July 1999

RICO Report
By Barry Tarlow

    Barry Tarlow is a nationally prominent criminal defense lawyer practicing in Los Angeles, CA. He is a frequent author and lecturer on criminal law.
    He was formerly a prosecutor in the United States Attorney’s Office and is a member of The Champion Advisory Board.

    The author wishes to thank Blair Berk and Shereen Charlick, members of his firm, for their invaluable assistance in the preparation of this column.
Yes, Daubert’s ‘Gatekeeping’ Function Does Apply
To Soft Expert Testimony

In Kumho Tire Co. v. Carmichael, 119 S. Ct. 1167 (1999), the Supreme Court answered the question posed by Justice Rehnquist’s dissent in Daubert v. Merrell Dow Pharmaceuti-cals, 509 U.S. 579 (1993): does Daubert apply to nonscientific, or, to “soft-expert testimony” as well as scientific evidence? See, The Champion, RICO Report (June 1998) (asking: “Does Daubert apply to soft expert testimony?”). Answering affirmatively, in a fairly resounding decision, the Supreme Court went to the Daubert well one more time, and may have actually (unwittingly) drawn some water for criminal defense lawyers. Justice Breyer authored the near-unanimous decision, where the Court decided “how Daubert applies to the testimony of engineers and other experts who are not scientists.” Id. at 1171. Answering this question resolved a circuit split, where several circuits, notably the Fifth, Eighth and Third had indicated that Daubert applied to all expert testimony while the Second, Ninth, Tenth and Eleventh circuits held that Daubert applies only to the admission of “scientific” expert testimony. See, The Champion, RICO Report (June 1998) (comprehensive discussion of each circuit’s position re Daubert and soft expert testimony). The Court concluded that Daubert’s strictures apply not just to “scientific evidence,” but to all forms of proposed expert testimony. Kumho, 119 S. Ct. at 1171, 1175.

In Kumho, the plaintiffs, eight members of the Carmichael family sued a tire manufacturer and distributor (collectively Kumho Tire) after their car suffered a blowout and overturned. See Kumho, 119 S. Ct. 1171; Carmichael v. Samyang Tire Co., 131 F.3d 1433, 1434 (11th Cir. 1997), rev’d, sub. nom., on other grounds. All the family members suffered “significant trauma” and one family member died as a result of the accident. Id. After the accident, the Carmichael family gave the tire to an expert on tire failure. Id. This first “tire expert” examined the tire, and concluded that its failure was not the fault of the Carmichaels; rather, it was caused by a design or manufacturing defect in the tire. Carmichael, 131 F.3d at 1434. However, this tire expert became ill before his deposition, and his boss took over, and was deposed. The new tire expert only examined the Carmichaels’ tire for the first time for one hour immediately prior to the scheduled deposition and confirmed the original conclusions.

Some additional facts which likely made a difference, at least to the district court and the Supreme Court, were: (1) that the tire from the minivan was made in 1988 and installed on the minivan sometime before the Carmichaels bought it in 1993; (2) although the Carmichaels had only owned the minivan for two months before the accident, they had driven it about 7000 miles; (3) the tires were not in good shape as the tread was quite worn-down, in some places, “to nothing at all;” and, (4) the tire tread had several punctures which had been inadequately repaired. Kumho, 119 S. Ct. at 1172. In spite of the “tire’s age and history,” the great majority of which the tire expert did not know, he still concluded that a manufacturing or design defect caused the blowout. Id.
The technology underlying this conclusion was hotly disputed in the district court. After Kumho Tire Co. moved in limine to preclude the tire expert’s testimony, asserting that it was unreliable, the district court agreed, that under Daubert, it had a “gatekeeping” role to fill. Id. at 1173. Concluding that Daubert applied, notwithstanding the fact that the tire expert’s proposed testimony was “technical” and not “scientific,” it employed the four Daubert factors, found none of them were satisfied and excluded the proposed testimony as unreliable. Id; see also Carmichael v. Samyang Tires, Inc., 923 F. Supp. 1514, 1521-22 (S.D. Ala. 1996).

However, upon the Carmichael’s petition, the trial court ultimately reconsidered applying Daubert, concluding that it had been too “inflexible.” Kumho, 119S.Ct. at 1173. It considered the matter anew but still found that the “methodology employed . . . in analyzing the data obtained in the visual inspection and the scientific basis, if any, for such an analysis,” was not sufficiently reliable. Id.

On appeal, the Eleventh Circuit reversed, holding that the Daubert analysis only applies where an expert relies “on the application of scientific principles,” rather than “on skill-or experience-based observation.” Id. (quoting Carmichael, 131 F.3d at 1435-36). It remanded to the district court to determine, pursuant to Federal Rule of Evidence 702, whether the testimony was “sufficiently reliable,” “relevant to assist a jury,” and for determining under Federal Rule of Evidence 403 whether “its probative value is substantially outweighed by its likely prejudicial effect.” Carmichael, 131 F.3d at 1436. Interestingly enough, in its rush to distance Daubert from this soft-expert type testimony, the Eleventh Circuit failed to acknowledge that the district court, on reconsideration, had essentially already considered “reliability” in a context broader than just the factors contained in Daubert. See Kumho at 119S.Ct. at 1179; Carmichael, 131 F.3d 1433. We can only speculate that the reason for the Eleventh Circuit’s reaching out for this Daubert issue was to send a message that soft-expert testimony should not be subject to Daubert’s strictures.

