United States v. William J. Kilroy

SILBERMAN, Circuit Judge,

dissenting:

I agree with my colleagues that when the government promised the appellant “retroactive use immunity” as part of his August 14, 1985 plea bargain, it agreed, as a matter of contract, not to use any of his immunized statements against appellant — just as would be so if the government had obtained his testimony under 18 U.S.C. § 6002 . It seems rather straightforward and obvious to me that the government, again as a matter of contract, therefore assumed the burden of showing in any subsequent prosecution exactly what it would have to show had the immunization been granted by order rather than as part of the plea agreement. That the government shoulders the “heavy burden,” United States v. North, 910 F.2d 843, 854 (D.C.Cir.1990), under Kastigar v. United States, 406 U.S. 441 , 92 S.Ct. 1653 , 32 L.Ed.2d 212 (1972), of proving the negative— that the prosecution has not used the “compelled testimony as an ‘investigatory lead,’ ” nor used any “evidence obtained by focusing investigation on a witness as a result of his compelled disclosures,” id. at 460 , 92 S.Ct. at 1664 -65 — is part of the protection given an immunized witness. The appellant was granted that legal position by the government, and he has a legitimate right to assert it fully in this court. I therefore do not understand why the majority comes so grudgingly, see maj. op. at 685, to this position.

Be that as it may, I do not think the government has met its burden. The district judge’s finding that it did — insofar as it could be described as a finding of fact — is clearly erroneous, but I think the district court (and the majority) decided the issue incorrectly as a matter of law. Both the district court and the majority seem to ask whether there was adequate evidence to convict the appellant without drawing upon the immunized testimony or, put another way, to require that the appellant must show that there is a “but for” relationship between the two indictments, see maj. op. at 688. That approach seems to me to reverse the burden (the majority’s initial discussion does suggest that it believes that the burden properly belongs on appellant rather than the government) and moreover applies a standard not called for in Kastigar and which we did not apply in North . The question is not whether the use of immunized testimony, was prejudicial to a defendant, but rather was any use made. If it was, than the government loses — without regard to whether the use was outcome determinative.

The first question which the government is bound to address is: Was Kilroy targeted for investigation by the Labor Department because of his prior immunized testimony? This question breaks down into two parts: Was Kim’s decision to audit NSCC’s funds influenced by knowledge of Kilroy’s Las Vegas indictment, and, even if not, was DOL’s (and the U.S. Attorney’s) decision to focus an *690 investigation on Kilroy “wholly independent” of its confessed knowledge of the indictment. In that respect, the government concedes that, because of the plea agreement, the first indictment and Kilroy’s interviews are treated as an immunized package. Kilroy had been interviewed before he was indicted as well as after, but all of his statements were immunized. And his pre-indictment statements are deemed to have led to his indictment in Las Vegas. That is what the parties mean by retroactive immunity. If, then, an Assistant U.S. Attorney in Las Vegas had called Pitt, the Labor Department investigator in Washington, to “alert” Pitt to Kilroy’s indictment, there would be no question but that the Kastigar test would have been violated.

Instead, the Labor Department was contacted by Williams, an independent auditor whose audit turned up the missing $573,000 and the fact of Kilroy’s responsibility. Williams was, in turn, told by James Kim, NSCC’s comptroller, to conduct the audit shortly after Kilroy had been indicted. It is not clear from the government’s brief whether it concedes that if Kim had learned of the indictment and thus was prompted to ask Williams to conduct the audit, Kastigar would have been violated. But, I believe the question is answered by Kastigar itself— whether or not one calls this sort of use “non-evidentiary.” 1 Kastigar said that the immunized witness must not be placed in a position by reason of his testimony that is any less advantageous than if he had invoked the Fifth Amendment. That surely means that if Kilroy’s testimony is deemed to have been immunized and to have led to his indictment and a private person targets Kilroy for investigation based on knowledge of the indictment, Kilroy’s testimony was used against him. That a private person, rather than a government official, is one who gains the knowledge and then presents his suspicions — in this case the fruits of an investigation — over to the government is of no significance to the immunized witness. In both eases the immunized witness is worse off— considerably worse off — than he would have been if he had insisted on his right against self-incrimination.

