLES ALAN WRIGHT & KENNETH W. GRAH AM, J R., FED ERAL PRACTICE AND PROCEDURE § 523 9, a t 4 32 (1 978 ). 54. See 22 id. § 5239, at 428. Even Wigmore describe d th e moder n comm on law character evidence rule in inclusionary terms: The usual black-letter description of the character evidence rule found in the literature and in codes tod ay . . . sim ply a sse rt s t ha t it is im pr oper for a tr ier of fact to d et er mi ne wh at a p er son did on a particular occasion by asking and an swe ring th e qu est ion of w he th er th at per son ha s t he sor t of dis posi tion that would incline him toward t he doin g or not -doin g of t ha t a ct. Phr ased in this wa y, the ru le seem s at first s ight to ha ve a wide am bit. Modern an al ysi s, h owev er , give s t he cha ra cte r evidence rule a rela tively restr icted mea nin g. It is ge ne ra lly s aid th at th e pr open sit y rule does not forbid an y use of e vide nce th at re flect s ba dly on th e ch ar act er of a per son , and ther e is no such pr ohibition even when it happens that the acts of the per son shown ar e not those that are our ultima te conce rn in the la wsuit in qu est ion but ser ve on ly a s a ba sis for a scert ai ni ng som e ot he r m at te r of fa ct that does ha ppen to be of more im mediate concern. The rule against character evid en ce, it is t hou gh t, pr ohi bit s t he us e of ev ide nce th at re flect s D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1561 to th e inclu siona ry a ppr oach in Rule 404(b) merely states the rule in a manner more faithful to its historica l origins, and perhaps states it in a manner more in conformity with actual practice. Regardless of the tru e his torical deriva tion of the ru le, th e attribution of such extraordinary significance to the semantic dist inction between th e inclu siona ry an d exclusionary approaches also fails to withstand reasoned analysis. One must keep in mind that, regardless of t he form of the rule, the dist inction present in both Ru le 4 04(b) a nd its p redecess or common law rules between in adm issible “propensity” evid en ce and admissible “nonchara cter” evidence only makes sense if that distinction is real. There must be an articulable, comprehensible boundary between the two; otherwise the plethora of motions, objections, judicial opinions and liter ature on th e subject is t ru ly the modern , legal equivalent of the emperor’s new clothes. If the distinction between wha t is admissible noncharacter evidence and inadmissible character evidence is rea son ably clea r, t hen wh at pos sible differ en ce ca n it make which is labeled the rule and which is labeled the exce pt ion ? Ea ch case sh ould come out correctly regardless of which half of the dichotomy enjoys the status of “the rule.” A glass that is filled half way and a glass that is filled completely and then emptied by half both contain the same quantity at the end of the process. One possible respon se to this point is that the exclu sionary form of th e r ule, by r estricting the t heor ies of releva nce t o a specified, closed list of possibilities, will result in th e exclusion of other crimes evidence in cases wher e admissibility would have been prem ised u pon a th eory of relevance not included in the tr aditiona l list embodied in the second sent ence of Rule 404(b). The flaw in this respon se is em pirica l. An exa min ation of the annotations following Rule 4 04(b) r eve als that decision s admitting evidence u nder Ru le 4 04(b) do so, alm ost invar iably,

on th e char acter of a pers on whe n a nd on ly when th at evidence is offered in order to s how th at th e per son a cted in a cert ain way on a par ticula r occas ion becau se he ha d a pr ope ns it y or in clin at ion to a ct t ha t w ay . If t he evidence th at re flect s ba dly on th e ch ar act er of a per son is offer ed for any other purpose, it does not offend the prohibition against the us e of ch ara cter evidence. 1A WIGMORE, supra note 34, § 54.1, at 1151-52. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1562 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 under one of the specified theories of relevance. It seems that after several cen turies of exp er im en tation , t he u niverse of possible “noncharacter” theories of rele va nce h as becom e substan tia lly closed, t he in clu sionary formulation of the rule notwithsta nding.55

The rea l a nsw er to the object ion stated above, and in tu rn the explanation for the attention devoted to what ought to be a tr ivial distinction as to form, is that the boundary between what Rule 404(b) and like rules allow and disa llow could ha rdly be less clear. The distinction required by Rule 404(b) has been chara cterized as so per plexing56 and vague57 that there is litt le

55. One nota ble exception to this observation is United S tates v. Procopio, 88 F.3d 21, 29–30 (1st Cir . 1996), cert. d eni ed, 117 S. Ct . 620 (1996), and cert. denied, 117 S. Ct. 1008 (1997). In Procopio, several defenda nt s we re cha rged w ith var ious offenses, including conspiracy, arising out of the armed robbery of an armored truck in 1991. Th e govern men t wa s ab le to in tr oduce evid ence of gun s, ha ndcu ffs, a state police uniform and badge, and a police scanner seized in 1993 from an apa rtm ent wh ich was the r esidence of one defendant and th e tempora ry residence of another defendant. Over the objection of these defendants that this was imper miss ible character evidence becaus e it revealed a p rop en sit y to com mi t r obbe ries, th e cou rt s allowed th e evide nce on t he th eor y th at it d em ons tr at ed a “crim ina l as socia tion ” between th ese t wo defenda nt s. Specifically, th e th eory of admissibility was that the ass ociation of th ese tw o defe nd an ts in 1 993 w as som e evide nce of th eir as socia tion two years ea rlier. The d efendants ad mitt ed their association with each oth er in 1991 in the m onths following the robbery, but th e cou rt s r ule d t ha t t he 1993 as socia tion inferred from th e ph ysi cal evid en ce h ad sp ecia l r ele va nce beca us e it, unlike the admitt ed 1991 association, was a criminal association. Had the case involved the sam e charges against a single defendant and the physical evidence of other similar crimes been found in that defendant’s apartment two years after th e date of the charged offense, t he evidence would pre sum ably h ave been excluded a s ina dmis sible cha ra cter e vidence. Th e only dist inction b etween t his hypothetical scen ar io an d t he act ua l fact s of Procopio is t ha t t he lat te r s itu at ion involved multiple actors. E vidence su ggestin g th e gene ra l propen sity of a sin gle individual to commit robberies may be char acter evidence which is inadmissible under the first sen ten ce of Rule 404(b), but apparently the same evidence suggesting the general propensity of two individuals to commit robberies together is noncharacter evid en ce ad mi ssi ble u nd er th e se cond sen te nce of Rule 404(b). What t he cour t labels as evidence of a crimin al association m ight jus t a s we ll be des ign at ed a s gr oup character evidence. Argua bly, an oth er excep tion to t he obse rvat ion in the text i s t he “doctr ine of chan ces,” discussed at infra notes 63–65 and accompanying text. As will be seen, however, all or vir tu ally all of th e “doctrine of chan ces” cases ha ve been , or could ha ve bee n, p lace d wit hin one or m ore of th e t ra dit iona l categor ies of adm issibility catalogued in Ru le 4 04(b ). 56. See 2 WEIN STE IN E T AL., supra note 14, ¶ 404[08], at 404–44. 57. See David J. Kar p, Evidence of Propensity and Probability in Sex Offense Cas es an d O th er Ca ses, 70 CHI.-KENT L. REV. 15, 35 (1 994 ). D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1563

consistency in th e decisions of courts at temptin g to apply t he rule.58 Courts often ha ve difficulty identifying an d applyin g th e distinction between the admissible and the ina dmissible.59 The confusion is ea sy to understand. The example chosen earlier, rega rding proof of iden tifica tion by means of a unique modus operandi, is actually one of the relatively comprehensible exa mples of allowable Ru le 4 04(b) evid en ce. F or most other exam ples, t he distinction requir ed by Rule 404(b) is far more obscure.60 The dist inction is pa rticu la rly elu sive, if n ot illusive, in cases in which the evidence is admit ted to prove some level of scien ter , such as k nowle dge, a bsen ce of mistake or accident and, in particular, intent.61 There are countless examples of the in trodu ction of evidence of prior ba d a cts, osten sibly admitted, not to show a propensit y t o engage in the sp ecified conduct, but rather to demonstrate the intent to engage in the specified con du ct;62 this distinction is, at best,

58. See id.; Kar p, supra note 57, at 34-35. 59. See 22 WRIGHT & GRAHAM § 5239, supra note 53, at 431. This difficulty, in the view of some comm ent at ors, h as led some cour ts in some cas es to a dmit evidence that prope rly sh ould h ave be en e xcluded . S ee, e.g., Imwink elrie d, supra note 7, at 578; see also Procopio, 88 F.3d at 29–30. 60. S ee, e.g., 1A WIGMORE, supra note 34, § 54.1, at 1152, 1155–56; Richard B. Kuhns, The Prop ens ity to M isu nd erst an d t he C ha ract er of S pecif ic A cts Ev id ence, 66 IOWA L. REV. 777 (1981); Lee E . Teit elbau m et al., Evalua tin g th e Prej ud icia l E ffect of Evidence: Can Judges Identify the Im pact of Improper Evidence on Juries?, 1983 WIS. L. REV. 1147 (1983); Uville r, supra note 12, at 882. 61. See 1A WIGMORE, supra note 34, § 54.1, a t 115 2 n.1; Or dover, supra note 18, at 147, 157. 62. For example, in Car son v. P olley , 689 F .2d 562 (5t h Cir . 1982), a civil rights act ion against law en force me nt officia ls for th e u se of exce ss ive for ce in th e con du ct of an ar re st , th e pl ain tiff a ppe ale d fr om a jud gm en t e nt er ed on a ju ry ver dict for t he several defenda nt s. On a ppea l, the court he ld t ha t it wa s r eve rs ible er ror for t he dis tr ict cour t t o ha ve exclu ded a p er form an ce eva lua tion report regarding one of the defendants, containing a comment that the defendant “needed t o ‘work on contr olling [his] te mp er an d p er son al feelings ,’ becau se he ‘tends to get int o argu men ts wit h inmates, lets his t em pe r fl ar e up too quick ly.’” Id. at 571. The court ru led th at th is report “tended to show [the defendant’s] intent to do harm to [th e p la in ti ff] when booking him at t he jail, and, therefore, was admissible under th e ‘int en t’ excep tion to the general rule against chara cter eviden ce.” Id. at 572. Apparent ly, prior incidents of angry en counters with inmat es is not evidence of a ge ne ra l dis pos it ion to s uch encounters, but is evidence of a specific intent to harm an individual that the defendant had n ot yet met a t th e time of the incidents forming the basis for the report. In United S tates v. Hadaway, 681 F.2 d 21 4 (4t h C ir. 198 2), a pr osecu tion for aiding and abetting the th eft of televisions from an interstate shipment, evid en ce of the defenda nt ’s crim inal involvement in three subsequent thefts of interstate shipments of, respectively, 1,000 cases of tuna fish, Goodyear tires, and a tra iler D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1564 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 extr emely su bt le. Evid en ce of prior , sim ila r crim es is routinely adm itted despite t he almost imperceptible distinction between, for exam ple, t he propensity to sell drugs an d the in ten t t o sell drugs.

containing pipe t obacco an d wome n’s boots, wa s ad mit ted , not to show a general propensity to commit such crimes, but to show that the defendant possessed the requisite “in t en t , k n ow le dg e, m ot iv e, pl a n , a n d p r ep a r a t io n ” t o aid and abet the comm iss ion of the e ar lier, ch ar ged offense . Id. at 219. In United States v. McCollum, 732 F.2d 1419 (9th Cir. 1984), a prosecution for attempted ba nk rob ber y in wh ich th e defendant claimed to ha ve acte d un der h ypnosis, t he cour t r uled that e vid en ce of a twelve-yea r-old armed robbery would have been admissible, not to establish the character of the a ccused, bu t r at her to pr ove the defenda nt ’s intent to commit th e charged offense. In Uni ted S ta tes v. P ollock , 926 F.2d 1044, 1047–49, (11th Cir. 1991), a pr osecu tion for va rious dr ug offens es i nvolvin g a v er y la rge qu an tit y of cocain e fou nd secreted in a car dr iven by t h e d ef en da n t , t h e c ou r t , a ft er a ck n ow le dg in g t h a t “t h e margin between [propen sity a nd in ten t] is n ot a b righ t lin e,” Id. at 1048, ruled that the defendant ’s five-year-old conviction for conspiracy to import ma rijuan a was adm issible to prove conspiratorial intent. In United S tates v. Parziale, 947 F.2d 123 (5th Cir. 19 91), a cas e in which th e defen dan t wa s cha rged with , inter alia, conspirin g to import, and att empting t o import, mar ijuana, evidence of the defen da nt ’s pr ior attem pt to import ma rijuana was adm issible on the issue of the defendant’s intent, an issue sufficient ly raised solely by t he de fenda nt ’s plea of n ot guilty. In United S ta tes v. Brown, 34 F. 3d 5 69 (7 th Cir . 199 4), a dr ug pr ose cut ion in wh ich th e defendant did not dis pu te int en t b ut ra th er clai me d n ot t o have been involved in the attem pted purcha se of cocaine from an under cover police officer which was the subject of the cha rged offen ses , th e cou rt su st ain ed t he ad mi ssi on of t est im ony a s t o th e defendant’s involvement in four pr ior s ale s or pu rch as es of d ru gs, p ur por te dly for th e establishment of the defendant’s intent. In Un ited S ta tes v. H ern an dez , 84 F.3d 931 (7th Cir. 1996), the defendant depart ed from a n a irpla ne a nd r etr ieved a s uit case la ter found t o contain large quan tities of cocaine and heroin, leading to charges of possession with intent t o distribute both su bst an ces. At t ri al , t he govern me nt wa s p er mi tt ed t o in tr odu ce evid en ce th at , on a pr ior occa sion , the defen da nt wa s st oppe d en te ring th e cou nt ry in an automobile con ta inin g a l ar ge qu an tit y of ma rijua na . Th e a ppe lla te cour t affirmed the defendant’s conviction, ruling th at t he pr ior crime evidence was admissible, not to show the defendant’s general propensity to deal drugs, but rat her his int ent to do so. The cour t r eason ed th at th e large quantity of marijuana seized on th e pr ior occas ion a llowed t he infer ence of an in ten t t o dist ribu te t ha t m ar ijua na , wh ich in turn allowed the inference that the defendant was still of a mind to distribute wh en he wa s fou nd wit h d iffer en t d ru gs on a d iffer en t occa sion. See id. at 935. Of course, t he in itial in feren ce in th is chain of reason ing wa s ava ilable dir ectly from the lar ge quant ity of cocaine and heroin found in th e suitcase in question. Mor e pers ua sive is th e court ’s admission of th e eviden ce on an “absence of mist ake” theory, i.e., to refute any claim that the defendant was an ignorant courier. In any event, the dis tin ction between “the defendant is a drug dea ler because he has done it befor e,” and “the defendant intended to distribute these drugs because he has done it befor e,” remains paper thin. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1565

3. The doctrine of chances Perhaps the r ea l hinter la nd of Rule 404(b) metaphysics is the doctrine of chances. Th e doctrin e is pr em ised on the improbability of multiple coincidences.63 In ma ny cases, the doctrine is used to introduce evidence r elevant to pr ove scienter;64 su ch evidence would ther efor e fall within one or more of the specified, permissible categories of Rule 404(b), such as inten t, knowle dge, or absence of mistake or accident. For example, suppose that an individual charged with receipt of stolen property defends on the claim that he or she was unaware of the stolen character of the property. It would be a most unlikely coincidence if in fact the defendan t wa s in innocent pos sess ion of st olen pr operty on multiple occasions. Therefore, evidence of such other occasions could be admissible under the doctrine of chances to pr ove t he r equisit e le vel of mens rea. In other cases, the doctrin e ca n be used t o establish actus reus element s of the charged offense. Suppose, for example, a provider of childcare ser vices is ch arged wit h ph ysica l a bu se of a child and defends on th e grounds tha t t he child’s injuries resulted from an accidental fall and not from any assault by the defendant. Evidence that a significant number of other children had also sustained an unusual rate of similar in juries wh ile in the defendant’s care could be admissible u nder the doctrin e of chances to pr ove a ss ault ive condu ct by t he defenda nt. Such evidence wou ld seem to fit squarely wit hin the “absen ce of mistake or accident” category of permissible evidence under Rule 404(b). If, in the same circumstances, the defense was instead premised upon t he possibility of th e injuries ha ving been inflicted by an other a ssa ilant, the same evid en ce of similar injuries suffered by other children in the defendant’s care could be in trodu ced, consistent with Ru le 404(b), to prove iden tit y. The doctrine of chances, th en, is often simply a different way of articulating how eviden ce m igh t come wit hin one or more of the specified per missible theories of adm issibilit y

63. See Edwar d J. Im winkelried, The D isp ut e Over th e Doctrin e of Ch an ces, CRIM. JUST., Fa ll 1992, at 16, 18. 64. See 2 J OHN HEN RY WIGMORE, EVIDENCE IN TRIALS AT COMMON LAW § 302, at 241 (197 9). D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1566 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 contained in th e second sent ence of Rule 404(b). In any even t, it is clear that the doctrine of chances is treated as a viable theory of admissibility under Ru le 404(b).65 As such, though, it is fraught with all the difficulties that atten d t he im plem en tation of Rule 404(b) generally. Even t he defenders of the doctrine concede th at , in man y cases, the distinction between permissible “doctrine of chances” evidence and imperm issible character evidence is slim.66 And, of course, there are the usual complaints that courts use the doctrine of chan ces to let in much of what should be kept out.67 But the problem with th e doctrine of chan ces is a syst emic one and is not limited to an occasional misapplication. Take the prototypical case for the doctrine of chances; i.e., the case based upon mu ltiple accusations. A defendant is cha rged with rape, and defends on the claim that the sexual encounter was consensual. The government wishes to introduce the testimony of other women who will testify that, in separate incidents, each of them was forced to submit to nonconsensua l sexu al intercourse by the defendant. If offered to prove that the defendant is in fact a rapist, the evidence will be excluded by the character evidence rule. There is a fair chance, however, that such evidence could be admitted under the doctrine of chances. Because most people are never accused of rape, the probability of an individual being falsely accus ed of ra pe is low.68 Th er e-for e, it wou ld be an im pr oba ble coin ciden ce for an individual to be falsely a ccused, even once, of rape. The probability of an individual bein g falsely accused on mu ltiple occasions is increasingly improbable.69 Therefore, the argument

65. See Mar k Ca mm ack, Using the Doctrine of Chances to Prove Actus Reus in Child Abuse and Acqu ain ta nce R ape: People v. Ewoldt R econs id ered , 29 U.C. DAVIS L. REV. 355, 407–08 (1996); Imwinkelr ied, supra note 63, at 19–20; Imwinkelried, supra note 1, at 1136. The use of the doctrin e of chances as a t heory of admissibility under Ru le 4 04(b ) ha s be en en dor sed by th e Cr im in al J us ti ce Sect ion of th e Am er ica n Bar Association. See Raed er, supra note 4, at 348. 66. S ee, e.g., Imw ink elrie d, supra note 7, at 588, 602. 67. See id. at 588; Ordover, supra note 18, at 168. 68. See Cam ma ck, supra note 65, at 396. This proposi tion does not dep en d u pon there bei ng a h igh per cen ta ge of r ap e a ccus at ion s wh ich ar e well founded. Sim ply because mos t p eople a re not accu sed a t a ll—t ru ly or fals ely—i t follows that most people ar e not accus ed fals ely. Id. 69. Suppose, for exa mp le, th at th e p roba bil it y of a false rape accusation being targeted against a par ticular individual is one in one hu ndred. Considered D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1567 goes, the mere existence of multiple accusat ions is r elevan t in a manner conceptually dist in ct from the chain of reasoning ordinarily barred by the character evidence rule.70 Essen tia lly, the doctrine of chances allows the inference of guilt of rape based upon multiple a ccusa tion s of r ape, provid ed that, a lon g the way, there is no inference drawn that the defendant is a rapist because he has raped before. There are several problems with this concept. We can start by stepping ba ck and gazin g at the big p ict ure. I s t her e n ot something awry with rules of evidence that permit the trier of fact in a rape case t o infer guilt ba sed m er ely on pr ior accusations of rape, but, at lea st in pr in ciple, ordinarily will n ot allow the trier of fact to infer guilt based on the fact that the accused is a ctually gu ilt y of rape on pr ior occasions? In addition, consider the ext en t to wh ich the doctrin e of chances has pushed Rule 404(b)’s supposed distinction between ina dmissible character evidence and admissible noncharacter evidence. For example, one of the issues which mu st be resolved in a “doctrine of chances” scenario is the nu mber of prior incidents or accusations necessa ry t o raise a rea sona ble in fer en ce th at th ey ar e un likely a ccounted for by innocent coincidence.71 Su pp ose that, in our hypothet ica l r ape case, the court deter min es that a min im um of two prior accusa tion s of rape is necess ary t o qu alify the evid en ce a s n oncharacter evidence. On the basis of a p roffer from the government, the court determines to allow the testimony of two witnesses as to two independent accusations of rape committed by the defendant. On th e first day of tr ial, the first of these witnesses testifies as promised, concluding her testimony with the close of the trial until the next day. Given th e court ’s pretr ial determination that two such accounts are necessary to qualify for the doctrine of chances, as the case stands at that point, the

independently, the probability of a second false rape accusation being targeted against that individual would also be one in one hun dred. Consider ed together , the pr obability of both false accusations being made against the same individual would be the product of the two ratios, which would be one in ten thousand. 70. See Cam ma ck, supra note 65, at 397. 71. See id. at 397–98; Im wink elrie d, supra note 63, at 52; Paul Rothstein, Intellectual Coh eren ce in a n E vid ence Cod e, 28 LOY. L.A. L. REV. 125 9, 1 263 (199 5); see also State v. Allen, 725 P.2d 331 (Or. 1986) (admitting evidence in an ar son pr osecu tion of a p ri or a rs on, un der th eor y of doct ri ne of cha nces, as proof that the fire wa s n ot s ta rt ed acci de nt al ly, e ven th ough th er e was only on e p ri or in cide nt ). D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1568 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 defendant has been enormously prejudiced by t he in trodu ction of cha racter evidenced ba rred by Rules 404 (a) and 404(b). The next morning, the second witness appears in court and testifies as a nticip ated. Not only is t he t estim ony of the secon d witness not character evidence, but the testimony of the first witness from the previous day has now been m agically converted to noncha ra cter eviden ce as well. A premise of t he doctrine of chances specifically, as well as of Rule 404(b) generally, is that character evidence and noncharacter evidence are mut ually exclusive categories of evidence. Yet here we have a testimonial account that jumps from one category to the other based solely upon th e testimony of another witness whose test imony relat es to an entirely independent occurrence.72 Ultimately, th e flaw in th e doctrine of chan ces is tha t it collapses the slim barrier separating character and noncharacter evidence as supposedly distinguished in Rule 404(b). There a re r eally only two possible explanations that argua bly account for the persu asive influence of the multiple- accusers scenario. The first must sta rt with th e assumption that accusers are probably telling the truth.73 Arguably, then, multiple accusers corroborate each other and increase the probability th at each is telling the truth.74 This explana tion is not genera lly inaccurate; it simply does not get us anywhere. Even if we credit each of the mu ltiple accusers, that does not explain how they cross over and corrobor ate each other. If the only impact of the multiple- accusers scena rio is tha t we credit the specific t estim ony of each, then the only thing we learn from mult iple accusations is that the defendan t committ ed multiple crimes. There would be no reason why the second accuser’s testimony increases the