Reversing the Eleventh Circuit, the Supreme Court first reiterated the significant points from Daubert, specifically, that the expert testimony must be both relevant and reliable to be admissible. Kumho 119S.Ct. at 1174. While noting that Daubert addressed what it called “scientific expert testimony,” as that was the nature of the proposed testimony in Daubert, “the evidentiary rationale underlying Daubert’s ‘gatekeeping’ determination” was not limited to “‘scientific knowledge.’” Id. at 1174. Stressing the importance of this gatekeeping function, the Court noted that its objective was “to ensure the reliability and relevancy of expert testimony,” “to make certain that an expert, whether basing testimony upon professional studies or personal experience, employs in the courtroom the same level of intellectual rigor that characterizes the practice of an expert in the relevant field.” Id. at 1176.

According to the Court, while Rules of Evidence 702 and 703 give all experts, not just scientific experts, the additional “testimonial latitude” unavailable to ordinary lay witnesses, this is granted on the condition that their opinion will have a reliable basis, grounded in the teachings and theories of his or her discipline. Kumho, Id. at 1174. In order to ensure this level of reliability, Daubert’s teachings and test must be applied to all expertise, whether it be scientific or “technical” or “other specialized” knowledge. Id. These requirements must exist because all experts, not just scientific ones, unlike ordinary witnesses are given this wide latitude to offer opinions — including those not based upon firsthand observation or knowledge. Id.

However, Daubert’s four-factor inquiry requires some modification to apply to many non-scientific, technical or “other specialized” areas of purported expertise. See Daubert, 509 U.S. at 592- 94 (setting forth four factors in making reliability determinations: testing, peer review, error rates and “acceptability” in the relevant scientific community). According to the Supreme Court, this is what Daubert meant when it said that the inquiry it envisioned is “a flexible one.” Kumho, 119S.Ct. at 1175 (quoting Daubert, 509 U.S. at 594). While, this “flexibility” in Daubert’s application affords trial courts “considerable leeway,” id. at 1176, Kumho admonished that “a trial court should consider the specific factors identified in Daubert where they are reasonable measures of the reliability of expert testimony.” Id. Daubert’s “particular questions . . . will often be appropriate for use in determining the reliability of challenged expert testimony.” Id. (emphasis added). On the other hand, while urging application of Daubert’s four factors, Kumho was quite clear that the Daubert factors do not constitute a definitive and exclusive checklist. Id; But see id. at 1179 (Scalia, J., concurring) (“[t]hough, as the Court makes clear today, the Daubert factors are not holy writ, in a particular case the failure to apply one or another of them may be unreasonable, and hence an abuse of discretion”).

Applying this analysis to the facts at hand, the Court discussed in some detail how and why the district court evaluated the tire expert’s methodology and analysis in ultimately rejecting it as unreliable. See Kumho, 119S.Ct. at 1177 (Stevens, J. dissenting in part) (determining whether trial judge abused discretion in excluding evidence beyond scope of certiorari petition and more appropriately left for Court of Appeals). It appears most significant to the Court that the trial judge found that none of the Daubert factors had been satisfied, that its own analysis revealed no countervailing factors favoring admissibility, and that the parties had identified no additional factors. Id. at 1178. The Supreme Court noted that no other experts in the industry used the same test as this tire expert, and there were neither articles or papers regarding his particular theory nor was there any testing of his theory.

It will be interesting to see whether and how Kumho/Daubert will be rigorously applied in the criminal arena, especially in today’s “soft expert” heyday. Of course, prosecutors, represented by the Solicitor General, as one of the largest users and abusers of soft expert testimony, were quick to file an amicus brief in Kumho, pointing out that “there are many different kinds of experts” and “the factors identified in Daubert may or may not be pertinent in assessing reliability, depending on the nature of the issue, the expert’s particular expertise, and the subject of his testimony.” Kumho, 119S.Ct. at 1175 (quoting Brief for United States as Amicus Curiae at 19). They wanted to ensure that their favorite new toys, the soft experts (usually government agents) who testify that drug traffickers never entrust money or drugs to people ignorant of the criminal activity, or that there is no such thing as a “blind mule,” and the like, can continue to do so. See, The Champion, RICO Report (June 1998) at 48-50 (discussing Daubert’s applicability to this type of “expert” testimony); see also The Champion, RICO Report (Nov. 1997) (discussing use and dangers of drug modus operandi “expert” testimony; see also United States v. Cordoba, 104 F.3d 225 (9th Cir. 1997) (“expert testimony” that drug traffickers do not entrust large quantities of drugs to unknowing dupes); United States v. Webb, 115 F.3d 711, 716 (9th Cir. 1997) (“expert testimony” that defendant must have known was gun in engine compartment because criminals often hide guns there).

Notwithstanding the “flexibility” of the new and improved Kumho/Daubert test and the recognized inapplicability of some factors with regard to some experts, see Kumho, 119S.Ct. at 1175, we can only hope that Kumho will, at least halt, if not put an end to some of the abuses involved in the use of “soft expert” testimony, particularly since it is usually impossible to verify empirically and in most cases, impossible to cross-examine effectively because there is a complete dearth of empirical data — the “expert” can say whatever he or she wishes. However, we should remember that Kumho still cautions that “[n]othing in either Daubert or the Federal Rules of Evidence requires a district court to admit opinion evidence that is connected to existing data only by the ipse dixit of the expert.” Id. at 1179 (quoting General Electric Co. v. Joiner, 522 U.S. 136, 146 (1997). We had previously concluded that where Daubert is applied, it is “harder to introduce but easier to challenge” this type of soft expert testimony. See, The Champion, RICO Report (June 1998).

In any event, even if Kumho cannot be used to exclude the prejudicial soft expert testimony, it can work to the advantage of defense lawyers as well as prosecutors. The advent of the Daubert era has opened some doors to present expert witnesses in many areas, for example, testimony regarding differing cultural customs, tax and banking practices, and as we have suggested, “informer-cooperation” experts, to testify about what the plea agreement means and its true impact for the informer in the real world. See, The Champion, RICO Reports (Jan./Feb. 1994) (discussing introduction of “novel” defense expert testimony), (March 1997) (cooperation agreement expert testimony), (May 1996). See, The Champion, RICO Report (Dec. 1995) (discussing favorable polygraph test results.)