The majority, however, opines that even if Kim had directed an investigation of Kilroy because he learned of the prior indictment Kastigar would not have been violated. The majority’s view apparently is that, under such circumstances, it cannot be said that Kilroy’s immunized statement was “used” against Kilroy. I frankly do not understand why. It certainly cannot be because Kim was not a government official; we crossed that analytical bridge in North when we concluded that for Kastigar purposes it mattered not that North’s immunized testimony was “used” by non -government persons (witnesses and their counsel) whose interests were at least somewhat adverse to the defendant, even if the prosecutors were entirely unaware of the immunized testimony. Indeed, it could be thought that this case is a fortiori to North because Kim’s knowledge of the indictment leads directly (at least under the hypothesis) to the audit and hence to the government’s investigation whereas in North it was argued vigorously that the witnesses’ use of the immunized testimony did not affect the government at all. The focus of the word “use” is not on who uses the immunized testimony but rather on whether it was used against the defendant — by anyone. Put another way, the way the Supreme Court did in Kastigar , has the immunized witness been put in a less favorable position than he would have been if he had pleaded the Fifth Amendment and refused to testify? See Kastigar, 406 U.S. at 462 , 92 S.Ct. at 1665-66 .

To be sure, the Second Circuit in United States v. Helmsley, 941 F.2d 71 (2nd Cir.1991) (upon which the majority relies) took another tack. Although the court assumed a direct causal link between Ms. Helmsley’s immunized statements, The New York Times ’ article reporting her testimony, The New York Post’s investigation, and finally the U.S. Attorney’s indictment, the court nevertheless concluded that the fortuitous causal link in that chain circumvented the policies of the Fifth Amendment. Despite my enormous respect for Judge Winter’s an *691 alytical powers, I confess that his logic here escapes me. He gives as a hypothetical example, from which he reasons, the following:

If a grand jury witness testifying under a grant of immunity were recognized by a grand juror as the perpetrator of a bank robbery committed the week before, an undeniable causal link between the immunized testimony and á conviction for bank robbery would exist, but a plausible Fifth Amendment argument could not be made.

United States v. Helmsley, 941 F.2d at 82 . I think his premise is faulty; there is, in that example, no link between the bank robber witness’ immunized testimony and the grand juror’s recognition of him as a bank robber. The grand juror would have equally recognized the witness as a bank robber if the witness had come before the grand jury and took the Fifth Amendment. The Second Circuit, as well as the majority here, is obviously uncomfortable with the consequences of the press’ reporting of immunized testimony; it can be spread to those out of government who might conclude, based on the reporting, that the witness is a bad person and should be investigated for matters not directly related to the immunized testimony. That is, of course, true, but I do not understand either the Second Circuit or the majority to suggest that a government investigator, having learned of immunized testimony by reading of it in the newspapers, could start an investigation of the witness without leading the government afoul of Kastigar . Therefore, I do not see any principled reason to' permit the government to gain the benefit of circumventing Kastigar by employing a private actor, even unwittingly, to do what the government itself could not. Nor do I think the majority (and the Second Circuit) have a legitimate concern that applying Kastigar to such a situation would somehow extend use immunity beyond the Fifth Amendment protection it replaces. Surely, one of the advantages of taking the Fifth Amendment was preventing the possibility that yóur testimony would be used against you indirectly, through newspaper accounts that might lead investigators — public or private — to explore matters not directly germane to transactions about which a witness was questioned. Kast-igar made clear that the constitutionality of section 6002 depended on a witness not being put in a worse position by reason of a grant of use immunity than he would have been prior to passage of the statute. And the court emphasized that focusing an investigation on the witness because of his immunized testimony would result in just that worsening of position. See Kastigar, 406 U.S. at 460 , 92 S.Ct. at 1664-65 . Typically (one hopes) a witness’ immunized testimony -will not pass to newspapers, but surely it is the government and not the witness, given the underlying constitutional concerns, that must bear the risk of that occurring. 2

In our case, although the appellant understandably emphasizes The Baltimore Sun article as a possible reason why Kim ordered an audit, we really do not know why he did so. But it is not the appellant’s burden to provide the reason. The gap in the record represents a failure of the government’s proof. Of course, Kim is dead, but the government did not even produce testimony from other company officials to attempt an explanation (I think it is virtually inconceivable that Kim did not discuss the matter with his associates). The government rested on the speculative testimony of a government employee, Wagner, that the company likely would have ordered an audit sometime- because of a regulatory change. 3 I do not think, under the circumstances, that testimony is enough to establish that Kim was not influenced by Kilroy’s immunized testimony.

Even if Km’s decision to order an audit of NSCC’s funds, however, was “wholly independent” of knowledge of Klroy’s Las Vegas indictment, Kilroy’s immunized testimony would still have been used against him in violation of Kastigar if Pitt’s decision (made in conjunction with AUSA Benner) to investigate Kilroy was influenced, at least in part, by knowledge of that indictment. Under Kastigar , the government has the burden of *692 disproving that influence. Yet the government concedes that Pitt was informed of that indictment at the October 1985 meeting with Williams and the NSCC management.