72. If, in this scenario, we imagine t ha t t he s econd wit nes s does n ot m at eria lize at tr ial on t he secon d da y, t he tr ial cour t w ould pr esu ma bly h ave to ch oose a mon g (1) instructing the jury to disregard the testim ony of t he firs t w itn ess from th e previous day, (2) declarin g a m istr ial, or (3) revisin g its pr etr ial orde r lower ing the th reshold for t he doct rine of cha nce s t o one pr ior a ccus at ion. 73. See Cam ma ck, supra note 65, at 399. 74. Even this h ypothesis as sum es th at t he defenda nt’s innocence can only be consistent with a conclusion th at th e accuse r is lyin g. But th is is not necess ar ily the case. The a ccuser m ight be mis ta ken as t o any n um ber of issu es, inclu ding identificat ion. Moreover , the accuse r’s testimony could b e flawlessly accura te a nd still not account for a reasonable mistake or ignorance on the part of the defendant that would negate t he requ isite level of scienter. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1569 lik elihood that the defendan t committ ed the first crime.75 Indeed, if this is all there is to the doctrine of chan ces in t he multiple-accusers scenario, then accusations of unchar ged crimina l incident s would not even be relevant . The second, and only genuine, exp la nation of th e per suasive in flu en ce of the mu ltiple-accuser s sit ua tion is tha t we credit precisely the character inference that is supposedly barred by the character evidence rule.76 In fact, the explanation for the doctrine of chances in the multiple-accusers context is simply a convoluted explana tion of the general propensity inference. Each separ ate accusation would have no bearing upon the accuracy of another allegation but for the conclusion that the multiple accusations demonstrate a cross-situational pattern of behavior, wh ich is bu t a varia tion on t he t aboo in fer en ce of a genera l propensity or character trait. We connect one accusation to another accusation of a simila r crime precisely because we credit, at least to some extent, the notion that people do beh ave con sist en tly wit h person al ch aracter traits.77 That is the only reasonable explana tion for why even the doctrine of chances is limited to multiple accusations of identical or similar crimes. It is the r ea son wh y t her e is lit tle or no corrobor ative effect when the mu ltiple accusations involve significantly discrete crimes. The doctrine of chan ces ma y in fact be tr uly consistent with Rule 404(b), but u nfort un at ely tha t consistency stems from the fact that, at least in some cases, the doctrine of chances is the latest theory for allowing character

75. This corroborative impact would only be true if the mult iple accusers were testifying as to th e sam e incident, which of course, is never t he case in a Rule 404(b) situation. 76. See Cam ma ck, supra note 65, at 399. 77. See id. at 400. Professor Rothstein has expressed this conclusion most succinctly: The essen ce of this pr obable gu ilt ar gum ent is that there is a disparity between the chances, or probability, that an innocent per son wou ld be charged so ma ny t imes an d th e cha nces, or proba bility, t ha t a guilty per son would be ch arged so ma ny t imes. If there is s uch a dispa rit y, however , it is only because a guilty person would have th e propensity to repeat the crime. If it w er e n ot for th e pr open sit y to r epe at , th e ch an ces, or t he proba bility, that an innocent person and a guilty person would be charged repe at edly would be iden tical. H ence, the ar gum ent hin ges on pr opensit y and runs afoul of the first sen ten ce of Rule 404(b). The effort t o reconcile the permission in the Rule with th e prohibition in the Rule has failed. Rothstein, supra note 71, at 1262–63. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1570 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 evidence to reach the ju ry wh ile main tain in g t he piou s fiction that we follow the character evidence rule.

4. Anoth er look at th e su bsta ntive 404(b) prerequ isites As has been discussed,78 th e th ird prerequisite to the admissibility of evidence un der Rule 404(b) is that th e tr ial court must be persuaded that there is a legitimate, noncharacter purpose to such evidence. The current morass that this task has engendered has arisen precisely because the distinction s allowed by t he r ule, and t hus r equir ed of th e courts, ar e either bar ely per ceptible or ent irely illusory.79 We admit wh at essentia lly is character evidence by labeling it otherwise, en ga gin g somet im es in the m ost contorted analyt ica l gym nast ics to preser ve an en tir ely theor et ica l a dh er en ce t o the character evidence rule.80 Th e schola rly p ursu it of su ch noncharacter theor ies as t he doctrin e of ch ances is not unlike the in divid ual wh o, a fter hours of gazing upon an ordinary paint ing, pronounces that if one stands at a particular angle at a part icular tim e of day using very part icular lighting, the painting will reveal itself to be somethin g quit e extraordinary. The problem is t ha t judges m ust, in th e ordinary light of common sense and experience, attempt to implemen t theories that are often only operative in the a rtificia l light of th e scholar’s laboratory. It should not be surprising that the application of Rule 404(b) h as n ot resulted in a coherent and consistent body of law. One remaining question is whether the misadventures that attend the case law implementin g Rule 404(b) ar e entir ely the results of the difficult subject matter, or whether they are in part the products of willful disrega rd of t he limit ation s u pon admissibilit y sp ecified in Ru le 4 04(b). S ome in sigh t in to this question might be attained by exploring the fourt h prer equisite

78. See supra note 23 and accompanying text. 79. For this rea son, some commentat ors have advoca ted abandoning the “admiss ible categor ies ” app roa ch of R ul e 40 4(b) in favor of a cas e-by -cas e det er mi na tion based upon the assessment and balancin g of t he probative value and potential pre judicia l effect of the prior crim es evid ence. S ee, e.g., Allan, supra note 45, at 271–74; St one, supra note 41, at 984. This appea rs to be the current dir ect ion of th e cha ra cter eviden ce ru le in E ngla nd. S ee Director of Pub. Prosecut ions v. Boardman, [1995 ] App. C as . 421 ; see also Allan, supra note 45, at 253. 80. See 1A WIGMORE, supra note 34, § 54.1, at 1152–56. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1571 to admissibility under Rule 404(b). S pecifica lly, t he t ria l cou rt must balance the proba tive, n oncharacter va lu e of such evidence aga in st the poten tia l preju dice of t he ju ry u sing such evidence to deter min e t he ge ner al propensity, or chara cter , of the accused, taking into consideration that such evidence is, seem ingly wit hout exce pt ion , accompan ied by a limiting instr uction juxt apos in g t he per missible noncharacter use and the forbidden character treatment of the evidence.81 One of th e in her en t difficulties in st riking the required balance for admission of evidence un der Rule 404(b) is that there is frequent ly a parallel relationship between what the rule designates as relevant and what the rule designates as prejudicial.82 For ma ny of the th eories of admissibility under Rule 404(b), the stren gth of the claim for adm issibilit y increases in pr oportion to the similarity between the charged crime and the prior crime.83 Unfortu nately, the greater the similar ity between the two crimes, the more compelling the forbidden conclusion that the accused is predisposed to commit th at very crime. It is pa ra doxical, then, bu t n ot su rprising, t ha t the circumstance in which the danger of prejudice is great est —i.e., when such evidence is offered against the criminal defendant84—is also the circumstance in which such evid en ce is most pr oba tive and s ign ifica nt.85 Consequently, despite the fact that the character evidence rule is designed prima rily to protect t he crimin al defendan t, a perusal of the annotations to Rule 4 04(b) r eve als that the va st majorit y of Rule 404(b) cases involve evidence offered by t he prosecution against the accused. And becau se t he compet ing va lues vary

81. See supra notes 24–25 and accompan ying te xt. Becau se th e lan gua ge in Ru le 404(b) is permissive (“may . . . be admissible for other purposes”), even if eviden ce fits with in one of th e cate gories of adm issibility, it mu st still be exclu de d if t he “danger of un fa ir pr ej u di ce ” “su bs t a n t ia ll y o u t we ig h [s ]” “i t s pr ob a t iv e v a lu e. ” F ED. R. EVID. 403; See als o 2 WEINSTEIN ET AL ., supra note 14, ¶ 404[18], at 404–108 to 404–109. 82. See Ston e, supra note 41, at 954. 83. For example, the m ore similar th e modus operandi in the t wo crimes, the stronger the inference that the identity of the actor in the second crime is the already identified actor in th e first crime. 84. See Ston e, supra note 41, at 983. 85. As the similarity of the multiple crimes increases, th e distinction between the char acter an d non char acter uses of the evide nce ar gua bly disappear s. See supra notes 79–80 and accompanying text. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1572 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 proportionally rather than inversely, it is not merely the cases at the margin that present difficult issues. So how do courts actually resolve th e requir ed ba lan cing test? Th er e a re some limit ation s u pon the information that can be gath ered from the published opinions. Because the Double Jeopardy Clau se effectively elimina tes governmen t a ppea ls from trial court rulings disallowing the in trodu ction of evidence under Ru le 404(b),86 the reported appellate decisions are invaria bly cases in wh ich the t ria l cou rt admitted the challenged evidence under Rule 404(b). Moreover, the bala ncing test required by Rule 403, even in th e Rule 404(b) context, is one as to which there is a grea t deal of app ellate cou rt deference to the discretion of the trial courts.87 Consequently, what appea rs anecdotally to be a high percentage of appellate decision s affirming th e allowance of evidence under Rule 404(b) might not represen t the a ctual success r ate of prosecutors seeking to introduce such evidence. Never th eless, some observations can be made. First, appellate courts readily defer to trial court discretion. Perhaps because th e requir ed ba lan cing test is often so difficult, there is a conspicuous willingness of at least some appellate courts to ground their decisions predominantly, if not singularly, upon the deference accorded to the trial courts.88 This suggests a virtual abdication of an y mea ningful appellate a ut hority on t his issue. In some cases, the repudiation of meaningful appellate review is express, and not merely implied.89

86. In any case in which th e evidence is excluded and the defenda nt is acquitted, no appeal can be hea rd because double jeopardy would bar a r etrial. In any case in which the evidence is excluded and th e defendant is nevertheless convicted, the government has n o reason to appeal a t all, and t he defendant ha s no cause to raise th e Rule 404(b) issue on a ppea l. 87. See 2 WEIN STE IN E T AL., supra note 14, ¶ 404[18], at 404–21. 88. S ee, e.g., Unit ed St at es v. Cun nin gham, 103 F. 3d 553 , 55 7 (7t h C ir . 19 96), cert. denied, 117 S. Ct . 1481 (1997); Un ited Sta tes v. Pr ocopio, 88 F.3d 21, 30 (1st Cir. 1996), cert. d eni ed, 117 S. Ct. 620 (1996), an d cer t. d eni ed, 117 S. Ct. 1008 (199 7); Unit ed Sta tes v. Her nan dez, 84 F.3d 931, 935 (7th Cir. 1996); United States v. Parziale, 947 F.2d 123, 129 (5th Cir. 1991); Oregon v. Allen, 725 P.2d 331, 336 (Or. 198 6). 89. For example, in United States v. Brown, 34 F.3d 569 , 57 4 (7t h C ir . 19 94), the court stat ed tha t, absent “a categorical rule” requir ing exclus ion of the Ru le 404(b) evidence, “we must defer to the judgmen t of the trial cour t.” And in United States v. Hadaway, 681 F.2 d 21 4, 21 7 (4t h C ir. 198 2), t he cour t a nn oun ced that, “[g]iven the wide discretion permitted the district judge, it is fruitless to contend that the eviden ce was impr operly a dmit ted .” D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1573

This hands -off app roach used by a pp ellate courts in reviewing the seemingly endless parade of appeals by convicted defendants who su ffered the r evelat ion of their uncharged criminal conduct before the juries which convicted t hem ha rdly suggests a deep a nd abiding commitment to the character evidence rule, wh ich is su pp osed t o pr otect defendants from such exposure. It often seems as though once a so-called noncharacter use for the prior crimes evidence has been identified and adequately defended, the contest is, as a practical matter, over. It is as if char acter evidence is ordinar ily excluded, not because it is pr eju dicial, bu t r at her because it is just not relevant enough. But if its probative value can be enhanced incrementa lly by advan cing a th eory of relevan ce in addition to the suggestion of a general propensity, the bar has been exceeded an d the bala ncing test is rarely an obstacle to admissibilit y. Second, th ere is a flip side of this ph enomenon in cases in which the court has been persuaded that there is a real noncharacter use for the prior crimes evidence. In these circumstances, to the extent that the potential prejudice in the equat ion is merely th e da nger of th e jur y infer rin g a defendant’s gener al propen sity to engage in such crimina l conduct, the arguments for exclusion often fail to carry the day. Many judicial opinions seem to require more than this ever- present danger, such as someth ing par ticular ly inflammatory about the prior crimes evidence.90 Again, the devot ion to guarding against the danger of the general pr opensity in fer en ce is unimpressive.

90. S ee, e.g., United States v. Boyd, 53 F.3d 631 , 637 (4th C ir. 1995) (finding that the evidence of the defendant’s prior drug use was not unduly prejudicial in a drug prosecution “because the evidence of [the defendant’s] personal use of marijuana and cocai ne did not in volve cond uct any more sensational or disturbing than the crimes with which he was ch ar ged”); United St at es v. Melin g, 47 F.3d 1546, 1557 (9th Cir. 1995) (finding that tapes of the defendan t at tempt ing to hinder a n inv est iga tion were not unfairly prejudicial because “the profanity in the tapes was rela tively m ild by tod ay ’s st an da rd s”); Ca rs on v. P olle y, 6 89 F .2d 562 , 57 3 (5th Cir . 1982) (finding that, in a ci vil r igh ts act ion for th e u se of exce ss ive force in effecting an arrest, ad mi ssi on of th e de fen da nt ’s per form an ce ev al ua ti on r eport in dica ti ng the defendant’s pr ior losses of tem per a nd exp ressions of host ility towa rd de ta inees was not u njus tly prejudicial beca us e “th er e we re no h orr ifyin g de ta ils t ha t w ould pre dictably inflame the ju ry ’s pa ss ion ”). D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1574 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998

Third, th ere ar e ma ny cases in which Rule 404(b) evidence is admit ted, despit e an a ppa rent ly weak case for t he probat ive value of the evid en ce in connection with its proffered, “noncharacter” use.91 As noted earlier, this is particularly apparent in some decisions admittin g evidence under Rule 404(b) to show intent or other levels of mens rea.92 But the examples of this phenomenon are by no mea ns s o lim it ed. F or example, notwithst anding t he fa ct that it is the excep tion al person wh o could not use som e a dd it ion al cash, eviden ce of the defendant’s uncharged, costly criminal appetites sometimes finds its way to the jur y on th e improbable theory that it supplies a motive for the defendan t’s charged crime, where the latter is on e h avin g the a llu re of some pecu nia ry (and sometimes even nonpecunia ry) benefit.93 In many of these cases, one is left wit h the im pr ession that, not only is t he genera l ban against character eviden ce not a priority, bu t, t o the contrary, the courts often embrace opportunities to allow character evidence in under t he gu ise of treat ing such evidence as someth ing els e entir ely. Fourth, there often is no serious att ention to what the character evidence r ule d eems t o be pr eju dicia l be cause of a

91. Appar ent ly, even some defenders of the chara cter evidence rule would agree with th is obser vat ion. S ee, e.g., Imw ink elrie d, supra note 7, at 577–78. 92. See supra notes 61–62 a nd a ccompan ying te xt. 93. In Meling, 47 F.3d at 1557, a p rosecution in wh ich the defenda nt wa s charged wit h v ar iou s offe ns es ar isi ng from his attempt t o kill his wife in order to collect $700,000 on a life insurance policy, e vid en ce of t he de fen da nt ’s pu rch as e of a n expen sive firearm was deemed to be more probative of the defendant’s motive (greed) than prejudicial (propensity for violence). In Boyd, 53 F.3d at 636–37, a prosecu tion for conspira cy and p ossession with int ent to dist ribu te m ar ijuan a, te stim ony that the defendant used m ar iju an a a nd coca in e was foun d t o be mo re pr oba ti ve of t he defendant’s motive t o finance a nd supply h is own dr ug u se th an it wa s pr ejudicial in linking the defendan t to criminal dr ug activities. Moreover, in Cun nin gham , 103 F.3d 553, the defendan t, a nu rse, faced crimin al ch ar ges ba sed on h er all ege d t he ft of Dem er ol from hospital syringes. The pr osecution was per mi tt ed t o in tr odu ce ev ide nce that, four yea rs ear lier, t he de fenda nt ha d been addict ed to Dem erol, th at her nu rsin g license had been suspended because of her prior theft of Demerol and that she had subsequ ently fal sifi ed dr ug te st re su lt s in orde r t o m ai nt ai n t he re in st at em en t of t ha t nursing license. With bar ely a passin g reference to th e requisit e balan cing test, the cour t determ ined tha t th e value of the evidence to prove a motive for the theft (the defendant’s addict ion) exceeded a ny pr ejudicial suggest ion of a propensity to steal Deme rol. Ev en as su mi ng th is t o be tr ue , giv en th e e xpl icit evi de nce of th e defendant’s addiction, it is u ncl ea r w ha t t he in cre me nt al va lu e of t he evi de nce of the prior theft of Demerol was for the stated pur pose, especially as balanced against its tendency to suggest a pr opens ity t o pilfer t he d ru g. See id. at 556–57. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1575 mechanical, and arguably disingenuous, fait h in th e efficacy of limit ing in st ruction s t o the ju ry. Of course, the entire jury system is pr edicated on t he assumption th at jurors can, a nd will, follow the instr uctions from th e tr ial jud ge.94 This assumption ext en ds to lim it in g in st ruction s, i.e., in st ruction s restricting jurors’ consideration of evidence t o it s cor rect scop e in circumst ances in which the evidence could be used for bot h a proper and an improper pur pose.95 Judges wh o adm it evidence under Rule 404(b) routin ely assume (or so they say) that any potential prejudice from such eviden ce is controlled by an instr uction directing the jury’s consideration of such evidence to its sp ecified “non character ” context an d pr ohibiting consideration of that evidence as bearing upon the character of the accused.96 But on what possible ba sis d oes this ass umpt ion res t? Any meaningful consideration of prejudice must focus, not on the theory constructed by the law and contained in the jury instructions, bu t rather on the a ctual u se to be made of su ch evidence by th e jur y.97 So the question one m ust a sk is, “How probable is it that jurors will understand and follow instructions?” For exa mple, are juries capable of considering evidence of the defenda nt’s prior dr ug sales as bea rin g on the defendant’s intent to sell drugs but not on the defendant’s propensity to do so?98 Can a jury likewise consider the defendant’s prior theft of drugs from hospital inventory as bearing on the defendant ’s motive for stealin g drugs from hospital inventory but not on the defendant’s propensity to