Additionally, Kumho appears to encourage “special briefing” or “other proceedings” when needed to investigate the expert’s reliability. See Kumho,119S.Ct. at 1176. This could present an important opportunity to challenge the “soft expert” testimony by pre-trial pleadings, or even where the challenge is unsuccessful, one may obtain a pre-trial hearing, where the expert is forced to elucidate his or her methodology and experience, subject to cross-examination. Certainly, this can only assist in terms of obtaining one’s own expert to evaluate these conclusions or in preparing to investigate and cross-examine the witness at trial.

It cannot be overemphasized, however, that Kumho was decided in the civil context. History has demonstrated that what is good for the civil goose cooks the criminal gander. See, e.g., United States v. Castro, 89 F.3d 1443, 1452 (11th Cir. 1996); United States v. Viola, 35 F.3d 37, 42-3 (2d Cir. 1994); MCM Partners, Inc. v. Andres-Bartlett & Assoc., Inc., 62 F.3d 967, 979 (7th Cir. 1995) (examples of circuit courts refusing to apply Supreme Court’s “operation and management” from civil RICO case of Reves v. Ernst & Young, 507 U.S. 170, 178 (1993) to criminal context where would assist criminal defendants).

Of course, time and litigation will have to tell what effect Kumho/Daubert will have in criminal cases. Hopefully, it will further the search for the truth by eliminating some of the questionable “non-expert” soft-expert testimony that has been previously introduced in criminal trials.
Fifth Circuit Sets Standard For Determining Breach
of Non-Prosecution Agreement

Non-prosecution agreements, like plea bargains, are contractual in nature and are regularly interpreted consistent with general principles of contract law. United States v. Moulder, 141 F.3d 568, 571 (5th Cir. 1998); United States v. Ballis, 28 F.3d 1399, 1409 (5th Cir. 1994); United States v. Fitch, 964 F.2d 571, 574 (6th Cir. 1992); United States v. Brown, 801 F.2d 352, 354 (8th Cir. 1986). Traditional contract principles dictate that when a party lives up to its end of a bargain, the partycontracted with is likewise bound to perform its promises.

Similarly, in the context of a non-prosecution agreement, where a defendant “materially breaches” his commitments under the agreement, the government is traditionally released from its reciprocal obligations. Ballis, 28 F.3d at 1409; United States v. Crawford, 20 F.3d 933, 935 (8th Cir. 1994). Unlike general contract law, however, when the government believes that a defendant has breached the terms of a non-prosecution agreement and seeks to be relieved from the bargain of non-prosecution, due process restricts the government from unilaterally nullifying the non-prosecution agreement. United States v. Verrusio, 803 F.2d 885, 888 (7th Cir. 1986); United States v. Tarrant, 730 F. Supp. 30, 32 (N.D. Tex. 1990).

In order to rescind such an agreement, the government must prove to the court by a preponderance of the evidence that: (1) the defendant breached the agreement, and (2) the breach is sufficiently “material” to warrant recision. See, United States v. Price, 95 F.3d 364, 367 (5th Cir. 1996) (determination of whether government’s actions have breached terms of plea agreement requires demonstration of underlying facts that establish breach by preponderance of evidence); United States v. Witte, 25 F.3d 250, 263 (5th Cir. 1994), aff’d, 115 S. Ct. 2199 (1995); United States v. Packwood, 848 F.2d 1011 (9th Cir. 1988). Although the law in the area of breach of non-delineated agreements has been fairly well-delineated by the circuit courts, the issue of whether a defendant’s breach of the conditions of a non-prosecution agreement is “material” so as to allow the government to be released from it obligations is less than clear.

In a ruling which may shed new light on the issue, a panel of the Fifth Circuit Court of Appeals recently determined that the conviction of a cooperating RICO conspiracy defendant, prosecuted under 18 U.S.C. § 1962(d), must be reversed, based on the improper revocation of his transactional immunity agreement. United States v. Castaneda, 162 F.3d 832 (5th Cir. 1998).

Raymond Castaneda, the owner of an auto-repair shop and towing service in Brownsville, Texas, had conspired with a local county prosecutor’s investigator to solicit bribes from various individuals accused of drunk-driving in exchange for getting the charges dismissed or sentences reduced. Castaneda’s role in the conspiracy was that of middleman; his partner, who worked for the prosecutor’s office, was acknowledged by all parties to be the principal in the scheme to have criminal charges against defendants reduced or disappear. After the initiation of an investigation of the county prosecutor’s office by the FBI , an agent sought and received Castaneda’s cooperation in exchange for a non-prosecution agreement based on a generous transactional immunity agreement, the consideration of which was the obligation to “tell everything he knew” about the principal’s criminal activity.

Castaneda provided more than sufficient information against the investigator to assure a subsequent guilty plea. However, almost one year after Castaneda’s debriefing interview and near the end of the sitting grand jury’s deliberations, the government notified Castaneda that he had “failed to provide .. . relevant material information concerning criminal activities of which he was well aware,” and had therefore violated the transactional immunity agreement resulting in a revocation of the government’s promise not to prosecute. Less than 24 hours after the letter was sent, the grand jury returned a seven-count indictment against Castaneda, as well as the principal target of the investigation against whom he had cooperated.

During his trial and after his conviction for RICO conspiracy and sentence of 33 months, Castaneda unsuccessfully challenged the government’s recision of its agreement, arguing that he had provided considerable accurate and incriminating information about the principal, and that any omissions were essentially inadvertent or duplicative and therefore did not amount to a material breach. Castaneda, 162 F.3d at 835. Judge Jacques L. Wiener, Jr., writing for the Fifth Circuit panel, agreed with Castaneda, reversing his conviction, vacating his sentence and remanding the case for entry of a judgment of acquittal. Id at 839-40. Judge Wierner ruled that in the context of general contract law “a breach is not material unless the non-breaching party is deprived of the benefit of the bargain.” Here, the government got everything it wanted and more, and in the “11th hour” turned against its own cooperating witness unfairly. Id.