In early October, 1985, Pitt received a letter from Williams, NSCC’s outside accountant, stating that he had conducted an audit of certain pension funds sponsored by the NSCC and that there were funds missing. Williams requested a meeting with Pitt to discuss his findings, and the meeting was scheduled for October 17, 1985, between Pitt on the one hand, and Williams, Mozer (NSCC counsel), and NSCC’s assistant comptroller on the other. At the meeting, the participants discussed with Pitt the findings reached by Williams, namely that funds were missing and that they had been taken by Kilroy. Williams told Pitt that Kilroy’s attorney had confirmed the amount of funds missing. Significantly, Pitt received from one of the participants a copy of Kilroy’s indictment in Las Vegas.

That afternoon, without conducting an investigation of his own, Pitt called Benner, the Assistant U.S. Attorney, and arranged to meet with him to discuss a possible criminal investigation. When they met the following day, Pitt brought a copy of Kilroy’s Las Vegas indictment. Based on the indictment, Benner called Parry, the AUSA in Las Vegas, to determine, in part, whether immunity had been granted to Kilroy regarding a possible NSCC embezzlement. Benner was informed that Kilroy had only been given transactional immunity for dealings with Ost-erer and that Osterer was not involved in the NSCC embezzlement. Benner was also told by Parry that Kilroy had confessed in Las Vegas to the NSCC embezzlement. The civil investigation by Pitt was formally opened on December 13, 1985. 4

Wagner commenced his criminal investigation in December of 1986 and testified before a grand jury for the first time in March of 1987. That testimony was based solely on Pitt’s conclusions and report. For some reason, not clear from the record, the government did not pursue an indictment for quite some time. Wagner appeared three more times in 1990 before a grand jury, and an indictment followed closely thereafter.

In February 1988, that is, two years before Wagner’s 1990 grand jury testimony, Pitt had travelled to Baltimore and reviewed all of the FBI records relating to Kilroy’s debriefings, including those relating to the NSCC embezzlement, thus tainting himself irretrievably. He drafted a summary of his new-found conclusions and apparently placed that summary in a manila envelope in a file not accessible to Wagner.

At the Kastigar hearing below, however, Pitt testified that following his trip to Baltimore he had a conversation with Wagner in which he “gave him a summary of some of the stuff I looked at.” Pitt could only specifically recall telling Wagner that Mozer’s name had appeared in a suspected racketeer’s rolodex — a possibly incriminating piece of information about the lawyer that was never pursued. As the cross-examination of Pitt clarifies, Pitt did not testify that that was all he told Wagner; he simply could not recall anything else specifically. 5

*693 If Pitt’s decision to investigate Kilroy was based in part on the knowledge that Kilroy was indicted of various crimes of dishonesty in Las Vegas and if that indictment was in turn based on Kilroy’s immunized testimony, then Pitt’s decision to focus investigation on Kilroy indirectly resulted (in part) from Kilroy’s putatively compelled testimony. Kasti-gar would, therefore, have been violated. In other words, Kilroy was placed in a decidedly inferior position because of his immunized testimony.

When a government investigator actually obtains potentially incriminating information (in this case the Las Vegas indictment) directly traceable to immunized testimony and then decides to focus investigation on the protected witness, I rather doubt that the government can ever prove that that decision was “wholly independent” of the witness’ testimony. There is no need for me to so conclude here, however, because the government offered no testimony whatsoever to disprove the link between Pitt’s decision to investigate Kilroy and Pitt’s knowledge that Kilroy had been indicted on insurance racketeering in Las Vegas. 6 And circumstantial evidence suggests that a link may in fact have existed. Pitt, without any investigation on his part, contacted an assistant United States attorney (Benner) to arrange an immediate meeting to discuss a likely criminal case. Is that standard procedure in the Department of Labor when a private auditor complains of missing funds? Would an investigator at the Labor Department always contact an assistant United States attorney based solely on the accusations of a private auditor? On these questions, there is a complete failure of proof by the government. The government offered no testimony to establish what DOL procedures were. For example, if DOL always opened formal investigations whenever a complaint involving more than a given dollar amount arrived, that might be one thing. Pitt’s precipitous decision might well have been influenced, however, by his suspicions that Kilroy was a crook, suspicions grounded on the Las Vegas indictment. And Benner himself testified that he was told by an assistant United States attorney in Las Vegas that Kilroy had actually confessed to the NSCC embezzlement before his second conversation with Pitt authorizing the investigation and that he conveyed that information to Pitt.