94. See Par ker v. Ran dolph, 442 U.S. 62, 73 (1979); Edward J. Imwinkelried, The Ri ght t o “Plead O ut ” Issues an d B lock t he A dm iss ion of Prejudicial Evidence: The Differential Tr eatm ent of Civ il L iti gan ts a nd th e Crim in al A ccus ed a s a D eni al of Equal Prot ection , 40 EMORY L.J . 341 , 37 8 (19 91). 95. See Spencer v. Texas, 385 U.S. 554, 562 (1967). The authority for such a limiting inst ru ction is conta ined in t he F eder al Ru les of Evide nce. S ee F ED. R. EVID. 105. 96. S ee, e.g., Un ited Sta tes v. Her na nde z, 84 F.3 d 931, 935 (7th Cir. 1996) (“[T]he dis tr ict court instru cted the jury both before an d after t he 404(b) evidence was admitt ed tha t it could be considered ‘only on the question of the defendant’s intent, knowledge, ab sence of mi st ak e, or acci den t a s it re la te s t o th e ch arges in the indict men t.’ Absent any su bstan tial evidence to the contrar y, we presume t hat the jurors follow ed th eir in st ru cti ons.”). 97. See Kuh ns, supra note 60, at 796. 98. See, for example, the decisions discussed supra at n ote 62. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1576 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 steal drugs from h ospital in ventory?99 The an swer is that the chance that such instr uctions accomplish their intended purpose is pr actically nil. The inten sity of the struggle between prosecutors and defense coun sel over the a dm issibilit y of prior crim es evidence under Rule 404(b) exists because both sides recognize that so much more is at stake than just pr oof of motive or intent. Privately, many judges acknowledge that their limiting instructions accompanying Rule 404(b) evidence are almost certainly futile. Occa sionally, t his judicia l ca ndor ascends to a public forum a s well: The undue preju dice [e vid en ce a dm itte d u n de r Ru le 4 04 (b)]

engenders may be gua rded aga inst in only one w ay: by an inst ru ction . . . to the effect that the jury is not to consider the evide nce as going t o th e ch ara cter of the accused but only as going to identity. It may well be that the jury cannot make this dist inct ion in its collect ive m ind. I f th at is so it is unfort unat e, bu t it hap pe ns a ll th e t im e . . . .100

Even sch ola rs w ho gen er ally su pp ort the cu rrent rule acknowledge that ther e is lit tle con fiden ce in the u tilit y of the accompanying limiting instructions.101 This conclusion should not be in the least surprising. There is some support for the proposition that jurors have difficulty understanding instr uctions genera lly.102 Her e, th e problem is much more a cute. As discu ss ed above , t he la w’s dist in ction s between the character an d nonchara cter tr eat men ts of prior crimes evidence a re fr equen tly obscu re, occa sionally imperceptible, and arguably, sometimes nonexisten t.103 Scholars disagree over the actuality of at least some of the distinctions engendered by Rule 404(b).104 It has been suggested

99. See United States v. Cunningham, 103 F.3d 553 (7th Cir. 1996), discussed supra at n ote 93. 100. United St at es v. D an zey , 59 4 F .2d 905 , 91 5 (2d Cir . 19 79). 101. S ee, e.g., Imw ink elrie d, supra note 1, at 1140–41; Imwinkelried, supra note 94, at 379. 102. See Imw ink elrie d, supra note 1, at 1140–41; Geoffrey P. Kramer a nd Dorean M. Koenig, Do Jurors Understand Criminal J ury Instructions? Analyzing the Results of th e Mi chi gan J ur or Com preh ens ion Proj ect, 23 U. MICH. J.L. REFORM 401, 402, 429 (199 0). 103. See supra notes 57–80 a nd a ccompan ying te xt. 104. Compare Imw ink elrie d, supra note 63, at 18–2 0 (de fen din g th e doct rine of chances as a non char acte r t heor y of releva nce), with Roth ste in, supra note 71, at D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1577 that some judges do not fully appr eciate th e dist inctions.105 Those of us who teach Evidence know that, when the syllabus rolls around t o Ru le 4 04(b), con siderable tim e a nd effort are required to rem ove even a por tion of the perplexe d expressions that st are back at us in the cla ss room. Yet jurors, with n o background in the law and no information as to the theory and purpose of Rule 404(b), are supposed to understand the dist inction on the basis of a one or two sentence conclusory direct ion bur ied a mong a ba rr age of other complex instructions? The proposition is really quite preposterous. Moreover, eve n if t he ju rors w er e somehow to comprehend the limiting instruction applicable to evidence admitted under Rule 404(b), there would still be a consider able question as to whether the instruction would be obeyed. Unremarkably, jurors are much les s likely to follow instr uctions tha t contra dict t heir own good sense and experience.106 Con sequ en tly, one of the major impediments to the efficacy of the Rule 404(b)-prompted limit ing instruction is that the prohibited use of the evid en ce flies in the fa ce of what jurors know to be true: that character is relevant to behavior.107 If limiting instructions in this cont ext a re so manifest ly inadequate (a conclu sion that has not escaped t he a pp reh en sion of th ose volunt ar ily connected with th e conduct of crimina l trials in this country), then the steady allowance of evidence of the accused’s prior crimes under Rule 404(b) is yet another tellin g indication that the commitm ent to the character evidence rule in practice falls a good deal short of the rhetoric. In fact, the considera ble extent to which evidence of the criminal defendan t’s character manages to meet Rule 404(b)’s criteria for adm issibilit y reflect[s] a ha lf-hearted and unprincipled compromise between

an inte re st in tr uth seek ing a nd a belief t hat we sh ould n ot judge pe ople or their acts by their cha racter . . . . (Refusal by som e courts to even acknowledge the extent of the effective

1262–63 (attacking th e distinction between th e doctrine of chances a nd ordina ry character evi de nce as illu sor y). 105. See supra note 59 and accompanying text. 106. See Kra mer & Koenig, supra note 102, at 429–30. 107. See Imw ink elrie d, supra note 94, a t 379 ; Susa n Ma rlen e Davies, Evidence of Ch ara cter to P rov e Con du ct: A R eass essm ent of R eleva ncy , 27 CRIM. L. BULL. 504, 511 (199 1). D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1578 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998

re pu dia tion of the character evidence rule is surely the m ost unprincipled comp rom ise of all; this compromise may be fa irly de scr ibed a s h ypocrit ical.)108

C. Rule 60 8(a) The extent to which the character evidence rule is compromised is not limited to the Ru le 404(b) context. As to any witness who testifies at trial, the opponent may call a character witness to testify, in the for m of opinion or reputation testimony,109 regarding the witn ess’s propensity for telling t he truth.110 Thus, should a criminal defendant testify in his own behalf, th e prosecution may call one or more wit nesses t o testify that, in the opinion of the witness or of the community, the defendant is dish onest. Because this type of attack upon the defendant’s credibilit y does not require that the defendan t open the door by first in trodu cin g eviden ce of h is or her own character, ther e is nothin g t hat the defenda nt can do t o forestall this attack except to remain off the witness stand. In this particular, then, the rules of evidence hardly demonstrate a commitment to protecting the defendant from such character evidence. To the contra ry, the law either allows eviden ce of the defendant’s dishonesty or at least t olerates a predictable disin cent ive to the exer cise of the defen dant ’s right to test ify.111 It is true that such cha ra cter eviden ce is almost always accompanied by a lim itin g instr uction rest ricting t he jury’s consideration of such evidence to the issue of the defendant’s credibility as a witness. Once again, however, the law would appear to demonst rate a n unwa rranted fait h in su ch instructions.112 Pa rt icular ly in cases in which t he defendan t is charged with a crime involving dish onest conduct, the connect ion between a gen era l propensity for dishonest conduct and a particular act of dishonesty might be too obvious to ignore. Should the defendant testify, and should the defendant’s character for truthfulness then be attacked, the defendant may

108. 1A WIGMORE, supra note 34, § 54.1, at 1156. 109. The expla na tion s of, a nd dis tin ction s be tw een , opin ion a nd re pu ta tion evid en ce can be found supra at n ote 8. 110. See FED. R. EVID. 608 (a). 111. See 3 WEINSTE IN ET AL ., supra note 14 ¶ 608[01], at 608–10. 112. See id. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1579 then call a positive character wit ness t o testify, in the for m or opinion or reputation testimony, that the defendant is a truthful person.113 However , just as in the situ at ion in wh ich the defen da nt calls a pos it ive character witn ess r egar ding his own character pursuant to Rules 404(a) and 405(a ),114 calling a character wit ness t o testify t o the defenda nt’s character for truthfulness opens t he door to cross-examination of the character wit ness a s t o sp ecific in st ances of dish onest acts of the defendant.115 Thu s, in cases in which th e defendant is charged with a crime in volving dis hon est conduct , the jur y migh t lear n of simila r, prior acts of t he defenda nt. Th e t heor y of admissibility is, once again, tha t this evidence is not direct evidence of the defendan t’s chara cter, but is, ra th er, simply relevant to the weight to be give n to the con clu sor y t estim ony of the character witness,116 and the jury will be in st ructed to that effect. But the efficacy of such a limiting instru ction in th is context is indist in gu ishable from the dubiou s circumscr iption accomplished by the parallel instruction accompanying cross- examination of character witnesses under Rule 405(a).117 Rule 608(a), then , pr ovides yet another illu st ration of th e law’s glaring neglect of the supposedly important character evidence rule. D. Rule 608(b) Any witn ess, including the criminal defendant, may poten tia lly be cross-examined concerning th e witness’s own sp ecific acts evidencing an untruthful character.118 This form of impeachment is limit ed t o the cross-exam ina tion itself, and any extr insic eviden ce th at migh t be necessa ry to complete the impeachment of the reluctan t witness is disallowed.119 Nevertheless, a testifying defendant might well be compelled, for reasons of conscience, potential perjury prosecution or

113. See FED. R. EVID. 608 (a)(2 ). 114. See supra notes 13–19 a nd a ccompan ying te xt. 115. See FED. R. EVID. 608 (b)(2). 116. See 3 WEIN STE IN E T AL., supra note 14, ¶ 608[06], at 608–100. 117. See supra notes 16–18 a nd a ccompan ying te xt. 118. See FED. R. EVID. 608 (b)(1). 119. See FED. R. EVID. 608(b). The re ason s for th is limit at ion ar e prim ar ily to avoid th e a lloca tion of sign ifican t t rial t im e a llowin g confli ctin g eviden ce t o resolve disputes as to uncharged conduct, and to prevent unfair surprise to the defendant forced to defen d aga inst allega tions of unchar ged condu ct. S ee 3 WEINSTEI N E T AL., supra note 14, ¶ 608[05], at 608–45. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1580 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 otherwise, to admit prior acts of dishonesty. Mor eover, t he questions themselves, even if followed by denials, might have a damaging impact.120 Aside from this difference as to form, a ll that was stat ed in the previous section regarding Rule 608(a) applies here as well. The accused might be furth er deterr ed from testifying in the face of unleashing evidence of specific instances of dishonest conduct.121 Such impeachment will tr igger t he u sual limitin g instruction, bu t, particu la rly in cases in wh ich the impeaching conduct bears some simila rit y t o the charged crimes, the direction to the jury will be of suspect utilit y.

E. Rule 609 Any witness who testifies may be impeached by the introdu ction of his or her prior convictions. Under the federal rules, admission of prior convict ions for crimes of dishonesty—such as p er jury, bribe ry or fraud—is allowable without pos sibilit y of exclu sion for prejudice, even in circumstances in wh ich the wit ness s ubject to impeachmen t is the criminal defendant.122 All other felony convictions may be admissible to impeach the wit ness, pr ovid ed the bala nce of probat ive value against unfair pr eju dice is resolved in favor of admissibilit y.123 Th e federal r ule, wh ich gen er ally conforms t o the practice in other American jurisdictions124 and longstanding practice under th e common law,125 is actually pr em ised upon a

120. See 3 WEIN STE IN E T AL., supra note 14, ¶ 608[05], at 61–62. 121. See 3 id. ¶ 6 08[05], at 62–63. 122. See FED. R. EVID. 609 (a)(2 ). 123. See FED. R. EVID. 609(a)(1). For all witnesses other than the criminal defendant, the ordinary bala ncing t est of Rule 403, which pres um es ad miss ibility unless the “probat ive value is su bsta nt ially out weighed by th e dan ger of un fair pre judice,” app lies. F ED. R. EVID. 403. In 1990, Rule 609 was amended to accommodate the special concerns regardin g potential prejudice when t he impeached witness is th e accus ed. S ee 3 WEIN STE IN E T AL., supra note 14, ¶ 609[04], at 609–42 & n.3. If the prior conviction is not a crime of dishonesty au toma tically ad miss ible under Rule 609(a)(2), the defen dan t m ay only be impeached “if the court determines that the prob at ive v al ue of ad mi tt in g t hi s eviden ce ou tw eigh s it s p re jud icia l effect t o the accus ed.” FED. R. EVID. 609(a)(1). For all im peachable convictions, there is a strong pr es um pt ion th at a conv ict ion is s ta le a nd in ad mi ss ibl e if more than ten years ha ve pa sse d fr om t he lat er of th e da te of convict ion or the date of r eleas e fr om incarceration. S ee F ED. R. EVID. 609 (b). 124. See 3A J OHN HEN RY WIGMORE, EVIDENCE IN TRIALS AT COMMON LAW § 926, at 750 (197 0). 125. See Kar p, supra note 57, at 29. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1581 theory which bla ta nt ly disr egards the reasons u nderlying t he character evidence rule. Specifically, th e reasoning u nderlyin g Rule 609 is t ha t prior crim es in dicate a gen er ally poor character, which in turn supports an inference that the individual would t est ify falsely.126 There are seve ral proble ms w it h this hypothesis. First, the connect ion between the exis ten ce of a prior conviction and testimonial dishonesty is tenuous at best.127 Because the impeaching conviction need not be for dishonest conduct, and further because it is not limited to the case in which the defendant testified falsely128 at a previous trial, the Rule 609 premise is simply ba sed u pon the defendant’s general bad character. This inference of perjurious behavior based solely upon gen er al crim inal condu ct is actually fa r less rea sona ble than the more specific logic (i.e., that a person who commits particular crim es is more likely tha n other persons to commit that sa me, p articula r crim e a gain) supposedly ba rr ed by t he character evidence r ule. Even in cases in wh ich the prior conviction is one based u pon dish onest condu ct, t he prior conduct might be unrelated to testimonial untruthfulness except for the very general propensity toward dishonesty. Sma ll wonder, then, that studies suggest that jurors do not vary th eir assessment of the defendan t’s credibility with the presence or absence of impeachable prior convictions.129 Second, despite the often unimpressive impa ct of th e impeachable conviction upon the assessment of the defendant’s credibilit y a s a wit ness, evidence admitted under Rule 609 does appear to have an unint ended impact. One simulated study suggests that evidence of prior convictions does increase convict ion rates, es pecia lly wh er e t he prior convictions are for

126. See 3 WEIN STE IN E T AL., supra note 14, ¶ 609[02], at 609–28. 127. See Uviller , supra note 12, at 886–87. 128. Argua bly there would be m uch gr eat er r elevan ce to a conviction result ing from a t ria l in wh ich th e defen dan t t est ified. See id. at 888. We ca n in fer from th e convi ction itself th at th e defend an t’s testim ony at th e ear lier t ria l was found to be un tr ut hful. 129. See Davie s, supra note 107, at 506, 520; Roselle L. Wissler & Michael J . Saks, On th e In effica cy of L im iti ng In str uct ion s, 9 LAW & HUM. BEHAV. 37, 41, 43–44 (199 1). One possible explanation for this is that, given the defendant’s motive to lie in his or her own defen se, t he d efend an t’s credibilit y is inh ere nt ly sus pect. See id. at 43, 47. C ons equ en tly , pr ior con vict ions mi ght p rod uce lit tle or n o increment al da ma ge to th e defend an t’s credibility. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1582 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 crimes similar to the charged crime.130 And, although the similar ity between the proffer ed im peachable convict ion and the charged crim e sh ould ordinar ily counsel against admissibility in cases in which th e balan cing test of Rule 609(a)(1) controls, such similar -crimes evidence still occasionally finds its way to th e jury under this provision.131 Moreover, if the prior conviction is for a crime of dishonesty, it must be allowed for impeachmen t u nder Rule 609(a)(2) even if the defendant is cur rent ly charged with the identical crime. Thus, a testifying defendant charged with, for exa mple, per jury, can be im peached wit h his or her pr ior perjury convict ion . Once again, the only shield pr otectin g the testifying defendant from the very evil su pposedly pr ohibit ed by t he character evidence r ule is t he limit in g in st ruction. A prior convict ion of the defendan t a dmitt ed under Rule 609 is invaria bly accompa nied by an instru ction tha t such evidence is to be considered in evaluating the defendant’s credibility as a witness, but not as eviden ce of the defendan t’s propensity t o have committed the charged crime.132 But empirical studies confirm that these instructions are hopelessly powerless to prevent jurors from using the prior convictions in assessing the defendant’s guilt.133 The fact that jurors do not consider prior convictions in accordance with the limiting instructions that accompany Rule 609 impeachment does not reflect poorly on either t he good sense or the good faith of jurors. Consider, first of all, whether the instruction itself is sensible. For example, suppose a defendant, on trial for perjury, takes the stand and is impeached with a prior conviction for per jury. The requisite limit ing instr uction would essent ially direct the jury to make the followin g distin ction : You ma y not consider the defendant’s prior per jury conv iction

as evid en ce t hat th e d efen da nt is gu ilty of the crim e of pe rju ry with which he is charged. You may, however, consider the

130. See Wissler & Sak s, supra note 129, at 41–46. 131. S ee, e.g., U ni te d S ta te s v. Alvarez, 833 F.2d 724, 727 (7th Cir. 1987) (admitting evidence of prior cocaine conviction to impeach defenda nt ’s testim ony in cocaine pr ose cut ion ). 132. See 3 WEIN STE IN E T AL., supra note 14, ¶ 609[05], at 609–84 to 609–85. 133. See Robert G. Sp ect or, Rule 609: A Last Plea for its Withdrawal, 32 OKLA. L. REV. 334, 352–53 (1979); Wissler & Sak s, supra note 129, at 44. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1583

defendant’s pr ior p er jur y conviction in e valuating the credibility of the defendant as a witness.