In its opinion, the panel articulated a logical way to analyze the concept of “material breach” by comparing it with the “converse” principle of substantial performance under contract law. Where a party’s nonperformance “is innocent, does not thwart the purpose of the bargain, and is wholly dwarfed by the party’s performance,” the courts have ruled that the party has substantially performed and the other party is not entitled to recision. See, Id. at 838. Ruling that the same approach is applicable in determining the materiality of a breach in the context of non-prosecution agreements, the panel concluded that the conduct identified by the government as a breach consisted of only relatively insignificant omissions, and did nothing “to frustrate the government’s prosecution.”

In a strong rebuke to the less-than-honorable conduct of the government in its dealings with Castaneda, Judge Wiener concluded:

    It ill behooves government agents and prosecutors to enter into agreements of transactional immunity with mid-level co-conspirators, milk them of substantial leads and information that literally make the government’s case against the big fish while coincidentally giving the government a lay-down winning hand against the cooperating co-conspirator; then, at the last moment, rely on some technical or relatively minor deficiency in performance to pull the rug from under the cooperating informant by claiming a breach and proceed to prosecute him in a slam-dunk case based largely on his own revelations. Yet, this is precisely what we perceive to have happened here, and due process cannot abide such behavior.
Id. at 839-40.

Prosecutors Licensed To Commit Misconduct in Fourth Circuit

We revisit what was formerly a defense victory after a federal district court in South Carolina dismissed an indictment against five state legislators due to egregious prosecutorial misconduct. See United States v. Taylor, 956 F. Supp. 622 (D. S.C. 1997); The Champion, RICO Report (June 1997). In support of what it recognized as an unusual and extreme exercise of its supervisory powers, the trial court made detailed, individualized findings reflecting a pattern of deliberate misconduct, including discovery violations, intentional withholding of exculpatory information, presenting perjured testimony, and making misrepresentations to the court. See id. However, on November 23, 1998, the Fourth Circuit reversed the dismissal of these indictments, holding that absent prejudice to the defendants resulting from the prosecutorial misconduct, the trial court had exceeded the scope of its supervisory powers. United States v. Derrick, 163 F.3d 799 (4th Cir. 1998).
The most troubling aspect of Derrick is not its primary holding – in light of the existing Supreme Court precedent requiring prejudice, in most instances, in order to dismiss an indictment against a criminal defendant, the appellate decision is not extraordinary. However, after reaching that conclusion, the Derrick court did not remand for a determination of prejudice. Instead, it summarily concluded that since there would be a new trial, albeit six years after the alleged commission of these crimes, the defendants could not be prejudiced. Derrick, 163 F.3d 799. Even more disturbing, the panel went to great lengths to strip all credibility from the district court’s factual findings of prosecutorial misconduct, thus, foreclosing the possibility that the dismissal of the indictments could stand.

In the decision, the Fourth Circuit notes that the Derrick case “arises from an FBI investigation into political corruption in the South Carolina legislature,” Derrick, 163 F.3d at 803. In fact, Derrick really arose out of the arrest of a South Carolina lobbyist, Ronald Cobb, who had been moonlighting as a drug dealer. See id. Cobb was caught attempting to invest in a drug deal with an undercover government agent. See id.; The Champion, RICO Report (June 1997). Preferring not to be prosecuted for his crimes, Cobb offered to prove true the FBI’s long held suspicions of widespread corruption in the state legislature. He was set free to set up state legislators in a sting operation. Taylor, 956 F. Supp. at 624.

What the defendants did not learn until six years and one remand later, was that Cobb had run roughshod over the FBI, taking over the investigation, flatly refusing to testify against a senator friend of his, while targeting his own enemies, and with government approval, going after black members of the state house. Id. at 653 (“I would make the comment [to the FBI] a lot of times that I could have the whole black caucus. There would be no more damn niggers”). Cobb had also engaged in blatant drug use and committed subsequent perjury about it, id. at 650, all with knowledge of the prosecution team.

“Cobb’s investigation” resulted in charges of extortion under color of official right and conspiracy to commit extortion, in violation of the Hobbs Act, 18 U.S.C. § 1951, filed against five South Carolina legislators. This particular trial, pre-empted by the dismissal of the indictment due to prosecutorial misconduct, would not have been the first in the case – indeed, three of the five defendants had already had convictions overturned on appeal for different reasons which included the playing of inadmissible tape recordings and giving of incorrect jury instructions. See Taylor, 956 F. Supp. 622.

On remand for the second trial, the defense team consisting of attorneys, Joel Collins, Jr., Lionel Cotton, James Bell, II, James Lengel, Jack Swerling and Thomas Brittain, began filing discovery motions alleging that Brady materials had been intentionally withheld. The district court was forced to become deeply involved in discovery, reviewing volumes of unproduced documents, in camera, holding several evidentiary hearings where government agents had to testify regarding the failure to produce certain FBI documents, and various other unsavory matters, including star witness Cobb’s other bribery attempts, “his deal” not to prosecute his friend Senator Lindsey, and his cocaine trafficking. At the same time, an Office of Professional Responsibility (OPR) investigation into the local prosecutors’ misconduct commenced, which, not surprisingly, ultimately resulted in an OPR finding of no “intentional wrongdoing.”