The latter half of the dichotomy essentia lly per mit s the jury to use th e defen dant ’s prior perjury conviction to determ ine whether the defendan t committ ed perjury in the instant case in which th e defendant denied committ ing perjury on the occasion which is the subject of the prosecution. How can jur ors be expected to even un derstan d, never m ind follow, such a convolu ted direct ion ? Take a more straightforward example. Suppose the defendant is on trial for robber y a nd is impeached wit h a pr ior robbery conviction. The juror thinks that the prior robbery is relevant to the issue of whether th e defendant commit s robberies. The law says that the prior robbery means no such thing, but only means that the defendan t lies while t estifying under oath. It ha rdly seem s consciona ble to say t ha t t he pr oble m her e lies wit h the ju ror. The heart of the matter is that the limit in g in st ruction given in this context “fails t o correspon d t o the a ctual effe ct of the evidence even in the minds of the most sober and conscien tiou s jurors.”134 Prosecu tors and defense lawyers are keen ly awa re of th is reality. This explains why a defendant’s prior convict ion s—esp ecia lly for crim es simila r to the charged crime—hea vily in flu en ce t he defenda nt’s decis ion as to whether to test ify.135 Thus, even before the recent amendments to the Federal Rules of Evidence allowing character evidence in sex offense prosecutions,136 the character evidence rule was already so fraught with gaping exceptions, the most subtle distinctions, and impossible jur y inst ru ctions so as to be pr actically eviscerated in numer ous circumst ances. In the words of one commentator, “the windings and twistings of the character evidence rule and its variou s excep tion s a re la rgely without ra tional explan at ion,” and t he existing practice as t o the ru le is an “effort[] to mak e sen se out of nonsen se.”137

134. Uviller, supra note 12, at 882. 135. See HARRY KALVEN, JR. & HANS ZEISEL, THE AMERICAN JUR Y 146–47 (196 6); Ad m iss ion of Cr im in al H ist ories , supra note 34, at 741. 136. See FED. R. EVID. 413 and 414. 137. 1A WIGMORE, supra note 34, § 54. 1, a t 1 156 (footnote om it te d). D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1584 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998

III. RECEN T EXPANSION Do recent additions to the Rules of Evidence remedy or compound the problem? As detailed earlier,138 the character evidence rule has undergone a reversal of fortunes in sex offense and child molestation prosecutions by the addition s of Rules 413 and 414 to the Federal Rules of Evidence. As to criminal prosecutions for sexual assault139 or child molestation,140 the n ew rules simply provid e t hat evidence of the defendant ’s prior commission of the same offense “is admissible.”141 Desp it e t he m anda tory la nguage of the Rules, the case law quit e sen sibly indicat es tha t t he vir tu ally omnipresent bala ncing formula of Rule 403 reta ins its vitality even in th e context of the new rules.142 But the new rules have triggered a realignment of the potential probative and prejudicial uses of prior crimes evidence.143 The inference of criminal conduct from similar unchar ged conduct—formerly a potential misuse of the evidence which had to be factored into the “prejudice” side of the Rule 403 equation—is now a proper and endorsed deduction in th e circumsta nces governed by the new rules. The new rules then (unlike some of t he well- established evidentiary rules which at least pay homage t o the character evidence r ule even wh ile failing t o observe it in practice) represen t an over t rejection of the ch aracter evidence rule as it relates to the specified crimes. Rules 413 and 414 can be attacked and defended by marshaling argu men ts for and a ga in st the ch aracter evidence rule generally. But such arguments miss the mark because, even if accepted, they prove either t oo much or too little. If t he genera l propensity conclusion makes sense generally, then there is no principled reason for limiting this use of character

138. See supra notes 3–6 and accompanying text. 139. See FED. R. EVID. 413 (a). 140. See FED. R. EVID. 414 (a). 141. FED. R. EVID. 413 (a) a nd 414 (a). 142. S ee, e.g., United S tat es v. Sumn er, 119 F.3d 658, 661 (8th Cir. 1997); United Stat es v. Me ach am , 115 F.3 d 14 88, 1 492 (1 0t h C ir. 199 7); Un ite d St at es v . La rs on, 112 F. 3d 600 , 60 4–0 5 (2d Cir . 19 97). 143. “The practical effect of the new rules is to put evidence of uncharged offenses in s exu al a ssa ult an d ch ild m oles ta tion cas es on the sam e footing as other types of relevan t evidence th at ar e not subject to a special exclus ionar y ru le.” Sumner, 119 F.3d at 662. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1585 evidence solely to cases involving sexual a ssa ult s or child molestation. On t he other ha nd, if the objection to the new rules is the predicted cataclysm that will atten d t he r eversa l of the character evidence r ule, eve n for certain crimes only, then it is difficult to understand the defen se of the la w a s it exist ed pr ior to the new rules, with its permeable resistance to the introduction of such character evidence. The truly notable feature of the new rules is t he exclusive focus upon sexual offenses. While it may seem odd that evidence other wise deem ed to be pr eju dicia l should become relevant based solely upon the specific crime charged, there is a fair amount of historical support for this proposition. Crime- specific exceptions to the char acter evidence rule date at least as far back as 1871.144 In ma ny jurisdictions, there developed a “lus tful disposition” exception to the comm on law character evidence rule, sp ecifically allowing evidence of prior sexual crimes to demonstrate a disposition to commit such offenses.145 Of course, Ru le 4 04(b) of the Federal Rules of Evidence contain s no “lustful disposition” exception to the character evidence rule, and, wit h the a dop tion of st ate r ules of evidence patt erned after Rule 404(b), the “lustful disposition” exception ha s argu ably been abandoned.146 Nevertheless, courts have continued to avoid the character evidence rule and to admit character evidence against the accused in sex crimes prosecu tion s. 147 In many cases, su ch evidence is admitt ed under t he questionable rationale that it proves intent or a common scheme or plan, or for som e ot her authorized p urpos e u nder Ru le 404(b).148 But whether such evidence is allowed by a perpetuation of th e “lus tful disp osition ” excep tion or by t he cr ea tive manipula tion of the exceptions aut horized by Rule 404(b), the fact r ema ins

144. In th at year , a Br itish sta tu te pr ovided (contra ry t o the gener al comm on law rule) for the admission of evidence to establish a general propensity to receive and possess st olen pr oper ty i n p ros ecu tion s for th at ver y offense . S ee Ston e, supra note 41, at 978–80. 145. See 1 CHARLES E. TORCIA, WHARTON’S CRIMINAL EVIDENCE § 188, at 795 & n.17 (1985); 1A WIGMORE, supra note 34, § 62.2, at 1335. 146. See Imw ink elrie d, supra note 1, at 1127. 147. See 1A WIGMORE, supra note 34, a t 133 4–35; Im wink elrie d, supra note 1, at 1127. 148. See 1A WIGMORE, supra note 34, at 1335–36. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1586 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 that th e “lustful disposition” notion is far from dead in actual practice.149 So why t he n eed, or even the desire, to revitalize, or at least rest ate, the repudiation of the character evidence rule in sex offense cases? It ha s been suggested that the new rules constitute a first step in a scheme to eradicate the character evidence rule in criminal prosecutions.150 In any event, the measures, as pa rts of more comprehensive bills, failed to emerge from Congress in both 1991 and 1992.151 In 1994, the rela tively few supporter s of the new ru les of evidence wer e able to secure pa ssage of these provisions in excha nge for providing the n ecess ary m argin of su pp ort for a la rger crim e bill in wh ich the new rules became incorporated.152 The new rules, considered independently, never commanded un fettered support in Congress,153 and, as mentioned earlier,154 have suffered extensive cr it icis m since t hey e mer ged from Congress.155 Oddly, then, Rules 413 and 414 enjoy the imprimatur that accompanies historical roots, but suffer t he stigm a of owing th eir exist en ce t o a polit ica l dea l. But these fa cts a lon e do not dictate the result of our appraisal of the new ru les. Not all tha t can claim a long hist ory is worth while, just as not everyt hin g that is born of political back scra tching is wort hless. An evalu ation of the merit of the additions to the Federal Rules requires an exam inat ion of the proffered justifications for these rules. Tha t exa min at ion revea ls the followin g t wo con clu sions: first, that both the attackers and defenders of Rule 413 and 414 have done an excellent job of demonstrating th at the new rules may ma ke very lit tle differen ce as a pra ctical m at ter, a nd

149. See Thomas J. Reed, R ead in g Goa l R evis ited : Ad m iss ion of U nch arg ed Mi scon du ct Evidence in Sex Offense Cas es, 21 AM. J. CRIM. L. 127, 159 & n.181, 189 (199 3). 150. See 23 CHARLES ALAN WRIGHT & KENNETH W. GRAH AM, J R., FED ERAL PRACTICE AND PROCEDURE § 541 1, a t 2 63 (S up p. 1 997 ). 151. See 23 id. § 5411, at 263–64. 152. See 23 id. § 5411, at 264; J am es J oseph Dua ne, Th e New Federal Rules of Evidence on Pr io r A ct s o f A ccu se d S ex Of fe n d er s: A Poorly Drafted Version of a Very Bad Idea, 15 7 F .R.D. 9 5, 9 5–9 7 (19 95). 153. See Dua ne, supra note 152, at 96. 154. See supra notes 4–6 and accompanying text. 155. See 23 WRIGHT & GRAH AM, supra note 150, § 5411, at 264–65; Duane, supra note 152, at 96–97. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1587 second, that the new rules constitute an entirely unprincipled addition to an a rea of the law alr eady brim min g with disingenuous foolishness. To begin, many of the proffered justifications for Rules 413 and 414 are premised on the notion that the evidence allowed by those ru les is needed to establish many of the ve ry is su es for which prior crimes evidence is permitted un der Rule 404(b). F or example, one argument for the new rules is that evidence of th e defendant’s desire for nonconsensua l sexual grat ification is essent ially ak in t o proof of motive.156 But on that theory, the evidence would have been adm issible to prove “motive” under Rule 404(b).157 A second reason put forth for the new rules is that th e admitt ed evidence is necessar y to disprove the defendant’s claim of consent,158 but the concepts of “intent” and “absence of mistake” in Rule 404(b) have been stretched to allow prior crimes evidence to disprove a claim of consent and thus to demonst rate a pr opensity t owa rd inten tion al assaults.159 A thir d claim for the n ecess it y of Rules 413 and 414 is that su ch evidence will establish a pa tter n of condu ct identifying the defendant as the assailant.160 A varia tion of this theory is based on th e doctrine of chan ces: the improbability of an in divid ual suffer in g m ult iple fa lse a ccusations of sexu al assaults.161 But again, Rule 404(b) already allows evidence of prior crimes t o prove “identit y” in this context, 162 and no additional ru le seem s necessa ry t o apply the doctrine of chances to such evidence.163

156. See 137 CONG. REC. 3240 (da ily ed. Ma r. 13, 1991). 157. Of course, und er Rule 404(b), the t ri al cour t m igh t h av e ex clu ded su ch evid en ce on the groun d tha t th e potentially prejudicial use of the evidence to establish a gen era l propensity t o commit s exua l assa ult s out weighed th e proba tive value of the evide nce to pr ove motive. U nder Rule 413, a ba lan cing test must still be applied, see supra note 142 an d accompa nyin g text , but th e gene ra l propen sity conclu sion need no lon ger be a voide d. N eve rt he les s, e ith er th is p roffer ed ju st ificat ion for th e r ule is d isin gen uou s, or it s ugges ts th at th e r ule , for t he stated pu rp ose of proving mo ti ve, is a n e nt ir ely un ne ces sa ry du pli cat ion of Ru le 4 04(b ). 158. See 140 CONG. REC. 12,990 (daily ed. Sept. 20, 1994); 137 CONG. REC. 3238 (daily ed . Ma r. 13, 199 1). 159. See Dua ne, supra note 152, at 98–99. 160. See 140 CONG. REC. H2 433 (da ily e d. Ap r. 19, 199 4). 161. See 137 CONG. REC. 3240–41 (daily ed. Mar. 13, 1991); Imwinkelried, supra note 1, at 1130–35 ; Kar p, supra note 57, at 20. 162. See Dua ne, supra note 152, at 97-98. 163. See Imw ink elrie d, supra note 1, at 1136. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1588 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998

To the extent that the debate over Rules 413 and 414 has focused on the overlap with Rule 404(b), numerous pa radoxes have mater ia lized. F ir st , in an app aren t effor t to demonstrate that th e new rules are “not so bad,” their proponen ts ha ve likened these rules in practice to Rule 404(b), thereby un witt ingly supportin g th e opposit ion argu ment that the new rules are entir ely unnecess ary. Secon d, in an app aren t effor t to demonstrate that Rule 413 and 414 ar e entir ely un necessa ry, th eir opponents have hastened to em ph asize t hat su ch evidence will frequen tly, if n ot normally, get in un der th e preexisting Rule 404(b), thereby un int ent ionally documenting that the new rules are “not so bad.” Third, the opponents of Rules 413 and 414, wh ile pointin g t o Ru le 4 04(b) in theor y in order to pr otest this su pp osedly dr amatic new viola tion of the character evidence rule, also point to Ru le 4 04(b) in pr actice t o advance the argument that the government has no need for the new rules. Be that as it may, we are still left with the quest ion of why there should be a separate—indeed a converse—rule for sex offenses. Any practical similarity between the results under Rules 413 and 414 and the results under Rule 404(b) proves nothing except the extent to which Rule 404(b) falls woefully short of vin dicatin g t he ch aracter evidence rule. If the similar ity to Rule 404(b) in pr actice is all that commends the change in the new rules, then there is certainly no principled dist inction between these n ew rules and com pa rable rules for robbery, drug offenses, or any other crimes. One argument for Rules 413 and 414 that at least addr esses the right qu estion is tha t sex crimes (including child molestation) are particularly difficult to prove because th ere is often no corrobora tin g evidence to support t he victim’s testimony, and because, in child molestation cases, the child’s testimony is often regarded with skepticism.164 There are, however , seve ral fa tal proble ms w it h this pos it ion . First, if the pur pose of the new rules is to assist the governmen t in cases built exclusively upon the testimony of the victim, then the rules should say exa ctly tha t. Bu t curr ently,

164. See 140 CONG. REC. 12,990 (daily ed. Sept. 20, 1994); 140 CONG. REC. H8991–92 (daily ed. Aug. 21, 1994); 140 CONG. REC. 10, 276 (daily ed . Aug . 2, 199 4); 137 CONG. REC. 3238–39 (daily ed . Mar . 13, 1991); Ka rp, supra note 57, at 20–2l. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1589 the robber y prosecution based solely upon t he victim’s testimony falls outside the r ules , wh ile the r ape case in wh ich the victim’s testimony is confirm ed by DNA evidence, a videotaped confession and twenty disinter ested byst an ders falls within the rules. Second, the claim t ha t sex offenses are singular ly difficult to prove is unsubstantiated. Many other crimes (such as homicides, burglaries of unoccupied p rem ises, t heft s of unguar ded property and so-called vict im les s cr im es) do n ot offer the government even the testimony of the victim, and sex offense prosecutions are frequen tly aided by m edical a nd scientific evidence.165 Third, and most importa nt , if chara cter evidence truly is prejudicial, then why allow it only in the cases in which the government’s case is presum ably weakest?166 If the idea is to give the government a lit tle e xt ra ass istance in cases in wh ich it needs t he h elp, what is the lim itation on this operating principle? If we mak e the government’s task a little easier when it only has a little eviden ce, sh ould we make it s job a lot easier when it has no eviden ce at all? One very plau sible exp la nation for a lack of strong eviden ce of guilt is actual innocence. It is simply counter in tuit ive to expose only the most plausibly innocent to supposedly dan gerously prejudicial evidence. The truth is that Rules 413 and 414 cannot be pr emised upon a desire t o corr ela te t he a dm ission of prejudicial evidence with a weak governmen t case. Even t he sta un chest an ticrime legisla tor wou ld not pu rpos ely vot e t o permit convict ion s on the basis of evidence not fairly correlat ed to factual guilt. Whether stated explicitly or not, the necessar y premise of the new rules is that the ch aracter evidence a t issu e is seen as p roba tive, n ot prejudicial. Con sequ en tly, a much more genuine defense of Rules 413 and 414 sh ould focus on what it is th at is part icular ly probative about character evidence in sex offense cases.

165. See Edwar d J . Imw ink elrie d, Undertaking the Task of Reforming the American Character Evidence Prohibition: The Importance of Gettin g th e Ex peri m ent off on t he R igh t Foot , 22 FORD HAM URB. L.J . 285, 299–300 (1995 ); Raed er, supra note 4, at 350-51. 166. See Dale A. Nan ce, Foreword: Do We Really Want to Know the Defendant?, 70 CHI.-KENT L. REV. 3, 1 3 (19 94). D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1590 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998

Some effor ts h ave b een made to just ify t he n ew rules alon g these lin es. It has been su ggest ed, for exa mple, by one of t he Con gr essional sponsors of the new rules that “a history of [child molestation] ten ds to be exce pt ion ally pr oba tive because it shows an unusu al dispos it ion of the defenda nt—a sexu al or sado-sexual interest in children —that simply does not exist in ordinary people.”167 It has also been suggested that persons who have a h istory of sexual crimes poss ess a “combin ation of [uninh ibited and undeterred] aggressive and sexual impulses” not shar ed by people gen erally.168 Once again, however, these arguments for Rules 413 and 414 do not ta ke u s where we need t o go. No doubt ma ny people who commit all sorts of crimes possess certain desires and disp osition s that are different from those people who manage not to run afoul of the penal codes. There is nothin g in pr inciple unique about sex offenders. Drug users may be addicted, thieves gr eedy, a ss aila nts violent, a nd s o on. It is simply n ot very difficult to identify character traits associated with particular categor ies of crim in al con du ct. To d o so d oes not explain wh y a differ en t rule for character evidence should contr ol for one su ch cat egory.169 On this point, the opponents of Rules 413 an d 414 ha ve taken the offensive. The counter argument is that, because recidivism for sexual offenses is not higher than for other crimes,170 and indeed is actually lower than for many other crimes,171 th e special valu e of chara cter evidence in sex crime offenses is thereby disproved.172 It is true that the absence of a higher rate of recidivism for sexual offenses removes one possible a rgu ment for special

167. 140 CONG. REC. H899 1 (daily ed . Aug. 21, 1994) (statement of Rep. Moli na ri ). 168. See Kar p, supra note 57, at 20. 169. See 1A WIGMORE, supra note 34, § 62.2, at 1345; Reed , supra note 149, at 219. 170. See Raed er, supra note 4, at 350; Reed , supra note 149, at 149 & n.117. 171. See 23 WRIGHT & GRAH AM, supra note 150, § 5412, at 274; Kater ine K. Baker, Once a Rapist? Motivational Evidence and Relevancy in Rape Law, 110 HARV. L. REV. 563, 578 (1997 ); Duan e, supra not e 15 2, a t 1 13; Da vid P . Leon ar d, The Federal Rules of Evidence and the Political Process, 22 FORDH AM URB. L.J . 305, 339 (199 5). 172. See 23 WRIGHT & GRAHAM, supra note 150, § 5412, at 274; Bak er, supra note 171, at 578 -79; Dua ne, supra note 152, a t 113 ; Leona rd, supra note 171, at 339. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1591 treatment of such crimes. It is not, however, true that the only way that a sh owin g of a rela tively gr ea ter pr oba tive va lu e for chara cter evidence in sex crime prosecutions can be established is by showing a great er degree of recidivism for such offenders. Suppose, for example, that the recidivism rate for both rape and theft is fifty percent, but that the rate for each crime committed by same-crime repeat offenders is twenty per cen t for thefts and eight y per cen t for rapes. In this cir cumst ance, p rior crimes evidence for rapes would be relatively more pr obat ive than prior crimes evidence for thefts because t he former gives us a r elatively mu ch greater lik elihood of identifying th e criminal in a particular case based upon prior conduct. Or suppose that fifty percent of all crimes (includin g both rape and theft) ar e committ ed by prior offender s wh o have pr eviously committed the same crime. Suppose also that there are 100,000 prior th eft offender s and only ten prior rape offenders. Entir ely independent of the r elat ive recidivism r at es, as well as the above-hypothesized repea t offen der ratios, prior crimes evidence in the rape case would be more probative than the same evidence in the t heft case simply because the raw numbers enable us to nar row down m ore substan tia lly a plausible gr oup of s usp ect s in the r ape scenario. Be this as it may, these alternative possible methods for demonstrating the super ior probative value of character evidence in th e prosecutions covered by Rules 413 and 414 are of no avail in defense of these provisions because there is no actual da ta yieldin g a ny of the a bove conclu sions. In fact, the new rules are sim ply devoid of a ny em pirica l supp ort.173 There is no basis wha tsoever for believin g t hat pr ior crim es evidence is more probative in th e situations governed by Rules 413 and 414 than it would be in the prosecutions of other crimes governed by the pr eexisting, contrary (in theory) character evidence rule.174 But even if recidivism rates or other data supported an inference of heigh ten ed pr obative value in sex crime prosecutions, the crime-specific amendments in Rules 413 and 414 would be unacceptable. Fir st, a difference in degree ough t not to account for a completely converse rule. If character

173. See J ud icia l Con feren ce, supra note 4, at 53. 174. See 1A WIGMORE, supra note 34, § 62.2, at 1345. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1592 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 evidence were marginally more probative in sex offense cases, one should expect at most a r elative relaxa tion of the character evidence rule, not what appears from the face of the rules to be a complete about-face from a rule of exclusion to a complete rule of inclusion. Second, the virtue and success of the F eder al Ru les of Evidence can be attributed precisely to their universal application, gener ally making no distinctions among litigants and subject matter, and imperviou s t o va ria ble polit ica l agendas and shifting statistics.175 Rules 413 and 414 are wrong for all the reasons that the Federal Rules of Evidence are gener ally right. Consistency and int egrit y dictate t ha t our r ule on character evidence not vary from crim e to crim e with t he winds of today’s part icular cause celebre.176 Fin ally, a s has been det ailed above,177 the state of the la w of evidence rega rding ch aracter evidence prior to the enactmen t of Rules 413 and 414 wa s already a h opeless ly complex, un workable and largely disin gen uous m orass . Like t he t op floor of a st ructurally unsou nd build in g, t he n ew rules not only do nothing to remedy the preexisting disorder, they also compound the problem by adding yet another unprincipled and indefensible distinction to the character evidence rules. If we are to do better than this, we need to pursue two goals. First, deconstruction is in order. Noth ing significant is to be gained by further building upon the present disarray. And second, we need t o discover the t rue, w orthwhile ju st ifica tion s, if any, for the chara cter evidence rule, an d thu s iden tify what is wort h preser ving a nd what is in need of discar ding.

IV. REASONS FOR THE CHARACTER EVIDENCE RULE Perhaps no single ration ale a ccounts for the character evidence rule.178 Variou s explanations for the rule have been advanced, in clu ding t he suggest ion th at th e ru le is, at lea st in part, “a senseless product of history.”179 Bu t before retreating to that conclusion, an examination of the more flatt ering pr offered

175. See Leona rd, supra note 171, at 341. 176. See Dua ne, supra note 152, at 112. 177. See supra notes 7–137 a nd a ccompan ying te xt. 178. See 1A WIGMORE, supra note 34, § 54.1, at 1150–51. 179. 1A id., § 54.1, at 1151. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1593

just ifications for the r ule is r equir ed. Th ese defen ses of t he character eviden ce ru le can be described in th e following five formulations: 1. character evidence is exclud ed b ecaus e it is essen tia lly

irrelevant;180 2. character eviden ce is exclud ed b ecaus e of th e like ly danger tha t such evide nce w ill be afford ed too m uch weight by juries;181 3. character eviden ce is exclu de d, a t lea st wh en offer ed against the accused, because of the likely danger that the jur y will convict t he d efenda nt t o exact punishment for the prior, uncharged misconduct;182 4. character eviden ce is exclu de d, a t lea st wh en offer ed against the accused, because otherwise the accused will not ha ve been provided fair notice as to the allega tion s of m iscon du ct a s t o wh ich a de fen se needs to be prepared;183 and 5. character evidence is excluded because it involves time-consuming vent ures into collateral matters.184

Ea ch of these rat ionales will be discussed in tu rn .