Nevertheless, the United States Attorneys Office for the District of South Carolina recused itself from further prosecution of the case and attorneys from the Main office of the Department of Justice’s Public Integrity Section (“PIS”) assumed responsibility for the case). It was only after all of this that the true and more complete picture of the misconduct was revealed. The district court itself admitted that it had erred in the early years of the case by failing to question and instead believing, without substantiation, the government’s misrepresentations to the court regarding discovery and the cooperating witnesses. Taylor, 956 F. Supp. at 658. Yet the Fourth Circuit cited these early 1991 rulings, denying defense claims of Brady violations and perjury as an indication that the district court’s 1997 decision to credit these same claims was an inexplicable about-face. See Derrick, 163 F.3d at 830 (noting that the district court had “abruptly reversed course” in changing its mind six years later).

Regardless of the appellate court’s incredulous assertion about the judge, it apparently took years for previously pro-prosecution Judge Falcon B. Hawkins to conclude that the prosecution committed misconduct and that it was “repetitious, flagrant and longstanding.” See Taylor, 956 F. Supp. at 660, 658. Judge Hawkins noted that the large-scale withholding of patent Brady materials was “unprecedented” and too pervasive to be anything but deliberate. See id. at 658-59 (for example, “it is unfathomable that the government could have possibly considered the drug investigation of Cobb prior to his employment as a confidential informant . . . anything but relevant”) (emphasis in original).

Regarding the testimony presented, the district court recounted numerous instances of perjury by several prosecution witnesses, repeated misrepresentations by government attorneys, and though “reluctant[ly]” it clearly found that what the prosecution did “in several instances constitutes subornation of perjury.” Id. at 660.

Naturally, the court found the continuous misrepresentations made to the bench to be most offensive, id. at 659, stating, “the various and repeated acts of the government were simply wrong,” “in bad faith, and its misconduct is not only greatly offensive to this court, but has interfered with this court’s duty to insure the proper administration of justice.” Id. at 661.

For these reasons, the district court exercised its supervisory power, holding the “evidence of the government’s misconduct, from the time this investigation commenced until the present, in its totality, is sufficiently egregious to warrant dismissal of these indictments with prejudice under the doctrine of supervisory power.” Taylor, 956 F. Supp. at 661; see also The Champion, RICO Report (June 1997)(discussing Derrick case). The court stated that “the resultant injustice to these defendants cannot be fully remedied by new trials,” but it did not discuss whether there had actually been prejudice. It dismissed the indictments without addressing prejudice and without citing recent authorities, rather relying on some older cases.

It is clear that the district court believed it was on firm ground in exercising its supervisory power to dismiss the indictment in the face of what it deemed to be “egregious misconduct” by the government without raising any other issues or making any other findings. See Taylor, 956 F. Supp. at 658 (citing case where court found it need not address prejudice or due process violations where misconduct is so aggravated as to warrant dismissal). Thus, in its already lengthy decision, rightly or wrongly, it did not believe it necessary to address prejudice or any constitutional issues, i.e., whether the misconduct rose to the level of a due process violation or interfered with the defendants’ ability to present a defense in violation of the Sixth Amendment. See Taylor, 956 F. Supp. at 623 (court “believes there is sufficient precedents to dismiss the subject indictments without addressing these issues”); id. at 658 (discussing possible constitutional violations as result of misconduct but finding those issues “need not be addressed”).

Against that complicated backdrop, the Fourth Circuit held that based upon Supreme Court precedents, not addressed in the district court’s decision, that court could not exercise its supervisory power to dismiss indictments due to prosecutorial misconduct, “absent a showing that the misconduct prejudiced the defendant.” Derrick, 163 F.3d at 807. In light of these precedents, among others, the Fourth Circuit’s conclusion that prejudice is required to dismiss indictments on this basis, is not surprising. However, there are certain persuasive contrary authorities. See, e.g., United States v. Kojayan, 8 F.3d 1315, 1325 (9th Cir. 1993) (court remanded for district court to decide whether to dismiss under supervisory powers as “sanction for the government’s misbehavior” without referencing prejudice,); United States v. Bernal-Obeso, 989 F.2d 331, 337 (9th Cir. 1993) (without reference to harmless error or prejudice, court remanded for inquiry into whether informant lied to DEA stating: “[s]hould the court uncover egregious wrongdoing by the government . . . nothing in this opinion forecloses consideration by the court of dismissing the indictment for outrageous government conduct”).

Of course, in both Kojayan and Bernal-Obeso, it would have been easy to adopt the Fourth Circuit’s “new trial cures all” approach, discussed below. However, these cases, decided by some of the Ninth Circuit’s most conservative jurists, Judge Kozinski in Kojayan and Judge Trott in Bernal-Obeso, declined that option instead, permitting the district court to exercise its supervisory power to dismiss.

In Derrick, the Fourth Circuit simply assumed that the district court failed to make any findings of prejudice even though Supreme Court precedent required it, because it could not do so. See Derrick 163 F.3d at 809. Rather, than remanding for the appropriate inquiry, it concluded that “any prejudice that arguably existed as a consequence of discovery violations is fully remedied by this court’s orders of new trials” and “any prejudice . . . at their original trials . . . would also be fully cured by retrial.” Id. Further, it decided, without citation to the record, that the district court did not find prejudice for this precise reason — because the government (though it took them six years) had finally produced all the discovery materials, the withholding of which constituted the prejudice, thus, “the defendants now have all of the discovery materials to which they could possibly be entitled — and considerably more — available to them for use at their retrials.” See Derrick, 163 F.3d at 809.

Not only does the appellate court misstate the district court’s position since the district court did not make findings regarding prejudice only because it did not believe it needed to, but it also ignores the fact that the trial judge actually indicated that there was prejudice when he stated that the defendants’ Sixth Amendment rights were violated and that the injustice could not be remedied by a new trial. See Taylor, 956 F. Supp. at 658, 661. In addition, this “new trial cures all” position utterly fails to acknowledge that had there been no discovery violations, the defendants could and should have been acquitted at their original trials. Apparently, depriving them of the years-earlier opportunity for vindication is not considered prejudicial. In any event, these determinations were not for the Fourth Circuit to make, especially in light of the voluminous record and the trial court’s intimate familiarity with the evidence in the case. At best, the case should have remanded to the district court, for a determination of prejudice.