180. See Migue l Angel Men dez, California’s New Law on Cha ract er E vid ence: Evidence Cod e S ection 352 an d t he I m pact of R ecent Psy chol ogica l S tu di es, 31 UCLA L. REV. 1003, 1041 (1984) (“[S]cientists question whether character evidence has any proba tive va lu e a t a ll.”). 181. See Old Ch ief v. United Sta tes, 51 9 U.S. 172, 180 -82 (1997); Michelson v. United St at es, 335 U .S. 4 69, 4 75–7 6 (194 8); 2 LOUISELL & MUELLER, supra note 28, § 136, at 128–29; 1A WIGMORE, supra note 34, § 29a, at 979, § 58.2, at 1212–15; 23 WRIGHT & GRAHAM, supra not e 15 0, § 54 16, a t 3 07; Adm ission of Criminal Histories, supra note 34, at 730–31; Im wink elrie d, supra note 63, a t 18; Im wink elrie d, supra note 1, at 1138; Imwinkelr ied, supra note 165, a t 289 –90; Im wink elrie d, supra note 7, at 581–82; Me nde z, supra note 180, at 1044; Ra eder , supra note 4, at 349; Stone, supra note 41, at 957. 182. See 2 LOUISELL & MUELLER, supra note 28, § 136, at 130; 1A WIGMORE, supra not e 34 , § 29a , at 979, § 57, a t 1 185, § 58.2 , at 1212 –15; Adm ission of Criminal Histories, supra note 34, at 730–31; Dua ne, supra note 152, at 110; Imwinkelried, supra note 165, at 290–91; Imwinkelried, supra note 7, at 580, 582; N an ce, supra note 166, at 11; Ra eder , supra note 4, at 349; Reed , supra note 34, at 166. 183. See 2 LOUISELL & MUELLER, supra note 28, § 136, at 130–31; 1A WIGMORE, supra note 34, § 58.2, a t 121 2-15; Edwa rd J . Imw ink elrie d, The Need to Amend Federal R ul e of E vid ence 4 04(b ): Th e Thr eat to t he Fu tu re of th e Fed eral Ru les of Evidence, 30 VILL. L. REV. 1465, 1489 (1985); Kar p, supra note 57, at 27; Reed, supra note 34, at 167; St one, supra note 41, at 957-58. 184. See 1A WIGMORE, supra note 34, § 58.2, at 121 3 n.2; Ku hn s, supra note 60, at 777; Na nce, supra note 166, at 11. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1594 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998

A. Relevance The ar gum ent th at chara cter evidence is irr elevant is just plain wrong; an exh au stive list of authorities on th is point—many supportive of the general ban on character evidence185—is in accord. The notion that character traits in flu en ce behavior is so much a part of common sense and everyday exp er ien ce186 th at th e contr ar y suggestion is really quite r emarkable.187 Imagine returning home from work t o find a shattered living room win dow and a br ick restin g a mid br oken gla ss on the living r oom floor . Without more, t her e would be a wor ld of equally plausible culprits. But suppose that the neigh bor hood contained an individual who had been apprehended on several occasions in the recent past throwing br icks thr ough t he living room win dow s of ot her homes. Wh ile by n o mea ns d ispos it ive of that individual’s com plicit y on the occa sion in qu estion , such information would unquestionably be relevant .188 Only a fool would fail to include such information in a report to the police, and only the disingenuous would claim that they would not report it beca use of th eir comm it men t to the fa ncy that chara cter eviden ce is irrelevant . In fact, the law’s trea tm ent of chara cter evidence is confir mation of the relevance of such evidence.189 The criminal defendant is unqu alifiedly p er mit ted to in trodu ce opin ion and reputation evidence of his or her own character,190 as well as that of the crime victim,191 and sur ely such evidence would be

185. S ee, e.g., Old Ch ief, 519 U.S . at 180-8 2; Mi chels on, 335 U.S. at 475–7 6; 2 LOUISELL & MUELLER, supra note 28, § 135, at 117; 1A WIGMORE, supra note 34, § 54.1 , at 1212 ; Adm ission of Criminal Histories, supra note 34, at 709; Imwinkelried, supra note 165, at 289. 186. See Davie s, supra note 107, a t 511 ; Imwin kelr ied, supra note 94, a t 379; Uviller, supra note 12, at 883. 187. See Na nce, supra note 166, at 3. 188. The test of relevan ce is simp ly whet her th e eviden ce has “any ten dency t o make th e ex ist en ce of a ny fact th at is of consequence t o th e d et er mi na ti on of th e act ion mor e pr oba ble or les s pr oba ble t ha n it wou ld be without th e eviden ce.” FED. R. EVID. 401. On the question of relevance, it is simply im ma te rial w he th er or n ot the evidence in question, by itself, is sufficient to establish th e objective of its proponent. 189. See Na nce, supra note 166, at 3-4. 190. See FED. R. EVID. 404(a )(1); see supra notes 10–12 a nd a ccompan ying te xt. 191. See FED. R. EVID. 404 (a)(2 ). D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1595 disallowed if it wa s irrelevant .192 Character evidence is adm itted to est ablish or at ta ck th e credibilit y of a witness.193 Prior cond uct evidence a dm it ted under Ru le 404(b) m ay technically fall outside t he la w’s definition of character evidence, but its rele va nce is neve rthele ss pr em ised upon the assumed cons isten cy of an individua l’s behavior.194 The same is true of evidence of h abit , a dm it ted under Feder al Ru le of Evidence 406.195 Evidence of past conduct to establish propensity is routinely considered in determ ining pretr ial release a nd sen tencing.196 Evidence excluded by the character evidence rule is no less relevant than the evid en ce a dm it ted pu rsu ant to the va riou s exce pt ion s to, an d distin ctions from, the r ule of exclusion .197 Whether based upon common anecdotal experience or statistical evidence of high recidivism rates,198 we quite sensibly assign significance, for exa mple, t o the prior br ick toss es of th e above-hypothesized neighbor because he or she has demonstrated a disposition to cross a line that m ost people will not cross.199 Nevertheless, some commentators have suggested that the nonexistent, or at least trivial, probative value of character evidence is est ablished by psychological studies su ggesting that character is a poor pr edict or of behavior , which is determined to a greater extent by situational factors.200 There are th ree flaws in this pos it ion .

192. See 1A W IGMORE, supra note 34, § 55, at 1157-59. 193. See supra notes 109–137 a nd a ccompan ying te xt. 194. See Roth ste in, supra note 71, at 1264–65. 195. See 1A WIGMORE, supra note 34, § 54.1, a t 115 6; Roth ste in, supra note 71, at 1264–65. 196. See Adm ission of Crimin al H ist ories , supra note 34, at 735–36; Kar p, supra note 57, at 51. 197. See 1A WIGMORE, supra note 34, § 54.1, at 1157 n.2. 198. See Bak er, supra note 171, at 578–79 & n.80; Raeder, supra note 4, at 350. 199. For th is rea son, contra ry t o what Mend ez suggests , it should not be “troublin g . . . that jurors may give greater weigh t t o eviden ce of m iscon du ct a nd dishonesty tha n to favorable evidence.” Mendez, supra note 180, at 1045. It is a happy circum sta nce th at most people do not commit crimes. Therefore, crim ina l, rat her tha n noncrimina l, conduct is distinctive and especially relevan t. Moreover, it is per fectl y se ns ible to focu s u pon ne gat ive con du ct a s a tr ue ind icat or of character. People we know to be honest do not generally lie, but people we know to be dishonest do often t ell t he tr ut h. I t is th e pa tt er n of even occas ional fabr icat ion th at , in comm on exp er ien ce, ca n q ua lify on e for th e la tt er re pu ta tion . 200. See Imw ink elrie d, supra note 7, at 582; Men dez, supra note 180, at 1052. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1596 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998

First, th e more recent psychologica l literature supports the proposition that character evidence is indeed r elevant .201 It suggests that character traits do have predictive value,202 even in cross-situational circumstances.203 Second, th e cita tion of psychological litera tu re in th is context ha s all the flavor of a welcome discovery of a new just ifica tion for travelin g on a ship th at actua lly left the shore a very lon g t im e a go. The ch aracter evidence r ule, in one for m or another, has been around for a few centuries.204 It seems un likely that it originated from a prescient anticipation of late twentieth centur y psychological theories suggest ing t ha t t he concept of behavior based on character traits is all premised upon misconception . Mor eove r, t he shift in g consensu s a mong psychologists has made it rather precarious to premise one’s position upon this literature.205 Third, even if empirical research did suggest that character traits are poor indicators for predicting human behavior, it would sh ed lit tle ligh t on the question of the lega l r ele va nce of character evidence. This is because it is not the task of the jury to predict the defendant’s behavior. If we could pr edict human behavior, we could presumably dispense with our current after- the-crime syst em of adjudication and inst ea d d evise som e preemptive method of crime control. But we cannot predict behavior, an d consequ ent ly wha t m ay be of enormous interest to psychologists as theories of huma n beha vior ha s itself little relevance to the evidentiary issu e of relevance.206

201. See Davie s, supra note 107, at 506, 516–17, 533; Imw ink elrie d, supra note 27, at 37. 202. See Davies, supra note 107, a t 516 –17; Lyn n McLa in, New Federal R ules in Sex Off ens e Cas es, 25 MD. B.J . 34, 37 (1 995 ). 203. See Andre w E. Ta slitz, Myself Alone: Individu alizin g Ju stice Th rough Psychological Ch ara cter Ev id ence, 52 MD. L. REV. 1, 33 (1 993 ). 204. See supra notes 42–52 a nd a ccompan ying te xt. 205. Compare Imw ink elrie d, supra note 7, a t 5 82 (w ri ti ng in 199 0: “[T]h e empirical stud ies ind icat e t ha t t he gen er al con st ru ct of ch ar act er is a re lat ively poor pr edi ctor of a per son’s conduct on a given occas ion. . . . In light of the a vailable studies, we can have little confidence in the construct of character as a predictor of condu ct.” (footnote s om it te d)) with Imwinkelried, supra note 27, at 45 (writin g in 1994: “[T]he pr eva ilin g . . . i nt er act ion ist th eor y . . . cu ts in favor of liber alizing t he [admission of] chara cter e vidence,” but “we should pe rh aps be hes itant to legislate on the ba sis of th e cu rr en t p opu la ri ty of th e in te ra cti onist th eor y.”). 206. In this respect, what Profess or I mw ink elr ied ha s ca ut ione d in 1994 ab out relying upon a cur ren tly popu lar psychological th eory is a bsolut ely correct . See Imwinkelried, supra note 27, at 45. It wa s a lso cor re ct a t t he ti me of Professor D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1597

To illustrate, we can return to our hypothetical brick- thrower. The fact that our suspect has th rown bricks t hr ough living room win dows on several prior occasions hardly enables us to pr edict wh en he or sh e will do s o again. In fact, we can add to our hypothesis that, during the same period of time that our suspect threw four bricks through four windows, he or she passed by some four thousand windows without flingin g anything in their direction. Even assuming that past performance is an accurate in dicat or of futu re behavior, a prediction th at a par ticular window will suffer a shattering fate on a particular occasion at the hands of our suspect would have only a bou t a one in one t housa nd ch ance of bein g cor rect . Does this render evidence of our actor’s prior adventures irr elevant ? Of course not. To insist that evidence ha ve predictive value as a condit ion of rele va nce is to im pos e u pon such evidence at the a dm issibilit y stage a requir em en t of sufficiency that need on ly be m et by t he cu mula tive total of a party’s evidence when that party rests its case.207 The question should not be whet her the character evidence is sufficient to establish guilt ; invaria bly that will not be th e case. The rea l issue is whether t he char acter evidence could be a cont ribu tin g piece of evidence in a case esta blishing guilt.208 For example, in a pr osecution of our br ick-chuckin g susp ect for the most recent incident, without more than the defendant’s history of similar crimes, the government could not survive the defendant’s motion for judgm ent of acquitta l at th e close of its case. But suppose that the government’s case also included testimony that the defendan t was, minu tes before the crime, seen hea ding in t he direction of the relevant home, ca rrying a br ick and muttering threats against the resident of that home. In add it ion , im agin e t estim ony t hat, m in utes after the cr ime, the defen da nt wa s s een runnin g in a direct ion awa y fr om the relevant home carrying nothing. Now the governmen t’s case is surely one that will reach the jury, wh ich migh t well con vict on such evidence. And although the eyewitness accounts of the

Imwinkelried’s art icle four years earlier when a different theory of hu ma n b eh avi or was th en in vogue. See supra note 205. 207. See Davie s, supra note 107, at 518. 208. See David J. Kar p, R espon se to Prof essor Im wi nk elri ed’s Com m ent s, 70 CHI.- KENT L. REV. 49, 52 (1 994 ). D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1598 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 defendan t’s conduct immediately before and after the crime ar e most importan t, t he prior crimes evidence is a lso plain ly relevant, as it logically strengthen s the government’s case. As a final variation on our hypothetical, suppose that the defendant is only one of three suspects with a track record of throwing br icks t hrough win dow s. Th is alt er s t he a nalys is not at all. Because th e chara cter evidence makes it increment ally more likely that one of the three prior offenders (as contrasted with the general population of those who have historically abstained from throwing bricks through windows) committed this crim e, s uch evidence is relevant. The fact that there are two others equally dama ged by such character evidence goes only to the weight ass ign ed by t he ju ry t o this evidence and mer ely increases th e sign ificance of th e eyewitness accounts, testimony which incriminates only the particular defendant.209 Character evidence, then, is manifestly relevant. Or, perhaps a better way to state the proposition is as follows: evidence which is sensibly and logically relevant does not become irr elevant simply because it is cha ra cter eviden ce. If there is a good r eason to exclude such evidence, it will have to be for some reason other than the bogus claim that such eviden ce is not relevant .

B. Da nger of Ov ervaluat ion The second (and more soph istica ted) argu men t for the character evidence r ule is t hat jurors t en d t o overva lue the significa nce of such evidence.210 The initial question that must be an-sw er ed in eva lu atin g t he m er it of this argu men t is: How

209. See Tas litz, supra note 203, at 68. Professor Taslitz also hypothesizes the sit ua tion in which, because of distin ct, ind ividua l aggr egat es of char acte r t ra its, t he lik elih ood of th e de fen da nt ha vin g committed th e char ged crim e is only 4.5%, while the likelihood of another individual ha ving committed th e charged crime is on ly 0.9 %. Even though th e character evidence does not make it likely that the defendant committed the crime, it is clearly relevant because this eviden ce does ma ke it five times more likely that the defendan t committed t he crime tha n th at t he other individual commit ted th e crim e. See id. As in the circumstances hypothesized in the text, th e govern men t would requ ire m ore tha n jus t t his char acter evidence to establish a prima facie case. 210. See supra note 181. In part icular, the argum ent is that jurors, having concluded th at an individu al has a propen sity t oward a pa rt icular beh avior, will overestimate the va lue of such evidence in resolving th e question of wh et he r or not that same individu al b eh ave d in accor da nce wit h t ha t ge ne ra l pr open sit y on t he occas ion in qu est ion. S ee, e.g., Imw ink elrie d, supra note 7, at 581. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1599 do we know that this is true? Perhaps we should inquire even more fundamentally: How could we know th at this is true? The starting point for addressing these inquiries is the often-asserted proposition that jurors are influenced by character evidence and, in particular, by evidence of prior similar crimes committed by the accused in a criminal case.211 This proposition is almost certainly true and, in any event, can be accepted as true for the purpose of our inquiry. But to state that jurors are persuaded by character evidence, without more, is merely to state that such evidence is relevant. Jurors are un doubt edly also persuaded by DNA evidence, fingerprints, confessions, eyewitness accounts, and the film from surveillance camera s, but it would be ludicrous to exclude such evidence sole ly on accoun t of its p er su asive effect. To state the matter in a different fashion, eviden ce is not unfairly prejudicial simply because it tends to improve the government’s chances of obt ain in g a convict ion .212 The pr oble m is that it is a huge lea p fr om the proposition that jurors assign value to character evidence t o the con clu sion that jurors over value chara cter eviden ce. One way to bridge this gap is to insist t ha t char acter eviden ce ha s no proba tive value at all, in which case the assignment of any weigh t to su ch evidence would constitu te “overvalua tion.”213 But that analysis fails for the simple reason that, as discussed above,214 character evidence often does have probative value. Nevertheless, there is no short age of scholar ly auth ority in support of the proposition that jurors tend to overvalue character evidence.215 Consequently, the proposition has acquired th e stat us of a certain convent ional wisdom, with commentators often either blithely assuming it to be true or citing to each other in what migh t be no more than a tautological house of cards. Consider first of all the paradox of the role of popular opinion in the equation supporting the proposition that

211. See Imw ink elrie d, supra note 1, at 1138; Im wink elrie d, supra note 7, at 581; Ordover, supra note 18, at 144. 212. See Admission of Cri m in al H ist ories , supra note 34, a t 731 ; Kar p, supra note 57 at 22. 213. S ee, e.g., Mendez, supra note 180, at 1041–44. 214. See supra notes 185–209 a nd a ccompan ying te xt. 215. See supra note 181. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1600 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 over va lu ation occurs. If the danger of overvalua tion is sign ifica nt, it must be that most, if not all, jurors overvalue character evidence. To illu st rate, s upp ose that the va lu e of evidence could be measured on a scale of one to ten, with ten being the greatest value. If the danger of overvalua tion is real, it must be that there is a correct va lu e of ch aracter evidence (assume “three” on our hypothetical scale), but that most, if not all, jurors will assign such evidence a higher value (assume “five” on our hypothetical scale). Now, if it is true that most jurors believe t hat the cor rect value of chara cter evidence is “five,” and if a consensu s of op in ion is a va lid pr oxy for accuracy, th en t he a ctual correct value of the character evidence is “five,” and there is no over va lu ation wh atsoever . F or the danger of overvaluation to be real, then it must be the case that what most people believe to be true (i.e., that character evidence has a value of “five”) is in reality false. If popular opinion is thus discredited as a measure of reality in the a bove context , t hen how ca n the propos it ion that jur ors overvalue character evidence be ju st ifia bly pr em ised on the notion that the stated proposition is commonly believed to be true? Th e exclu sion of relevant evidence ou gh t to be ba sed u pon something more than popular myth. But the fact is that, popular rh etoric aside, t her e is no ba sis for the con clu sion that jurors overvalue character evidence.216 Moreover, if the focus is