The panel not only improperly decided the prejudice issue, but went further, usurping the district court’s fact-finding function regarding the nature of the misconduct, and intentionally revising the case history to preclude finding a pattern of misconduct sufficient to warrant reversal. It did this because the Fourth Circuit was forced to recognize that the Supreme Court has left open the possibility that actual “prejudice” is not required where “a pattern of recurring violations by investigative officers that might warrant the imposition of a more extreme remedy in order to deter future lawlessness” existed. See Derrick, 163 F.3d at 809 (quoting United States v. Morrison, 449 U.S., 361, 365 n.2 (1981)).

The Supreme Court left “open” thispossibility, as part of the well-established, wide-ranging supervisory power doctrine which exists with respect to the administration of justice in federal judicial proceedings. United States v. Hasting, 461 U.S. 499 (1983); see also Chambers v. Nasco, Inc., 501 U.S. 32, 41, 46, 49-50 (1991), (court “has the power to conduct an independent investigation to determine whether it has been the victim of fraud,” and if inquiry reveals “that fraud has been practiced upon it or that the very temple of justice has been defiled,” it may impose sanctions irrespective of the existence of other means of redress). These powers permit courts to formulate rules not specifically required by the Constitution or Congress for three reasons: (1) to provide a remedy for violations of recognized rights; (2) to preserve judicial integrity by ensuring that a conviction rests on appropriate considerations validly before the jury; and (3) to deter illegal conduct. Hasting, 461 U.S. at 505.

In Hasting, where the prosecutor made an isolated but improper comment upon the defendant’s failure to rebut the prosecution case, it was clear that both the district court and courts of appeals were using the case to teach a lesson to the prosecutors of its district who had been “indifferent to the frequent admonitions of the court.” Hasting, 461 U.S. at 507. Thus, the Supreme Court noted that this third doctrinal goal — deterring illegal conduct — “is an inappropriate basis for reversal where . . . the prosecutor’s remark is at most an attenuated violation of [the defendant’s right to remain silent] and where means more narrowly tailored to deter objectionable prosecutorial conduct are available.” Hasting, 461 U.S. at 506. So it was under these particular circumstances that the Court issued the now oft-quoted edict: “[s]upervisory power to reverse a conviction is not needed as a remedy when the error to which it is addressed is harmless since, by definition, the conviction would have been obtained notwithstanding the asserted error.” Id.

Subsequently, in Bank of Nova Scotia v. United States, 487 U.S. 250, 250 (1988), the Court again held that “a federal court may not invoke supervisory power to circumvent the harmless-error inquiry” but also held that this harmless error inquiry applies only to nonconstitutional errors and that certain fundamental errors were not subject to the “harmless error” analysis. Id. at 255. More important, for our purposes, the Court noted that, in applying a prejudice requirement, “we are not faced with a history of prosecutorial misconduct, spanning several cases, that is so systematic and pervasive as to raise a substantial and serious question about the fundamental fairness of the process . . . .” Nova Scotia, 487 U.S. at 259 (emphasis added).

Hence, post-Hasting, several courts have taken the position, predicated both upon Hasting’s factual scenario and the Nova Scotia caveat, that “resort to a court’s supervisory power has not been foreclosed altogether as a means to remedy government misconduct not injuring the defendant”— it “may not be entirely a dead letter.” See, e.g., United States v. Santana, 6 F.3d 1, 10 (1st Cir. 1993); see also CACJ Forum, Vol. 20, No. 4, Tarlow, The Supervisory Power to Dismiss: An Essential Solution to the Problem of Prosecutorial Misconduct (1992) (containing an extensive discussion of solving prosecutorial misconduct problem by increased use of judicial supervisory powers).

One of the proponents of this principle, the First Circuit, has stated that the Supreme Court left open “the possibility that the goal of deterring future misconduct would justify using the supervisory power to redress conduct not injuring defendants if the conduct is plainly improper, indisputably outrageous, and not redressable through the utilization of less drastic disciplinary tools.” Santana, 6 F.3d at 10. While Santana, where the government distributed 13.3 grams of heroin and failed to recover it, did not warrant further amplification or application of this doctrine, that court clearly left these determinations “for another day.”

Although recognizing the door left open by Hasting and Nova Scotia, the Fourth Circuit was quick to close it, holding that the district court never really found a “pattern” of intentional misconduct, never really made any such “findings,”and, that even the district judge tried to make such “findings,” they were erroneous. Derrick, 163 F.3d at 810 & n.11. Its disagreement with, and reversal of, the district court’s particularized findings of misconduct which were based upon the trial judge’s firsthand participation in discovery, as well as the witnesses’ testimony at the evidentiary hearings, appears factually wrong, and as a legal matter, contravenes all principles of appellate review. See United States v. Barrera-Moreno, 951 F.2d 1089, 1091 (9th Cir. 1991) (dismissal of indictment on supervisory powers is reviewed for abuse of discretion).

In reaching (and overreaching) to arrive at this conclusion, the Fourth Circuit plays semantical games with what it refuses to recognize as the district court’s findings. In fact, these “non findings” constitute 26 of the almost 40-page lower court opinion and recite numerous instances of prosecutorial misconduct, spanning years. See Taylor, 956 F. Supp. at 632-657. For example, Derrick found that because the district court stated that FBI Special Agent Clemens “was not entirely truthful in his testimony,” see Taylor, 956 F. Supp. at 648, instead of finding explicitly that the agent committed perjury, the district court had only “implied such perjury” and never made a finding of it. Derrick 163 F.3d at 810 n.11.

In the same vein, the district court’s statement that the government’s actions “suggest a total avoidance of pursuing information that might have proved adverse,” see Taylor, 956 F. Supp. at 660, means that, it had merely “implied but did not actually find that the government’s investigation [into its informant’s illegal activities] was inadequate.” Derrick, 163 F.3d at 810 n.11. Even more ludicrous is the Fourth Circuit’s conclusion that, with regard to a particular government witness, the district court was “unprepared to make a specific finding” of perjury because it only stated that this government witness’s testimony “would have to be perjured unless alternatively the government lied to the court at [the witness’s] sentencing hearing.” Derrick, 163 F.3d at 811 n.11; see Taylor, 956 F. Supp. at 660.

Similarly, the Fourth Circuit notes that the district court stated that the government “allow[ed] testimony from Cobb, Clemens and Greer that it knew to be untrue stand uncorrected,” rather than “characterizing the testimony . . . as perjurious” — thus, this could not constitute a finding of prosecutorial misconduct. Derrick, 163 F.3d at 811 n.11 (emphasis in original). Since Derrick could not play its semantical games with the district court’s clear and unequivocal finding that the prosecution “suborn[ed] perjury” in several instances, it then chastised the district court for not specifying which of the three government witnesses’ testimony, (the perjurious content already set forth at length in the district court decision), gave rise to the subornation of perjury conclusion. See Derrick, 163 F.3d at 811 n.11 (recognizing that the district court was “seemingly finding that the government had suborned perjury ‘in several instances’ but without identifying whether the referenced perjury was committed by Cobb only or also by Greer and Clemens”).

The appellate court’s usurpation of the fact-finding role is even more egregious in light of the trial court’s acknowledgment that the six year long record is “so vast as to make it virtually impossible” to cite to all the arguments and counter-arguments regarding misconduct. Taylor, 956 F. Supp. at 632. In its selective fact-finding mission, the Fourth Circuit picked a few violations, minimized the prosecution’s actions and whitewashed the entire course of misconduct. The panel was required to spend pages and pages of its unauthorized de novo review, explaining exactly how the district court made these “numerous errors” in its misconduct findings. See Derrick, 163 F.3d at 834.

Of course, this review was without the benefit of observing the witnesses, hearing the testimony and reading the volumes and volumes of discovery, all of which was done by the district court. See Taylor, 956 F. Supp. 622. The appeals court incorrectly states that the district court blindly followed defense suggestions regarding the misconduct without conducting its own independent investigation or making its own credibility decisions. Derrick, 163 F.3d at 810. Even a cursory reading of the district court’s decision spanning, as we noted, near forty pages, unequivocally refutes that statement. See generally Taylor, 956 F. Supp. 622.

Of course, if this was a defense appeal, this inordinate and highly unusual level of review (a “careful parsing of the district court’s lengthy opinion” as Derrick calls it) would have never been considered let alone conducted. It is of small comfort that some of the Fourth Circuit’s words may come back to haunt them. For example, where government witness Cobb repeatedly refused to characterize the thousands and thousands of dollars he “paid” to his friend Senator Lindsay, as bribes, the district court labeled this as perjury, but, according to the appeals court, it was merely Cobb’s “subjective belief” that he was not bribing Lindsay. Derrick, 163 F.3d at 830-31. Of course, part of the deal Cobb struck with the prosecution guaranteed that he would never have to testify against Lindsay — the appellate court fails to address why Cobb would need this agreement if he “subjectively” believed that he had never bribed Lindsay. The district court arrived at its conclusion about Cobb, by not only witnessing his perjured testimony but also by reviewing the FBI 302s which repeatedly characterized Cobb’s conduct with Lindsay as a bribe or a “payoff.” See id. However, for future reference, perhaps, even providing a basis for a jury instruction in some other case, the Fourth Circuit finds — “the fact that the FBI 302 characterizes the payments to Lindsay as ‘payoffs’ does not mean that Cobb himself so characterized the payments. The words recorded in the FBI 302 are the FBI agent’s characterization of what Cobb said, not the words that Cobb actually spoke.” 163 F.3d at 831.

The Fourth Circuit’s opinion improperly sanitizes the government’s conduct to justify its finding that neither real prejudice nor any pattern of misconduct existed. It calls the district court’s findings of intentionally withheld Brady materials, misrepresentations made to the court, subornation of perjury, and allowing a drug-addicted, racist criminal-turned-informer to ride herd on the FBI and the United States Attorney’s Office, “erroneous” and “unsupported by the record.” See id. In fact, it is the appeals court’s conclusions which lack record support and, in truth, constitute an abdication of any sense of responsibility to ensure fairness in the judicial proceedings.

Unfortunately, these are not the only federal judges who have turned a blind eye to prosecutorial misconduct present in our criminal justice system. Abandoning the responsibility to eliminate this misconduct encourages rogue prosecutors to travel beyond the edge of the envelope and, ultimately derails the search for the truth.
Tainted Verdicts Protected By Local Rules
Preventing Post-Trial Juror Interviews

The seepage of prejudicial extrajudicial information into a jury’s deliberation process in criminal trials is all too common. The factual inquiry into whether such information has tainted the deliberation usually can only be undertaken in states and federal districts where counsel are free to interview jurors after a verdict. See, Marino v. Vasquez, 812 F.2d 499, 507 (9th Cir. 1987). While jurors may often refuse to talk to defense counsel or not admit what, in fact, has actually occurred, prejudicial extrajudicial information can rarely be discovered if juror debriefings are prohibited. In addition, juror interviews are essential for lawyers to evaluate their job performance and improve their litigation skills.