216. See Kar p, supra not e 57 , a t 2 7. One au th ori ty som et im es cit ed a s emp ir ica l support for t he pr oposi tion that jurors overvalue character evidence is The Am erican Jury. S ee KALVEN & ZEISEL, supra note 135. For example, Professor Imwinkelried has written: As par t of th e Chica go Ju ry P roject, r esearchers attempted to determine the im pa ct of a d efe nd an t’s p ri or cri mi na l r ecor d on th e p roba bil it y of conv ict ion. The re se ar che rs foun d t ha t conv ict ion ra te s w er e s ign ifica nt ly gr ea te r a fter a jury learn ed that t he defenda nt h ad a crim inal record or had been charged with even a minor crime. The researcher s concluded that jur ies aware of prior misconduct employ an entirely “different . . . calculu s of probabilities” to determ ine the defenda nt’s guilt or innocence. Imwinkelried, supra note 183, at 1487–88 (footn ote s om itt ed). I n fa ct, P rofe ssor Imwinkelried is incorrect as to what the r esearchers were inten ding to determ ine, what they did det ermine, and what inferences can be drawn from what they did determ ine. Gathering information from sur veys completed by judges who presided over criminal jur y tr ial s, P rofe ssor s Ka lven an d Zeis el a ctu all y sou ght “to an swe r t wo basic questions: First, what is th e magnit ude an d direction of the disagreemen t between judge and jury? And, second, what are the sources and explanations of such dis ag re em en t?” KALVEN & ZEISEL, supra note 135, at 55. The authors found that D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1601

placed on the question of whether jurors overvalue character evidence (as opposed to the almost meaningless question of whether jurors value chara cter evidence), ther e simply is no empirical support for the overvaluat ion hypothesis. 217 The fact that the overvaluation hypothesis is w it hout any demonstra ble su pp ort sh ould not rea lly be su rpr isin g. F or the hypothesis to be true, we would need t o be able to determine

judge and jury a gre ed on th e ver dict in 75.4% of crim ina l tr ials an d th at a h un g jur y resulted in a nother 5.5% of such t ria ls. See id. at 56. Thus, excluding hung juries, judg e-jur y disagr eeme nt on guilt or inn ocence occurre d in only 19 .1% of all tria ls. This 19.1% is divided bet ween th e 16.9% of trials stu died in which th e judge would ha ve convicted in cases in whi ch t he ju ry act ua lly a cqu it te d, a nd th e 2. 2% of t he tr ials in wh ich th e judge would have a cquitt ed in ca ses in which the jur y actu ally convicted. See id. Even when cases in which t he ju ries h un g an d cases in wh ich the judge an d ju ry dis agr eed only as to t he cha rge we re incl ud ed i n t he qu an tificat ion of judg e-jur y disagr eement s, su ch disa gree men ts occur red in only 30% of all cases. See id. at 110. This disagreement was found to be “massively in one dir ection,” with t he jury acquits/judge convicts split occurring more tha n six times as often as the jury convi cts /judge acqu its s plit. Id. at 58. The a ut hors th us defined t he cas es resulting in the former split as “normal disagreemen ts” and the cases r esulting in the latt er split as “cross-over disa gree men ts.” Id. at 110. The bulk of the book, in cluding t he por tion err oneously relied u pon by P rofessor I mwin kelr ied, is an att empt t o discover “reasons for th e nor ma l disa gree men t . . . .” Id. at 119. In the port ion of the book relied upon by Professor Imwinkelried, the authors concluded from the empirical data that there existed a strong correlation between normal disagreemen ts an d close cases involving serious charges in wh ich the defendant testified and ha d no cr imin al r ecord. See id. at 177– 81. (P rofe ssor Imwinkelried’s reference to the impact of th e jury learning that the defendant “had been charged with even a minor crim e,” Imwin kelr ied, supra note 183, at 1487, complete ly misreads the data in The Am erican Jury. Professors Kalven and Zeisel do distinguish between “serious crimes” and “minor crimes,” but the distinction refers to the crime for which t he de fenda nt was t hen on tr ial, not to the nature of the defendant’s prior crim ina l recor d. S ee KALVEN & ZEISEL, supra note 135. a t 180 .) What this means is th at t he abs ence of a cr imin al r ecord, in combination with other factors, influenced th e jur ies to a cquit in cases in which t he ju dges would ha ve convicted. The cor olla ry of th is con clus ion i s t ha t, i n s ome cas es (d epe nd ing up on t he alignment of th e ot he r fa cto rs me nt ion ed ab ove) in which the ju ry doe s le ar n of th e defendant’s criminal recor d, the jur y is more likely to behave as does th e judge. If we accept th e ju dge ’s view a s t he “corr ect ” valu at ion of th e ev ide nce , t he n t he em pir ica l data, far from demonstr ating that the jur y overvalues prior crimes evidence, actu ally shows that the jur y overvalues t he absen ce of prior crimes evidence. If this were true, then a rule which excludes prior crimes evidence actu ally crea tes t he su pposedly un desir able danger of overvaluation, albeit in the defendant’s favor in most cases. Of course, on e migh t qu ite r eason ably qu ar rel wit h t he pr emise that th e judge assigns the “correct” value to such character evidence. But if the judge does not know the “correct” value of this evidence, then who does? And if no one know the “correct” value of character evidence, then on what possible basis can it be asserted that jurors overvalue such evidence? 217. See Admission of Crimina l H ist ories , supra note 34, a t 710 ; Davies , supra note 107, at 533. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1602 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 two things. First, we would need to be able to deter min e h ow jurors actually assess character evidence.218 Second, we would need to be able to deter mine t he correct valuation of character evidence as a benchmark for mea su rin g t he a pp ropr ia ten ess of the ass essm en t of su ch evidence by jurors.219 Unremarka bly, we simply la ck the ability to make either one of these necessary determinations. As to the fir st requir em en t—the det er min ation of how jurors ass ess ch aracter evidence—t he immediate problem is the unstated assumption that jurors ar e fairly homogeneous in th eir appraisal of such evidence.220 Jurors, however, represent virtually the entire range of human attitudes and perceptions, so it is hardly u nexp ect ed that em pirica l studies in dicate that jurors ha ve very diverse r eactions to, and interpretations of, items of evidence.221 And even if this rather presumptuous stereotyping of jurors wer e tr ue, judges or legal scholars would still have to determ ine wha t the juror assessment of character evidence would actually be.222 There is, however, nothing about a legal ed ucation that qu alifies one to predict human reactions to evidence,223 and again the empirical data does not suggest that lawyer s and judges ar e part icular ly capable of accura tely assessing the reactions of jurors to evidence.224 As to the second requirem ent—the determination of the correct valuat ion of chara cter evidence—similar problems exist. Empirical data suggests that there is no more agreement among lawyers and judges as to how to assess evidence than exists among people in gener al.225 And even if t her e wer e some consensus among t hose wit h a legal ed ucation about the correct value of character evidence, what reason would there be to believe that assessment to be superior to that of those individuals with different training an d experience?

218. See Teit elbau m et al., supra note 60, at 1153, 1193. 219. See 1 WIGMORE, supra note 34, § 10a, at 688. 220. See Teit elbau m et al., supra note 60, at 1153, 1193. 221. See id. at 1163. 222. See id. at 1153, 1193. 223. See Victor J . Gold, Limiting Jud icial Discretion to Exclude Prejudicial Evidence, 18 U.C. DAVIS L. REV. 59, 85 (1 985 ). 224. See Teit elbau m et al., supra note 60, at 1166, 1176, 1193–95. 225. See id. at 1165, 1172, 1193. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1603

[T]h e claim th at th e judge ha s su per ior ex per ien ce or in sigh t is

usually simply gratuitous an d u ns up por ted. H ow does sitt ing in a cou rt room , as opposed to, for exam ple, w ork ing in a fact ory or rea ding n ovels or raising a family or anything else con fer such superior experience and insight? Might one not say with equa l force that courtr oom “experience” or le ga l pr actice or tr aining dist ort s in sigh ts into h um an m otiv at ion an d character?226

In the end, th e proposition that jurors overvalue character evidence is unsupported and unsupportable. No empirical data suggests that jurors overvalue such evidence. Moreover, there is no qu antifica tion of the va lu e t hat jurors a ss ign to su ch evidence, and there is no reason, or even consensus, as to the value that ought to be assigned to such evidence. The flaws in the overvaluat ion hypothesis do not end t here, however. What is fundamentally unacceptable about the overvaluation hypothesis is the prem ise t hat the va lu ation of relevant evidence should not be exclusively within the provin ce of the jury. This pr em ise is one of th e la st ves tiges of a n antiqua ted distrust of the jur y and its ability t o perform its essentia l fa ct-fin ding fu nction . If, as is often the case with char acter evidence, evidence is relevant and the sole issue is the weight to be attributed to such evidence, it is axiomatic tha t such evidence is admissible, and it is for th e jur y to determ ine what weight to assign to the evidence.227 Many older, common law evidentiary rules (particularly those related to the competency of interested witnesses) excluded eviden ce for fea r that jurors w ould not appreciate the reduced value such evidence should be given, but the history of the law of evidence is a movement away from such exclusionary rules.228 It is true, of course, that courts have authority to exclude relevant evidence if the probative va lue of such eviden ce is substan tia lly outweighed by the danger of unfair prejudice.229

226. 1 WIGMORE, supra note 34, § 10a, at 688 n .22. 227. See 1A WIGMORE, supra note 34, § 29, at 976–78. 228. See CLEARY ET AL ., supra note 15, § 61, at 155, § 71, at 168; Karp, supra note 57, at 29. 229. See FED. R. EVID. 403. In addition to “unfair prejudice,” Rule 403 also speci fics “the da nger [s] of . . . confusion of the issues [an d] mislea ding th e jur y” as D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1604 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998

But, at least ordinarily, this authority is not understood to justify the exclusion of relevant evidence in cases in which the only possible ha rm is the potent ial for t he jury affording the evidence too great a significan ce. The notion of balancing probat ive value against unfair prejudice inherently presupposes that th ere exist two sepa ra te uses t o which th e evidence m ight be put, one legitimate and one illegitimat e, an d the question is whether the danger of the latter outweigh s t he va lue of t he former.230 The matter can be illustrated as follows:

EVIDENCE

Process of Process of impermissible permissible rea son in g fr om, or reasoning em otive rea ction to, from evidence evidence

Per missible in fer en ce Im permissible in fer en ce from evidence from evidence (PROBATIVE VALUE) (UNF AIR PREJU DICE)

In this two-pronged schem e, t he danger of unfair pr eju dice arises from a potential misuse (as opposed to a misweighing) of the evidence. O nce t his pot entia l m isuse is iden tified, the cou rt is requir ed to ga uge t he sign ifica nce of, an d t he likely ju ry rea ction to, such evidence, but only for the purpos e of performing the una voidable task of bala ncing the va lue of t he evidence in supporting the permissible inference against the

grounds for e xclu din g relevan t e vide nce if th e ba lan ce so d icta tes . Id. In pra ctice, “unfair preju dice” is under stood to in clude a ll three grounds for exclusion, see 22 WRIGHT & GRAH AM, supra note 53, § 5215, at 273–74, and tha t sam e parlan ce is used here. 230. For exa mp le, e vide nce of des ign cha nges m ad e t o an all ege dly d an ger ous product following an injury resu lting from that product is not adm issible to prove that the product wa s defective as originally designed, but that sa me evidence can be adm issible to p rove t he fea sib ilit y of a sa fer de sig n i f th e m an ufa cturer claims that the sup erior design was not possible . S ee F ED. R. EVID. 407. In s uch an inst an ce, a single item of evidence could be employed for two distinct purposes. He nce, the value and significance of the evidence for the latter, permissible purpose must be balan ced against the da nger of the evidence for the former, impermissible purpose. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1605 danger of it s u se in su ggest in g t he entir ely dist in ct impermissible inference. By contrast, if the only claimed potential for pr eju dice is the danger that such evidence will be overvalued, there is but one use to which the evidence could be put, and the “balancing” suggested by the langua ge of Rule 403231 is no more possible than is the use of a seesaw by a single child. Unlike th e preceding, two-pronged illustration representative of the true Rule 403 ba lan cing pr ocess, the over va lu ation hypothesis can be illust rated by a single-pr onged sch em e, a s follows :

EVIDENCE

Process of permissible reasoning from evidence

Per missible in fer en ce from evidence (PROBATIVE VALUE)

Ove rva lu ed in fer en ce from evidence (UNF AIR PREJU DICE)

It is ext rem ely qu estion able that the a uthorit y t o exclude relevant evidence because its probative value is su bstan tia lly outweighed by a dan ger of unfair prejudice (the two-pronged model) should in clu de the power to exclude rele va nt evidence solely because that evidence might be overvalued (the single- pronged sch em e). As noted above , t he process of exclusion in the single-pr onged situation does n ot in volve any “ba la ncing”

231. See als o F ED. R. EVID. 403 ad visor y committee’s notes (“Situa tions in this area call for ba lan cing th e pr oba tiv e va lue of an d n eed for t he evidence against the harm lik ely to r es ul t fr om it s a dm iss ion .”). D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1606 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 whatsoever, and it is only a balancing of distinct uses of evidence which is the a ut horized judicial fun ction under Rule 403. Some further clue on this point might be gath ered by looking to the definition of “unfair prejudice” to discover whether that concept includes the danger of over va lu ation . Unfortu nately, “unfair prejudice” is not defined in the Federal Rules of Evidence,232 its mean ing is ra rely discussed by the courts,233 and consequently its exact mea ning is unclear.234 When “unfair prejudice” is discussed, the discussion almost invaria bly seems to envision t he t wo-pronged scena rio,235 alt hough this observat ion is by no means un iversal.236 In an y event , actu al judicial practice “rarely, if ever, exclud[es] evidence of subst an tia l probative value simply because the jury appears likely to give the evidence even more weight than it deserves or because the precise weight to be given is unclear.”237 This is su rely t he corr ect pr actice, for it preserves the jur y’s prerogative t o assign wh at ever weight it chooses to relevant eviden ce, and it curbs th e occasionally encroaching temptation of some tr ial judges to exclude evidence solely because they p er son ally are n ot persu aded by such evidence. Even if ther e existed a sou nd basis for believing th at character evidence is in reality less probative th an wha t is commonly believed, th e solution should not be the infringement of the jury’s aut onomy by the k eeping of such eviden ce from its consideration. The appropria te r emedy—the one selected in

232. See Gold, supra note 223, at 60, 73. 233. See 22 WRIGHT & GRAH AM, supra note 53, § 5215, at 277. 234. See Gold, supra note 223, at 65. 235. S ee, e.g., FED. R. EVID. 403 ad vis ory com mi tt ee ’s not es (“‘Unfair prejudice’ with in its context mean s an u ndue t endency to suggest decision on an improper basis, commonly, though not necessarily, an em otion al one.”); Old Chief v. United States, 519 U. S. 1 72, 180 (199 7) (improper basis); 22 WRIGHT & GRAH AM, supra note 150, § 5215, at 231 (illegitimate, as oppos ed to legitimat e, inference); 22 WRIGHT & GRAH AM, supra note 53, at 274–76 (per su as ion by il legi ti ma te me an s, s uch as inappropriate logic or un desir able e mot ion); Gold, supra note 223, at 67 (“Rule 403 can advance tr uth and fairn ess by excluding evidence that tends to induce the jury to th in k i llogica lly or e mp loy a n i mp rope r b ia s.”). 236. See 1 WIGMORE, supra note 34, § 10a, at 684 (“The primary aim of the pr eju dice ru le is t o preve nt jur y ‘misdecision,’” including “t h e jur y attr ibut [ing] greater evi de nt ia l va lu e t o th e e vid en ce t ha n i s w ar ra nt ed .” (footn ote om it te d)). 237. Richard O. Lem per t, Mod elin g R eleva nce, 75 MICH. L. REV. 1021, 1030 (197 7). D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1607 almost all compa rable circumstances—would simply be to provide the jurors with the information necessary to adjust their evalua tion of such character evidence. For exam ple, ther e now exists a body of literature calling into qu estion the r eliabilit y of eyewit ness identifica tion s, suggesting that these id en tifica tion s m igh t not be as a ccurate as commonly supposed.238 It would be ludicrous to suggest that, as a result of this literature, eye witness identifica tion testimony should be disallowed. At most, the sole reaction has been to permit expert test imony in appropriate cases concerning the foibles of eyewitness identifications.239 Even if it were supported by more t ha n m ere specula tion, th e over va lu ation hypothesis as t o character evidence would merit no greater accommodation than that allowed in the eyewitness iden tificat ion context. Ult ima tely, the overvaluation hypothesis confesses a fundamental mistrust of the jury in its assigned fun ction in th e criminal just ice system. Bu t if the jury can not adequat ely determ ine the proper weight to accord to eviden ce—especially evidence requiring no legal or technical expertise to comprehend and evaluate—then the problem we face is a much larger one t han can be add ressed by t he ch aracter evidence rule. In truth, the jur y system is too well established and too valuable to just ify t he con trary prem ise of t he ove rva lu ation hypothesis. And our faith in that jury system is not grounded on the presumption that twelve jurors will typically see the evidence just a s twelve la wyers would perceive it. To t he contrary, the fact that jurors may see the evidence in a ma nner dissimila r to the way lawyers would view it demonstrates an acceptable, even desira ble, infusion of the commu nit y’s perceptions a nd values int o the criminal justice system. Ult ima tely, th ere is neith er empirical data nor any sound reason to conclude that juries overvalue character evidence. The supposition that we as lawye rs n eve rthele ss know h ow all—or at least most —jurors perceive chara cter eviden ce, and

238. S ee, e.g., BRIAN L. CUTLER & STEVEN D. PENROD, MISTAKEN IDENTIFICATION: THE EYEWITNESS, PSYCHOLOGY, AND THE LAW (1995 ); ELIZABETH F. LOFTUS & JAMES M. DOYLE, EYEWITNESS TESTIMONY: CIVIL AND CRIM INAL (199 2). 239. S ee, e.g., MICHAEL H. GRAH AM, HANDBOOK OF FEDERAL EVIDENCE § 702.4 (1996 & Su pp . 19 98). D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1608 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 further that we as lawye rs h ave t he sin gu la rly cor rect view of such evidence, is not only baseless, it is also arrogant. Thus the search for a viable justification for the character evidence rule continues.

C. Da nger of Un ju st Pu nishm ent The theory of this rationale is that the jury, having learned of the prior miscondu ct of the defenda nt, will con vict the defendant of the charged crime as a means of punishing the defendant for his or her pr ior miscondu ct.240 Unlike t he danger of over va lu ation discu ss ed in the previous s ubsect ion ,241 the concern ar ticulat ed h ere un quest ionably qualifies a s “unfair prejudice” within the meaning of Rule 403.242 Even assuming that the jury is permitted to use evidence of the defendan t’s uncharged misconduct as evidence of t he defenda nt’s guilt of the charged crime, using tha t evidence for the purpose of punishing the defendant for such un charged misconduct is clearly a distinct an d impermissible use of that evidence.243 The fundamental problem that the unjust punishment hypothesis sh ares wit h the ove rva lu ation hypothesis is an absence of any sound basis for believing that the projected danger is a rea l on e. The fa ct that evidence of the defendant’s prior misconduct m ight influence jury verdicts proves nothing in th is regar d. Once it is accepted (as it should be) tha t simila r crimes eviden ce is relevant ,244 any correla tion of su ch evidence with convict ion s is n o more ca use for concern than the correla tion of convict ion s w it h scien tific evid en ce or confessions or any other form of relevant inculpatory evidence.245 And aside from this meaningless correlation, there is simply no empirica l basis for the speculative ass er tion that jurors w ill convict persons believed to be not guilty of the charged crimes in order

240. S ee supra note 182. Consequently, the concern underlying this rationale is of par ticula r, and poss ibly exclusive, in ter est r egar ding character evidence offered against the criminal defendant. 241. See supra notes 227–239 a nd a ccompan ying te xt. 242. See Imw ink elrie d, supra note 7, at 581. 243. The ar gume nt for p re jud ice in th is s ub sect ion thus clear ly fits with in t he two-pronged schem e discu ssed above. See supra note 230 and accompanying text. 244. See supra notes 185–209 a nd a ccompan ying te xt. 245. See Ad m iss ion of Cr im in al H ist ories , supra note 34, at 731. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1609 to impose sanctions for uncharged crimes.246 In fact, the most that can be said in support of the unjust punishment hypothesis is that it “has widely been pr esum ed” that jur ors will convict as a sanction for uncharged misconduct.247 In the absence of any documentation tha t jurors beh ave in accordance with the unjust punishment hypothesis, one would exp ect at least some sensible intuitive support for the presumption underlying the unjust punishment hypothesis. But in fact, th e idea th at every juror in a particular case (or at least a commanding majority of jurors s ufficient to coer ce acquiescence from a principled minority of the jury) would be willing to convict a defen da nt those jurors believe to be not guilty of the charged crime solely to punish for uncharged conduct is count erin tu itive. Th e notion th at it is unfair to punish someone for something other than the matter at issue is so straightforward and so fundamental that it is difficult to fathom th at an ent ire jury would agree to do just t ha t, even in the absence of the nomina l safeguard of the standard cautionary instruction to this effect. Moreover, whatever criticisms may be leveled at jurors, it is not generally suggested that an entire jury is so lacking in conscientiousness that it will violate its oath and fabricate its verdict in order to inflict an illegal and very serious pu nishm ent upon an individual. But perhaps the premise underlying th e unjust punishment hypothesis is th e vision of a more su btle influence. Perh aps it is not that t he jurors would willfully ma nipula te t heir verdict to exact extralegal retribution, but rather that the antagonism engendered by k nowledge of prior crim inal condu ct could effect an unconscious sanction. Here, of course, we have entered the world of pure speculation. By definition, each of us is incompetent to know what is in his or her own unconscious, so one wonders h ow we are to know what is in the unconscious of others (or even if such a thing as an unconscious exists). Moreover, this entirely conjectural variation is a further, and indeed aggravat ed, manifestation of the patronizing attitude un derlying the overvaluat ion hypothesis. Under t his theory, most if n ot all ju rors (w hich essentia lly amounts t o the

246. See Admission of Criminal Histories , supra note 34, a t 710 ; Kar p, supra note 57, at 27. 247. Duan e, supra note 152, at 110. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1610 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 genera l population) are consolidated into the single im age of a group of persons not only lacking the capacity to exercise fundamental fair ness but also the aptitude to appreciate what they are doing. Jurors, it is suggest ed, do not realize that they are actually punishing the defendant for uncharged misconduct; lawyers, however, by virtue of attending law school, myster iously know that this unjust punishment is really t aking place. But if jurors believe in stead th at th ey ar e simply using the uncharged misconduct eviden ce in eva lua tin g th e defendant’s guilt or in nocen ce of th e charged crime—a perfectly rational thought process—and th ere is absolutely no evidence to contradict th at self-assessm ent , why should we not accept this tacit assurance as true? Furthermore, we must be reasonably secure against the contrary possibility or we would never allow other crimes evidence to make it s w ay t o the ju ry a s oft en as w e do, such as through cross-examination of the defendant’s own character witnesses,248 through Rule 404(b) 249 and to impeach the defendant as a witness.250 If we really believe tha t t he jury is powerless to focus upon the cr im e a t issu e on ce it lea rns of t he defendant’s uncharged misconduct, th en the well-established range of opportunities for the prosecutor to deliver such apparen tly intoxicating information to the jury under existing law ma kes no sense wha tsoever. Be that as it may, t her e is one m ea su re which wou ld guard against the conjectural danger of unjust punishment without forfeiting an entire class of perfectly relevant evidence. Because the concern is that the jury might be tempted to punish the defendant for prior, uncharged crim es, a ny such in clin ation should only occur (if at all) if the jury believes the defendant has someh ow escaped sa nctions for these prior tr an sgressions.251 To the extent th at prior crimes evidence is limited to condu ct wh ich has been the subject of pr ior convict ion s, such danger ought to be complet ely or su bstan tia lly eliminated. The counter intu itive notion that jurors would be

248. See supra notes 13–19 a nd a ccompan ying te xt. 249. See supra notes 20–108 a nd a ccompan ying te xt. 250. See supra notes 118–137 a nd a ccompan ying te xt. 251. See Adm ission of Cr im in al H ist ories , supra not e 34 , at 734; see also Imwinkelried, supra note 165, at 290–91. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1611 irr esist ibly influenced to convict an innocent defendant to punish for uncharged crimes is all the more implausible in cases in which the jurors are well aware that the accused has already suffered the law’s sanctions for the prior misdeeds. The unjust pu nishmen t hypothesis thus m er it s s ome consideration because, at least in theory, it does isolate a legitim at e concer n. Th er e is , however , n o ba sis for believing that this concern is substantial in actua lity. Neverth eless, th at concern can be accommodated short of embracing the character evidence rule in its full scope, by overruling the rule as to relevant unchar ged condu ct which resulted in crim ina l convict ion s. Consequently, the search for a persuasive justificat ion for the chara cter eviden ce ru le continu es.

D. Fair N otice If a criminal defendant’s character may be attacked by proof of all sorts of misdeeds, real or imagined, then it might be unfair to saddle the accused with the task of preparing to defend against a lifetime of allegations.252 Especially because such allega tion s m ay be false, it could be im pos sible for the defendant to anticipate, a nd t o arrange t o rebut, such accusations.253 Such worries are le git im ate. I ndeed, one a uthor has suggested that, unlike recent concer ns a bou t over va lu ation and un just punishm ent , the problem of unfair notice actu ally accounts in part for the genesis of the character evidence rule.254 Of course, to the extent that t he objection is solely the absence of notice, the solu tion wou ld simply be a notice requirement rather than a broad exclusionary rule.255 Even with a notice requirement, defendants might complain that it is unduly burdensom e t o be requir ed to defen d a ga in st bot h charged and uncharged miscondu ct in cases in wh ich the defendant disput es both. Whether such a burden would be unfair or not, the bu rden is en tir ely relieved if the uncharged

252. See 2 LOUISELL & MUELLER, supra note 28, § 136, at 130–31; Stone, supra note 41, at 958. 253. See 1A WIGMORE, supra note 34, § 58.2, at 1212–13. 254. See Kar p, supra note 57, at 27. 255. For example, Rule 404(b) was amended in 1991 to a dd a p re tr ia l n oti ce requirement when t he prosecution inten ds to offer evidence against the a ccused under that provis ion. S ee F ED. R. EVID. 404(b) and advisory committee’s notes. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1612 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 misconduct has been the subject of a pr ior convict ion .256 First, the defenda nt ough t to be on notice a s t o his or her own pr ior convict ion s, and in any event , this informa tion is routinely provided in discovery.257 Second, by vir tue of t he fa ct that the defendant’s pr ior miscondu ct has been the subject of a pr ior convict ion , the defendant would be precluded from disputing that he or she committed the prior crime.258 Th us, lim it in g prior crimes evidence t o condu ct wh ich has r esult ed in pr ior convict ion s essentia lly elimin ates bot h the unsupported concern about the danger of unjust punishment as well as th e relat ively plausible concern about the danger of unfair notice.

E. Collateral Issu es The final ra tionale frequent ly offered in support of the character eviden ce ru le is t ha t a llowing t he part ies t o litigate whether or not uncharged misconduct actually occurr ed could be extr emely time consuming259 and could distract the jury from the relevant issue of whether or not the accused committed the charged crime.260 Th is consideration is the prim ary r ea son that impeachment of a wit ness by specific a cts of m iscondu ct not resulting in conviction is lim ited to cross-examin at ion, with proof of such misconduct by extrinsic evidence being disallowed.261 As with the unfair notice rationale discussed above, 262 th ere is some suggestion that this theory, rather than the currently popu la r theor ies abou t the danger s of over va lu ation and unjust punishment, explains the origin s of the character evidence rule.263 On ce a gain, however , if t he character evidence is limit ed to miscondu ct which resulted in an earlier conviction, the concern is eliminat ed.264 There can be

256. See 3A WIGMORE, supra note 124, § 980, at 828; Adm ission of Criminal Hi st ories , supra note 34, at 709–10, 729. 257. See FED. R. CRIM. P. 16(a )(1)(B). 258. S ee, e.g., Em ich M otor s Cor p. v. G en er al M otors Cor p., 340 U.S. 558, 568–69 (195 1); Un it ed St at es v. P ode ll, 5 72 F .2d 31, 35 (2 d Ci r. 197 8). 259. See Kuh ns, supra note 60, a t 777 ; Nan ce, supra note 166, at 11. 260. See Na nce, supra note 166, at 11. 261. See FED. R. EVID. 608 (b); 3 WEIN STE IN E T AL., supra note 14, ¶ 608[05], at 608–45. 262. See supra notes 252–258 a nd a ccompan ying te xt. 263. See Kar p, supra note 57, at 27. 264. See 3A W IGMORE, supra not e 12 4, § 98 0, a t 8 28; Adm ission of Criminal Hi st ories , supra note 34, at 709–10, 730. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1613 no dispute as to the fact that the defenda nt commit ted the prior crime,265 an d even in the rare instance in which there might be a disp ute a s t o the specific con duct t ha t support ed the prior convict ion , the record from the earlier gu ilt y plea or trial should provide an expedit ious resolut ion of the cont roversy. In all but the most unusu al cir cumst ances, proof of rele va nt pr ior criminal conduct for which the defendant has been convicted should be most econ omica l a nd n ot the least likely t o divert the jury’s focus from the pending charges. An exa min ation of the five r ation ales offer ed in su pp ort of the character evidence rule thu s reveals that these supposed justifica tion s ar e either err oneous, unsupported by an yth ing but speculative mistrust of the jury, or are at most plausibly legitim at e only to the extent of limit ing character evidence to conduct which has resulted in crimina l convictions. However, the case for the character evidence rule does not end here. Apparently unwilling to defend the rule solely on the grounds that it should be pr eser ved, some proponents of the ru le ha ve argued that it must be preser ved. Specifically, there have been var iou s contentions that the character evidence rule is constitutionally ma ndat ed. As will be shown, none of these constitutional theories is at all meritorious.

V. CONSTITU TIONAL CONSIDERATIONS The va riou s a ttem pt s t o cloa k the character evidence rule with a constitutional veil have included claims pr edicated upon due process, double jeopardy, the privilege against self- incrimination and Sixt h Amen dm en t notice requirements. Each of these will be addr essed in tu rn .

A. Due Process There is some judicial support for the proposition that due process has some application to character evidence, but th e analysis is spur ious. As a th eoret ical possibility, any evidentia ry ru ling by a tr ial judge (in clu ding r ulin gs on the admissibility of character evidence) could be so funda men ta lly unfair that due process could be implicated.266 But the United

265. See supra note 258 and accompanying text. 266. S ee, e.g., Tu cke r v . Ma kow sk i, 883 F .2d 877 , 88 1 (10 th Cir . 19 89). D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1614 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998

States Supreme Cou rt has n eve r su ggest ed that the a dm ission of prior crimes evidence to establish the defendant’s general propensity to engage in criminal conduct of the type charged would violate due process.267 In fact, the Court has held that the adm ission of prior crimes evidence under a state counterpart to Rule 404(b) does not offend due process.268 If evidentiary rules permit the introduction of the prior convictions of the accused for som e specified p urpos e, d ue process is not a license for judicial second -guessing of such rules.269 Moreover, the possibilit y th at th e jury will disr ega rd lim it in g in st ruction s and use prior crimes evidence to unjustly punish the defendant for uncharged miscondu ct does n ot pr esen t a const it ution al issue.270 Notwithstanding these admonitions, the United States Cou rt of Appeals for the Ninth Circuit has seen fit to hold that the introduction of character evidence in a st ate prosecution can violate t he defenda nt’s federal “cons tit ution al r igh t to a fundamentally fair trial as guaranteed by the Due Process Clause of the Fourteenth Amendment .”271 In McKinney v. Rees, 272 a case wh ich en ter ed the federal cou rt syst em on a petition for a writ of habeas corpus following a state prosecution which resulted in t he conviction of the petitioner for the m urder of his m other by s lit ting her throat wit h a kn ife of un specified dimensions, the state h ad introdu ced evidence that the petitioner had possessed knives tha t could ha ve inflict ed the wounds in qu estion ; that the pet it ion er wa s p roud of his k nife collect ion; that he h ad occasionally stra pped a knife to his body while wea rin g camouflage pant s (th e sa me type of pants worn by the murderer); and that he had used a kn ife to scratch the words “Death is His” on a closet door.273 The state had argued successfully before the state trial and appellate courts that the evidence was admissible under the state counterpart to Rule 404(b) to refute th e pet itioner’s claim that he was not in possession of a kn ife at the time of the murder

267. See Es te lle v. M cGu ir e, 5 02 U .S. 62, 75 n .5 (1 991 ). 268. See Lis en ba v. Cal ifor ni a, 314 U. S. 2 19, 227 –28 (194 1). 269. See Ma rs ha ll v. Lon ber ger , 45 9 U .S. 422 , 43 8 n .6 (1 983 ). 270. See Sp en cer v. T exa s, 3 85 U .S. 554 , 56 2–6 4 (19 67). 271. McKinney v. Ree s, 9 93 F .2d 137 8, 1 380 (9t h C ir . 19 93). 272. Id. 273. See id. at 1381–82. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1615 and to show that th e pet itioner h ad an opportun ity to commit the crime.274 Lacking any authority to disturb the state courts’ conclu sions that the evidence in question was not general propensity character evidence but rather fit within a so-called noncharacter theory of admissibility under the state rules of evidence, the McKinney court did essentially the sa me thing by recasting the id en tica l is su e a s on e of federal con st it ution al significance. The court concluded that much of this evidence was “offered to prove character, not eviden ce of opportu nit y,”275 and was “thus, irrelevant .” 276 Wor kin g ba ckwa rds from the premise that character evidence is inadmissible to the conclu sion that such evidence is th erefore lega lly irr elevant , th e court concluded th at “McKinn ey’s tr ial wa s impermissibly ta int ed by ir relevant eviden ce.”277 The decision can be viewed in one of two ways. First, to the extent that the court simply concluded that the evidence in question was inadmissible char acter eviden ce, the court blatantly acted wit hout jurisdiction a s a superappellate court on a qu est ion of pur ely st at e evidence la w. Second, to the extent that the court was really intending to constitut ionalize the character evidence rule, it did precisely what the United Sta tes Supreme Cou rt had directed it not to do.278 It so happens that the st ate in qu estion had in place the character evidence rule as a matter of state evidence law. But if a state evidence code were instead to allow character evidence to establish general propensity, and if the N in th Cir cuit is truly s er iou s a bou t the McKinney decision, then presumably such rules of evidence would be unconstitut ional on their face. That conclusion would be wholly fanta stic, as is t he McKinney decision itself. If it is truly the United States Constitution, and not just evidentiary ru les, which renders character evidence le ga lly ir rele va nt, a nd if du e process ca nnot

274. See id. at 1382. 275. Id. at 1383. 276. Id. at 1384. The evidence in question was cer ta inly not factua lly irre levan t. In a ca se in wh ich th e m ur de re r w iel de d a kn ife, evidence that the petitioner was with in tha t portion of the population which, even occasionally, carried and used a knife, cer ta in ly h ad a t en den cy t o ma ke th e id en ti ficat ion of th e petitioner as the mur derer mor e pr oba ble t ha n it wou ld h ave bee n w ith out such evidence. S ee F ED. R. EVID. 401. 277. McKinney, 993 F.2d at 1385. 278. See supra notes 271–278 a nd a ccompan ying te xt. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1616 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 tolerate a prosecution contain ing such legally ir relevant evidence, then even the F eder al Ru les of Evidence wou ld en joy no exemption from this constitutional mandate. Presu ma bly, then, Rules 413 and 414279 are unconstitu tional. Bu t a lso Rule 609, wh ich allows p rior convictions a s eviden ce of the defendant’s character for truthfulness as a witness,280 must violate due process. Th e same would be t rue for Ru les 608 (a)281 and 608 (b). 282 To the extent that Rule 404(a) allows the government to intr oduce chara cter eviden ce concerning t he defendant once t he defenda nt in trodu ces character evidence concerning him self or herself,283 surely it violates due process, at least in th e Ninth Cir cuit . If McKinney is to be taken ser iously, then every decision of every cour t a tt emptin g to apply the often ba rely perceptible (or, depen ding u pon the particular case and the reader’s par ticular point of view, illusory) dist inction between inadmissible char acter evidence and evidence which is admissible un der one of the various theories listed in Rule 404(b)284 or an equivalent state ru le is actually resolving a constitutional issue. Can something as fundamental as due pr ocess really tur n on whether th e evidence is more likely to dem onst ra te a n inten t t o deal dru gs or a propensity to deal drugs? It is hardly surprising that the Supreme Court warned the Ninth Circuit not to open that Pandora’s box,285 and it should not be sur prising when, given a futur e opportu nit y, the Court chastises the Ninth Circuit for trespa ss in g on to the states’ interpretation and application of their own rules of evidence. The simple truth is that there is no reason why due process should require th e exclusion of such factu ally r elevant 286 evidence. The bas ic fact tha t t he chara cter eviden ce ru le, at

279. See supra notes 138–177 a nd a ccompan ying te xt. 280. See supra notes 122–137 a nd a ccompan ying te xt. 281. See supra notes 109–117 a nd a ccompan ying te xt. 282. See supra notes 118–121 a nd a ccompan ying te xt. 283. See supra notes 11–19 a nd a ccompan ying te xt. 284. See supra notes 20–108 a nd a ccompan ying te xt. 285. See supra notes 267–270 a nd a ccompan ying te xt. 286. See supra notes 185–209 a nd a ccompan ying te xt. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1617 least in theory, has been ar ound a long tim e287 does not alone establish it as a funda men tal r igh t of const it ution al propor tion . The usua l explicit or im plicit assumption underlying the case for due process protection is the familiar one that juries will convict to punish for uncharged misconduct,288 resulting in a fundamentally unfair trial. To be sure, if the unjust punishment premise were est ablished, the conclusion of fundamental un fairness would be beyond peradventure. There would be no need t o dr ag t he Const it ution in to the fr ay, for no policy consideration could justify such a result even in the absence of constitutional limita tions. But, as pr eviously discussed,289 ther e is no rea son ed or em pirica l supp ort for the premise of unjust punishment, especially when admitted evidence is limited to prior convictions. That being th e case, a due process ar gum ent based en tir ely upon the sam e faulty assumption adds nothing to the equation. Unless the actual consequ en ce of admitting relevant evidence would be to deny the accused a fair tr ial, ther e can ha rdly be a constit ution al entitlement to exclude such evidence. One imaginative—and truly imaginary—variation on the due process theme is the claim that character evidence violates due process because it is in consist en t wit h the presumpt ion of innocence.290 Relying u pon Leary v. United States,291 the proponents of this argument have reasoned that due process requires that evidence offered against the accused mu st render the fact to be established more likely to be true than not,292 that prior crim es evidence does not make it more likely than not that the defendan t has committed the instant offense,293 and

287. See Mar k A. Sh eft, Federal Ru le of E vid ence 4 13: A Dan gerou s N ew Fron tier , 33 AM. CRIM. L. REV. 57, 7 9–8 0 (19 95) (s ug ges ti ng th at th e h ist ori cal sturdiness of the cha ra cter e vidence r ule is on e factor e nt itling it to const itutiona l status un de r t he Du e P roces s C la us e). 288. S ee, e.g., She ft, supra note 287, at 79–81. 289. See supra notes 240–251 a nd a ccompan ying te xt. 290. See Louis M. N at ali, J r. & R. St ephe n St igall, “Are You Going to Arraign His Whole Life?”: How Sexual Propensity Evidence Violates the Due Process Clause, 28 LOY. U. CHI. L.J . 1, 32 (1996 ); see also Weisse nberger , supra note 28, at 608 n.87. 291. 395 U. S. 6 (196 9). 292. See Na ta li & Stigall, supra note 290, at 25. 293. See id. at 27–28. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1618 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 that therefore the admission of such evidence violates due process.294 The argument is specious on its face. The argumen t illegitimat ely su bstit utes a su fficiency of the evid en ce standard for the appropriate standard for admissibilit y of evidence. It suggests th at each item of evidence introduced by the government in a criminal prosecution must meet a “preponderance of the evidence” standard simply to cross the th reshold of admissibility. App arently, then, the government may not introduce any piece of evidence unless that evidence, considered in isola tion , wou ld convince the trier of fact of the ultimate issu e in a civil case. No matt er how overwhelming t he cumulative effect of the governmen t’s evidence, under th is requirement it is en tir ely pos sible that no single it em of evidence in the gover nmen t’s case wou ld eve n be allowed for the jury’s con sideration . The fatal misst ep in the argument is the confusion of the law governing the u se of presum ptions against the criminal defendant with the ordinary test of relevan ce for the a dmission of evidence. A p resumpt ion is not itself evidence but is rather “a procedural rule w hich requir es the exis ten ce of fa ct B (presumed fact) to be a ss umed wh en fact A (basic fact ) is established un less a nd un til a certain specified condit ion is fulfilled.”295 Ordinarily, a party with the burden of establishing fact B (presumed fact) may take advantage of a n available presumption by introducing eviden ce of fact A (basic fact ) and thereby sa tisfyin g it s burden. Because due pr ocess requires that the gover nmen t in a crim in al ca se pr ove, beyon d a rea sona ble doubt, every fact necessary to constitut e the charged crime,296 the question has arisen whether the use of a presumption against a criminal defendant unconstitutionally relieves th e governmen t of its bur den of proving the presu med fact. In Tot v. United States,297 the Court app roved the u se of presumptions against the accused, but only where there is a rational connect ion between the basic fact and the presumed

294. See id. 295. 1 WEIN STE IN E T AL., supra note 14, ¶ 300[01], at 300–1. 296. See In re Win sh ip, 397 U. S. 3 58, 364 (197 0). 297. 319 U. S. 4 63 (1 943 ). D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1619 fact.298 The Court further refined this requirement in Leary v. United States299 by specifying that the rational connection test means that the presu med fact must be more likely than not true a s a conclu sion derived fr om the basic fact.300 The ill-conceived notion tha t due process r equires evid en ce to meet the “more likely to be true than not” test for admissibilit y is premised on a misinterpretation of this Leary test.301 But character evidence is not a presumption, and it involves no use of presumptions. It is simply evidence. As such, its admissibilit y turns on t he ordin ar y ru les governin g th e admissibility of evidence and not some extraordinary constitutional prerequisite. One relat ed defense of the char acter evidence rule is Professor Reed’s claim th at th e ru le “had it s roots in t he accusative, as opposed t o inquisitoria l, n ature of t he An glo- Amer ica n criminal process,”302 and the charge that a contrary rule would “transform[] the American criminal justice system from an a ccusative to an in quisitorial process.”303 “Under an accusative system,” states Professor Reed, “t h e state must establish that the accused did some act forbidden by law.”304 An inqu isitorial system, on the other hand, “assum es the accused committed the crime and imposes upon him the burden of est ablishing h is innocence.”305 Professor Reed’s ana lysis ea rn s high mar ks for hyperbole but not for much else. First, the essentia l historical dist inction between the accusatorial and inquisitorial systems of procedure

298. See id. at 467–68. 299. 395 U. S. 6 (196 9). 300. S ee id. at 36. The constitutional limitations upon the use of presump tion s against the criminal defendant h ave undergone additional modifications since Leary. A mandatory (but r ebut ta ble) pres um ption, a s dist ingu ished from a perm issive inference, must pass th e rat ional connection test on the face of the presumption, and not merely as applied to the particular circumstances of th e case . See Ulster County Cou rt v. Allen, 442 U .S. 140, 157 –60 (1979). Moreover , a presu mption th at is eith er irr ebut ta ble or that shifts to th e defendant the ul ti ma te bu rd en of pe rs ua sion (a s distinguished from m erely s hiftin g th e bur den of pr odu cing som e evide nce cont ra ry to th e pr esum ed fa ct t o defea t t he effect of the pr esu mp tion ) is u ncon st itu tion al notwithstanding a ra tional connection between the ba sic fa ct and the presumed fact. See Sa nd st rom v . Mon ta na , 44 2 U .S. 510 , 52 3–2 4 (19 79). 301. See Na ta li & Stigall, supra note 290, at 25. 302. Reed, supra note 2, at 713. 303. Id. at 714. 304. Id. at 713. 305. Id. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1620 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 had nothing at all to do with character evidence or the rules governing the adm issibilit y of eviden ce in gen er al for that matter. The essential features of the inquisitorial system were the combination of the prosecutorial and adjudicative function s in a single individual or instit ut ion, with proceedings typically being writt en and secret ive.306 By contrast, the accusatorial system separated the accusatory and investigative functions from the adjudicative function, and proceedings were typically oral and public.307 Second, even under Professor Reed’s definitions of the two systems of criminal justice, it is difficult to comprehend the relationship of the character evidence rule to the posed juxt apos it ion . Th e per ceived dist in ction appears to be one which focuses on the a ss ign men t of the burden of persuasion, but the allowance of a certain t ype of evidence does not alter the obvious requirement that the government prove the defendant guilty beyon d a rea son able doubt . Appar ent ly the ar gum ent is that th e effective consequence of allowing character evidence is to place the defendant at such a disadvant age a s to pra ctically assign to him or her the burden of proving his or her innocence. In th is respect, Pr ofessor Reed illustra tes th at th e climb is always easier if one starts at the top, for he has assumed precisely what is at issue with regards to the character evidence rule. The fact that the government is assigned the burden of persuasion does not mean that the government is forbidden from meeting t hat burden. Whenever th e government introduces substan tia l evidence in satisfaction of its bu rden, th e defendant might, as a practical mat ter , find himself or her self hard pressed to introduce some exculpatory evidence. That does not mea n t ha t t he days of the Inqu isition ar e once aga in upon us. The fact th at the case put on by the government includes character evidence involve s n o syst em ic per mutation from the typical case in which the government attempts to meet, and often su cceeds in meetin g, it s burden. Th e r ea l qu estion —the only question—is wh et her the proba tive va lu e of such evidence is outweighed by an y unfair prejudice, as discussed above.308

306. See JOHN HEN RY MERR YMAN , THE CIVIL LAW TRADITION 132 –38 (196 9). 307. See id. 308. See supra notes 178–265 a nd a ccompan ying te xt. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1621

There is no us efu l shortcut for r esolvin g the is su e of the character evidence rule.