Unfortunately, a number of federal and state jurisdictions have local rules prohibiting counsel from interviewing jurors. See, ABA Model Rule of Professional Conduct 3.5 (prohibiting lawyers from communicating ex parte with jurors “except as permitted by law”); Local Rule of United States District Court, M.D. Fla. Rule 2.04(c); Local Rule of United States District Court, E.D.N.Y., Rule 3.0; Local Rule of United States District Court, Wis., Rule 8.07. Although the rules provide that attorneys can contact jurors with the permission of the court, this permission is rarely granted unless it comes at the request of the prosecution. See, e.g. United State v. Chung An-Low, 851 F.2d 547 (2d Cir. 1988); see also, United States v. Harrelson, 713 F.2d 1114, 1118 (5th Cir. 1983) (upholding trial court restriction on juror interviews). Unless something is explicitly revealed to the court or a news reporter outside of deliberations, counsel in these jurisdictions can almost never demonstrate good cause to conduct juror interviews. Although the purported rationale for the local rules is that they are designed to protect jury privacy, in fact the purpose for the rules is inconsistent with a fundamental principle of our criminal justice system, which ensures the right of a defendant to confront those who testify against him or her. Federal and state courts in California have always permitted such interviews, and this author has never heard of a case where a juror was mistreated. Jurors can and often do refuse to talk if they do not want to discuss a case.

There may be no better example of the importance of juror interviews than in the recent case involving Charles Keating. United States v. Keating, Case. No. 93-50576, 1996 U.S. LEXIS 15276 (9th Cir. 1996); (related case), Keating, 147 F.3d 895 (9th Cir. 1998). In its unpublished opinion, the Ninth Circuit held that the trial court had erred in not holding an evidentiary hearing to determine whether there was a reasonable possibility extrinsic evidence had affected Keating’s conviction in 1992 on federal racketeering, fraud and conspiracy charges. Id. After remand, each of Keating’s federal convictions, and those of his son, “C III”, involving their activities at Lincoln Savings and Loan. Keating’s lawyer had conducted extensive juror interviews that convincingly demonstrated that the jury had been tainted by inadmissible prejudicial outside information. See, The Champion, RICO Report (Dec. 1997). The convictions were reversed after admissions that the jurors had obtained this extrajudicial knowledge of Keating’s widely publicized problems and had discussed his inadmissible prior state convictions during deliberations. The only reason this blatant juror misconduct that tainted the verdict was discovered was because his dedicated defense lawyer, Stephen Neal of Cooley, Godward, Castro, Huddleson and Tatum, was free to interview jurors, in California, after the verdict.

Unfortunately, the uncovering of misconduct in Keating stands in stark contrast to a recent Seventh Circuit opinion in Wisconsin. Diettrich v. Northwest Airlines, Inc., 168 F.3d 961 (7th Cir. 1999). In a lawsuit alleging age discrimination, a 28 year veteran salesman for Northwest Airlines claimed he was terminated during a company reorganization in 1992 based on his age. After a jury returned a verdict in Diettrich’s favor and the district court entered an award for damages and attorneys’ fees, Northwest moved for judgment as a matter of law and/or a new trial. Shortly thereafter, the district court granted Northwest permission to interview jurors after it requested to do so “to learn which trial techniques worked and which did not, so that the attorneys could use the information for future trials.” Id. However, during these interviews, Northwest also uncovered facts which supported its motion for overturning the jury verdict and the granting of a new trial. Although none of the information uncovered was evidence of misconduct by the jury, the information gleaned during the interviews would have supported a motion for new trial.

Nevertheless, the inclusion of the information in Northwest’s briefs asking the court to overturn the verdict was met with hostility by the Seventh Circuit panel. In an opinion authored by Circuit Judge Diane P. Wood, the panel found that Northwest attorneys had abused the court’s grant of permission to interview the jurors when it “relied on its lawyers’ post-verdict dialogue with the jury . . . to support its motion for overturning the jury verdict.” Id. at 964. The panel found that Northwest sought permission to interview the jurors based on false pretenses and went far beyond the scope of the district court’s permission, and thus engaged in “highly improper behavior.” Id.

Although the panel refused the plaintiff’s request to sanction Northwest by affirming the jury’s verdict, it referred the matter to the District Court for the Eastern District of Wisconsin, as well as the Supreme Court of Wisconsin and Wisconsin Board of Attorneys’ Professional Responsibility for investigation. The court also overturned the jury verdict in favor of Diettrich, finding that the circumstantial evidence in the case did not justify submitting the case to the jury in the first instance. Nevertheless, it refused to consider any of the facts uncovered during the course of the so-called “improper” jury interviews as part of its decision.

While it would clearly be improper for counsel to heedlessly ignore the ground rules a district court sets for conducting certain jury interviews, it cannot be said that jury interviews should be so restrictive as to only allow for interviews to be conducted which only educate lawyers for the purpose of future trials. If the purported purpose behind prohibiting juror interviews, protecting juror privacy, is already breached, why not allow for questioning relating to jury misconduct during jury deliberations? A defendant’s ability to establish prejudice from juror misconduct comes only from a close analysis of the substance and impact of the contamination, which can be most accurately determined from jury interviews. See, Jeffries v. Wood, 114 F.3d 1484 (9th Cir. 1987). There is almost never a demonstrable sound reason to refuse to permit counsel (remember, they are officers of the court) to conduct post-trial juror interviews. The local rules preventing juror interviews ought to be revised to permit such interviews unless a persuasive particularized showing can be made that these interviews should not be permitted.


    Readers are a vital source of information without which this column could not be supplied with current information on RICO developments. Information regarding developments in pending cases, decisions and interesting briefs and motions should be sent to:

    RICO Report
    Barry Tarlow
    9119 Sunset Boulevard
    Los Angeles CA 90069
    Phone (310) 278-2111
    Fax (310) 550-7055




National Association of Criminal Defense Lawyers (NACDL)
1660 L St., NW, 12th Floor, Washington, DC 20036
(202) 872-8600 • Fax (202) 872-8690 • assist@nacdl.org