B. Double Jeopardy A second constitutional provision occasionally suggested as a pos sible lin chpin for the character evidence rule is the proscript ion aga inst double jeopa rdy.309 The superficial appeal of the argumen t is that an accused who is convicted because the jury seeks to punish him or her for uncharged misconduct which had been the subject of a prior prosecution has been “subject for the same offen se to be t wice pu t in jeopardy of life or limb.”310 There are, however, two flaws in this argumen t. First, once again the constitutional argument is entirely dependent upon the n onconstitu tional prem ise tha t juries will unjustly punish the accused for uncharged misconduct. If that premise were cor rect , the defense of the character evidence rule would be com plet e wit h no ass istance r equir ed from the Con st it ution . But in fact the unjust punishm ent hypothesis fails for want of any nonspeculative basis,311 an d consequ ent ly the double jeopa rdy analysis does n ot advance t he ca use of th e character evidence rule. Second, like th e due process argument, the double jeopardy claim, if it sa ys anyt hin g at all, says far too much. If it is tr uly the Constitution, and not just lawmak ers’ prerogative, th at dictates th e cha ra cter evidence rule, th en a ll of the well- establish ed rules allowin g for character evidence u nder va riou s circumstances312 would also be constitutionally invalidated.

C. Priv ilege Again st S elf-In crim in at ion The claimed connection between the character evidence rule and the privilege against self-incrimination is that prior crimes evidence may compel the defenda nt to testify to refute the uncharged allega tion s, wh ich may em ph asize h is or her failure

309. S ee, e.g., 23 WRIGHT & GRAHAM, supra note 150, § 5412, at 275; Brauser, supra note 46, at 1583. 310. U.S. CONST. am en d. V. 311. See supra notes 240–251 a nd a ccompan ying te xt. 312. See supra notes 280–284 a nd a ccompan ying te xt. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1622 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 to testify as to the char ged offense.313 The argum ent is flawed in several particulars. First, the argument is certainly not a viable defense of the historical character evidence r ule, because whatever force it has is equally app lica ble to pr ior crim es evidence a dm it ted, for example, under a Rule 404(b)-type theory. Second, a defendant may feel strategically compelled to test ify be cause of th e strength of the government’s evidence, b ut this is not a compu lsion that is of an y constitut ional significance.314 The fact that the evidence, the use of which influences the defendant’s voluntary tactica l decis ion to testify, is character evidence is of no sp ecia l im por tance.315 Finally, to the extent that the uncharged crimes resulted in a prior conviction, the argum ent is inapposite because the defendant would be estopped by the judgment of conviction from denying the prior crimina l conduct.316

D. Notice The last constitut ional window dressing th at has sometimes adorned the character evidence rule is the argument that a contrary rule would violate the requirem ent that the accused “be informed of the nature and cause of the accusation.”317 The short answ er is that ther e is no inherent relationsh ip between character evidence and inadequate notice. To the extent that notice migh t be requir ed, a nd even to the ext en t that notice would be desirable, it is a very easy matter to impose that obligation upon the government.318 The constitutional and related arguments for the character evidence ru le add n othing of an y significa nce t o the fa te of t he character evidence ru le. The rule must stand or fall on its own, and, as discussed above,319 so viewed there is little to commend it, at least in its present form. What follows is a proposal for revamping the rules of evidence governing character evidence in accordance with what has been discussed thus far.

313. See Bra use r, supra note 46, at 1583 & n.1. 314. See Will ia ms v. F lor ida , 39 9 U .S. 78, 83– 84 (1 970 ). 315. See Ad m iss ion of Cr im in al H ist ories , supra note 34, at 749–50. 316. See supra note 258 and accompanying text. 317. U.S. CONST. am en d. VI; see Reed, supra note 34, at 167. 318. See supra notes 252–258 a nd a ccompan ying te xt. 319. See supra notes 178–265 a nd a ccompan ying te xt. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1623

VI. P RO PO SAL The prima ry pur pose of the ru les of evidence is t o allow the trier of fact to discover the truth concerning the matters at issue.320 Consequently, all relevant evidence should be allowed absent some plain and forceful reason not to do so.321 The Federal Ru les of Evidence represent the modern trend toward “relaxation of the bar riers to admitt ing r elevant eviden ce.”322 The character evidence rule is entir ely at varian ce with th e modern view of facilitating the search for the truth by admitting all relevant evidence whenever possible.323 To the extent that, a s s omet im es happ en s by t he process of objection s, jurors becom e a wa re t hat su ch evidence is being kept from them, they quite na tu ra lly become frustra ted at th e apparen tly inexplicable int erfer ence wit h t heir performa nce of their assigned task.324 And to the extent that such evidence is never mentioned in t he presen ce of jurors, t hey might reasonably but err oneously conclude that the accused has engaged in no prior criminal conduct, quit e rationally concluding that such evidence is so paten tly relevant that the prosecutor would have presented it if it in fact existed.325 We have discovered that, while ther e is still widespread support for the character evidence rule, the reasons advanced

320. See Leona rd, supra note 171, a t 341 ; Uviller , supra note 12, at 845. 321. See Imw ink elrie d, supra note 183, a t 1 485 (“The admission of all relevant evid en ce ordinarily promotes th e public interest in discovering the truth in litigation, and that interest is weightier than an y of the com pet ing in ter est s list ed in ru le 403.” (footn ote om it te d)). 322. Id. at 1478. 323. The rule excluding u nch ar ged mi scon du ct i s con tr ar y t o th e t re nd in evid en ce law toward free proof. For centuries, the m ovement ha s been toward a bolit ion of th ose e xclu sion ar y rule s t ha t h ave as th eir ba sis th e da nger of misleading the fact -find er . J ur ist s a nd sch olars alike increasingly have agreed with Bentham th at te chn ica l r ul es of evi de nce de sig ne d t o pr eve nt fact -fin de rs from making mi st ak es a re , at bes t, m ore tr oub le t ha n t he y ar e wor th . David P. Br yden & Roger C. Pa rk , “Other Crimes” Evid ence i n S ex Of fen se Ca ses, 78 MINN. L. REV. 529, 561 (199 4) (foot note s om it te d). 324. See Teit elbau m et al., supra note 60, at 1196. 325. See Karp, supra not e 57 , a t 5 2. T hi s r esul t i s a rg ua bly a sig ni fican t dis tor tion of th e t ru th . If t he re is a ny conclu sion on t his point that can be dra wn from the data collected and analyzed by Professors Kalven and Zeisel, it is that the ab sence of a criminal record is a significant factor in explaining why the jury acquits in cases in wh ich th e judge would ha ve convicte d. See supra note 216. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1624 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 for its con tin uin g existen ce do not wit hst and s er iou s examination an d ana lysis. Moreover, while there continues to be a zealous commitm ent to the r ule a s a th eoret ical model, the rule in pr actice is oft en honored in the brea ch. Th e r ules of evidence explicitly per mit the introduction of character evidence in several circumstances. Furthermore, the behavior of the courts in liber ally allowin g ch aracter evidence t o rea ch the juries under a variety of exceptions and guises confesses a recognition of the true value of such evidence. What is needed is a complete reconstruction of the rules governing character evidence. The ru les should recognize the true worth of such evidence and should m ore genuinely describe what we actually do with it. Tinkering at the margins of these rules will not be sufficient , an d addin g more unprincipled, ad hoc “exceptions” such as Rules 413-415 simply contributes to the problem. As a starting point, Federal Rules of Evidence 404, 405, 413, 414, 415, 608 and 609 sh ould be a bolished in their entirety.326 Th ese should be repla ced wit h the followin g fou r relatively straightforward rules. PROPOSED RULE ONE: Evidence of specific crimes, wron gs

or a cts t ha t h ave resu lted in crim ina l convict ions is ad m issible for any pu rpose for which such evidence is relevant, su bject t o Rule 403. Proposed Rule One is the crux of the matter. Several features of this proposed rule deserve comment. First, the rule governs only p rior miscondu ct wh ich has r esult ed in convict ion of a crime. Th is limita tion prevent s potent ially inefficient and distracting commitments of time to collateral matters.327 It also substantially redu ces wha tever un likely t em pt ation t he jur y migh t have t o convict solely to sa nction the defendant for what the ju ry m ay per ceive t o be unpu nished, prior miscondu ct.328 Second, Proposed Rule One is not a rule of automatic admissibility for pr ior convict ion s, for sever al r ea son s. Th e fa ct

326. The only ru le govern ing cha ra cter e vidence left un distu rbed in this proposal is Rule 4 12, th e ra pe sh ield la w. S ee F ED. R. EVID. 412. To the extent t ha t t his r ule is base d on the du al policies of rem oving th e tr ibula tions of testifying a s a victim of a sex offense a nd en coura ging th e rep ortin g of such crim es by victims, the ru le is beyon d the scope of any of the issues addressed in this article. 327. See supra notes 259–265 a nd a ccompan ying te xt. 328. See supra note 251 and accompanying text. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1625 of a prior conviction is a prerequisite for admissibility under Proposed Ru le O ne, b ut it is the con du ct underlyin g t he prior convict ion that will det er min e whet her the evid en ce of t he prior crime is relevant . To the extent that the evidence is offered to demonstrate a gr ea ter lik elihood t hat the a ctor behaved on the occasion in question consist en tly wit h his or her pr ior behavior , the pr oba tive va lu e of such evid en ce logica lly correla tes to the degree of similarity between the prior misconduct and the pending allegations.329 To the extent that the evidence is offered to attack the cr edibilit y of a wit ness, the proba tive va lu e of such evidence logica lly correla tes to the degr ee to wh ich the prior crime evidences a dishonest character. Thus for example, in a perjury prosecution , t he defenda nt’s prior dr ug conviction should ordinar ily be imm at eria l. And, in a drug prosecution in which the defendant does not take the stand, the defendant’s prior perjury conviction ordina rily sh ould be completely irr elevant . Moreover, in circumstances in wh ich the con du ct underlying th e prior conviction has only min imal probat ive value, the trial court migh t well exclu de the evid en ce because of the presence of any of the exclusionar y criteria specified in Rule 403.330 Third, no automatic time limit has been placed upon t he age of th e prior conviction in Pr oposed Rule One. Under the current law, there is a presumptive age limit for prior convictions used to impeach the credibility of a witness.331 Proposed Rule One, however, is not limited to the use of prior convict ion s for impeachment only. On e ca n cer tain ly envis ion conduct resulting in a conviction long ago which bears such a striking similar ity to the charged crim e so as to be extr emely relevant despite the intervening passage of tim e. Consequent ly, the tim e between the prior misconduct and t he cu rren t allega tion s should be trea ted as on e of t he fa ctors for the trial judge’s consideration under Rule 403. Fourth, if evidence is admitted against the accused under this provision, th e defendant would still be ent itled to a limiting

329. S ee, e.g., Uviller , supra note 12, at 882–85. 330. The danger of unfair pr ejudice would not include the possibility that the jury would draw the general propensity inference from the prior cr iminal misconduct. To the exten t th at t he evidence logically suggests such a general propensity, th is is precise ly what makes the evidence relevant. 331. See FED. R. EVID. 609 (b). D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1626 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 instr uction cautioning the jury not to convict the defendan t in order to punish h im or her for the prior misconduct.332 But this instr uction does not carry with it the sign ifica nt da nger s of misunderstanding or disobedien ce t hat accompa nies sever al of th e instr uctions tha t a re curr ent ly employed under the existing character evidence rules.333 Jurors ordinarily are con scien tiou s in attemptin g to comply with direct ion s r eceived fr om the trial judge.334 It is unreasonable, however , t o exp ect jurors t o compr ehend instructions which ar e extrem ely complex and appea r to be delivered solely for the legal pretense that the instructions were technically correct .335 Further , it is unrea son able to exp ect jur ors to comply with instructions that are contrary to what they know to be true.336 One reason why the current character evidence instructions hurled at jurors a re so hopelessly in effectual is tha t jurors kn ow from their own common sense and exp er ien ce that tru ly similar crimes evidence is relevant precisely to demonstrate a gener al cha ra cter tr ait , which in turn is rele va nt to the likelihood of r epea ted simila r condu ct.337 By contrast, there is no popular commitment to the notion that people should be pu nished for crim es for wh ich they are not on trial. Quite to the contrary, the cautionary instruction accompanying evidence admitted under Proposed Rule One is not only easy to underst and, it is also perfectly consistent with common, intuitive notions of fundamental fairness. Even to the unsubstantiated extent that a juror might be unconsciously tempted to punish for prior crimes, an instruction that alerts the juror to guard against this possibility is ver y likely t o have it s inten ded effect .338 PROPOSED RULE TWO: Evidence of specific crimes, wrongs

or act s w hich hav e n ot r esulte d in a cr im inal conv iction , an d not including conduct which is the subject of a ch ar ge, cla im or defense which is the subject of the proceeding, shall not be ad m issible unless the court d eter min es th at the probative

332. See Imw ink elrie d, supra note 1, at 1144–46. 333. See supra notes 13–18, 94–10 8, 112, 132–135 a nd a ccompan ying te xt. 334. See Imw ink elrie d, supra note 94, at 379–80. 335. See Kra mer & Koenig, supra note 102, at 433. 336. See id. at 429–30. 337. See Davies, supra note 107, a t 511 ; Imwin kelr ied, supra note 94, at 379. 338. See Teit elbau m et al., supra note 60, at 1196. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1627

value of such evidence outweighs any danger of unfair prejudice, confusion of the issues, misleading the jury, undue delay, waste of time or needless pr esen tation of cumu lative evidence. Proposed Ru le T wo would govern all s pecific instances of conduct not resulting in a criminal conviction offered as character evidence. It is envisioned that this rule would also, therefore, cover evidence h avin g special relevance because of a striking simila rit y t o the con duct at issue (such as some of the evidence curren tly a llowed under Rule 404(b) under the mislabel of noncharacter evidence), e ven though such conduct did not result in a crimina l conviction. Because this evidence la cks th e sa fegua rds and virtues of evidence of con du ct wh ich has resulted in criminal convictions,339 Proposed Rule Two contains a pr esumpt ion of exclusion, a s d istin gu ished from the presumption of admissibility in Proposed Rule One. The burden is placed upon t he proponen t of such evidence to sa tisfy the trial court that the probative value of the ch aracter evidence outweighs any dangers of unjust punishment or undue attention to collateral matters. PROPOSED RULE THRE E: Evidence of specific crimes,

wrongs or a cts of the accu sed offered by t he pr osecu tion in a criminal cas e, ot her th an th e cr im es or a cts wh ich are su bject of th e charges in the cas e, sh all not b e a dm issible unless, u pon request by the accused, the prosecution shall provide re as ona ble notice in advance of trial, or during trial if t h e cour t excuses pretrial notice on good cause shown, of the general nature of any such evidence it intends t o introduce at trial. The la ngu age of th is Proposed Rule Thr ee is taken ver bat im from th e 1991 amen dment to Rule 404(b) adding a pretrial notice requirem ent to that rule. Proposed Rule Three completely accommodates any concern, constitutional or otherwise, about providing the crimina l defendant with a fair opportunity to prepare a defense. Moreover, it provides the defendant with sufficient information to move in limine to exclude evidence of prior misconduct, thus avoiding the danger of unfair prejudice that can occur when objections are made and

339. See supra notes 327–328 a nd a ccompan ying te xt. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1628 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 sustained only after the jury ha s heard the impermissible, yet neve rthele ss pot en tia lly da magin g, question . A case can be made that such a pretrial notice requirement should apply t o all evid en ce of u ncharged miscondu ct and n ot just that of the criminal defendant. This would include, then, evidence of prior misconduct by th e victim of the crime, by the parties in a civil action, and by witnesses (which might be introduced to attack credibilit y). I have n ot in clu ded s uch a br oad notice requirement in Proposed Rule Three because the special pr otect ion s t o be afforded t o the cr iminal defendant are manifest, and beca use in other context s a pr et ria l n otice requirement might be too burdensome. For example, a party might not know t he id en tit y of all of t he wit nesses t o be called by the opponent, and therefore the circumstances might be too fluid reasonably to require a party t o sp ecify, in advance of trial, th e cha ra cter evidence it exp ect s t o use to attack the credibility of the opposition witnesses. PROPOSED RULE FOU R: Evidence of the character of a

person is not admissible by testimony as to repu tation or by te st im ony in th e for m of an opin ion, except t ha t t he accu sed in a crim in al pr ose cution m ay in tr odu ce such evide nce concerning his or her own character. The gen er al ba n on character evidence in the for m of opinion or reputation testim ony is just ified beca use su ch evidence is the weakest, least reliable and relatively most misleading form that character evidence can take.340 The exce pt ion for t he cr im inal defendant’s proof of his or her own good character is warranted for two reasons. First, the consequences of a criminal prosecution warrant special precautions against a n er roneous conviction. Second, it is especially difficult to demonstrate good character solely by introducing sp ecific in st ances of good conduct. A defendant accused of serial killings does not advance his or her defense sign ifica ntly by calling individuals to testify only that when each of them encountered the defendant, the defenda nt did n ot kill each of them.

340. See CLEARY ET AL ., supra note 15, § 186, at 550; Leonard, supra note 12, at 16; Uviller , supra note 12, at 885. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1547] CHARACTER EVIDENCE REVISITED 1629

There is n o pr ovis ion in Propos ed Ru le F our for the cr oss- examination of character witnesses as to specific instances of bad conduct of the person about whom th e chara cter witn ess is testifying. The omission is by design. If specific instances of misconduct qua lify for admissibilit y un der eith er P roposed Rule One or Proposed Ru le T wo, t hen ther e is no rea son not to allow th eir use in cross-examination of a character witness called by the accused pursuant to Proposed Rule Four. However, if the alleged sp ecific acts of m iscondu ct by t he defenda nt do n ot qu alify for admissibility u nder eit her Propos ed Ru le O ne or Proposed Ru le T wo, the government should not be permitted to present the sa me in formation to the ju ry u nder the pret ext of testing the weight to be given to the testimony of the positive character witness. There is also no pr ovision for perm itt ing t he government to call a character wit ness t o offer a nega tive opinion or reputation concerning the defendant’s character once the defendant ha s pu t on a positive cha ra cter witn ess a s ther e is under the current law.341 Fina lly, nothin g in Proposed Rule F our restr icts t he a bility of the defendant (or any other party) to introduce opinion or reputation testimony for a purpose other than to establish the character of a person. Thu s, for example, a defendant charged with a crime of violence who claim s t o ha ve a cted in self- defense wou ld be fr ee t o pu t on e viden ce t hat he or sh e (t he defendant) was aware of th e r eputation for violence of the victim of the alleged crime, because su ch eviden ce would not be offered an d received t o prove t he chara cter of the victim, but rather to prove the r easonableness of the defendant’s beh avior under the circumstances as he or sh e reasonably perceived them to be.

VII. CONCLUSION The character evidence rule cannot be just ified by any sound policy. It exists because of an entirely un justified mistrust of th e jury. The r ule a s w rit ten allows for many complicated exceptions tha t a re in consistent with th e supposed reasons for the ru le itself, and the r ule in practice is observed

341. See supra note 11 and accompanying text. D:\ 1998-4\ FINAL\ MEL-FIN.WPD Ja n. 8, 2001

1630 BRIGHAM YOUNG UNIVERSITY LAW REVIEW [1998 sporadically at best. The presen t approach to the cha racter evidence pr oblem is ent irely u nfa ir to ju rors, as it requires them to master a nd comply wit h obscure and count erint uit ive rules. A superior, streamlined and more honest approach can be obtained by substitu tin g four simple, compr ehensible, a nd workable rules for the current morass.