So Pitt was exposed to tainted information prior to deciding to focus investigation on Kilroy, and both Pitt and Benner knew of Kilroy’s confession of the NSCC embezzlement prior to the formal inception of the investigation. To be sure, Williams had informed Pitt at their first meeting that Kilroy’s attorney had confirmed the amount of funds missing, and that fact by itself might have led Pitt to act as he did. The absence of any proffered evidence by the government on this matter leaves me able only to speculate on these questions and, hence, unable to conclude that the government has met its Kastigar burden.

I am even more troubled by the government’s inability to document the extent to which Wagner was tainted by his conversation with Pitt following the latter’s 1988 Baltimore trip, and would reverse for that reason alone. In determining that there was no Kastigar violation, the district court relied heavily on its finding that “[a]t no time was the criminal investigator [Wagner] privy to information developed by the government *694 from Kilroy himself.” United States v. Kilroy, 769 F.Supp. 6, 10 (D.D.C.1991) (emphasis added). Even the government concedes that that finding is- erroneous. The government admits that Pitt gave Wagner at least one piece of information that Pitt had discovered after reviewing all of the tainted files in February of 1988 in Baltimore — that Mozer’s name had turned up in a suspected racketeer’s rolodex. Wagner was tainted, then, prior to his three 1990 grand jury appearances. The government continues to assert, however, that “once Pitt was exposed to the immunized statements, he carefully segregated that material and, with the exception of the one irrelevant lapse ... did not relate that information to Wagner.” That assertion distorts the record. As set out above, Pitt candidly admitted that he told Wagner more than simply this one, probably inconsequential, tidbit, that he gave Wagner a “summary of some of the stuff I found.” That statement, by the government’s own investigator, renders the district court’s finding that Wagner was not privy to any tainted information clearly erroneous. And Pitt’s inability to recall specifics makes it impossible for any court to rule out, and — more important — for the government to disprove, the possibility that Kilroy’s testimony was in any way “used” by the government. With the burden on the government, Pitt’s lack of memory is fatal. 7

‡ # i}:

I suspect that underlying the majority’s conclusion contrary to mine is the notion that where, as here, the independent documentary evidence of guilt is strong enough, and where private individuals had on their own uncovered at least some evidence of criminality, any governmental reliance on tainted information was unnecessary, perhaps even gratuitous, and thus reversal is unwarranted. But Kastigar requires the government to prove far more than that it could or indeed would have developed its evidence free of taint. Kastigar “imposes on the prosecution the affirmative duty to prove that the evidence” it gathered was in fact “derived from a legitimate source wholly independent of the compelled testimony.” Kastigar, 406 U.S. at 460 , 92 S.Ct. at 1664-65 (emphasis added). The government has not met this burden.

Of course, my opinion is based on the proposition that the reasoning of Kastigar , written by Justice Marshall, is applicable law. I believe that my position follows logically from that reasoning. I take it that the majority, by not responding to my dissent, essentially concedes its logic. I operate on the premise that notwithstanding Justice Holmes’ famous axiom the essence of legal reasoning, and thus law, is logic — which is why I believe that there is, at least theoretically, a right answer to every case (to be sure, as imperfect humans we may not always find it). I must admit, however, that some Supreme Court justices disagree with these propositions, explicitly or implicitly, and so the reasoning of prior opinions written by other justices may not be followed— even when those prior opinions are not overruled. Instead, often a case is decided in accordance with a majority’s present intuition. In that regard, I have great respect for my colleagues’ intuition on the issues presented by this case — -indeed, I originally shared it myself — so their intuition may well be a reliable predicator as to the ultimate Supreme Court resolution of those issues. But logic led me inexorably in another direction.

* * * * * *

Therefore, I respectfully dissent.

. See maj. op. at 688-89 n. 6.

. Otherwise, we would be creating a perverse incentive to leak immunized testimony to the press in hope that someone else would act upon such information.

. See maj. op. at 687-88.

. The majority mistakenly states that the civil investigation was opened in October 1985.

. Pitt's exact testimony, in relevant part, was as follows:

Direct

Q: Did you have any conversations with Agent Wagner following your trip to Baltimore and your review of these records that you've described?

A: Yes, Sir.

Q: Did you tell him — first of all, did you give him copies of the documents that you brought back?

A: No, Sir.

Q: Did you tell him what documents you had seen or copied up there?

A: My best recollection is I gave him a summary of some of the stuff I looked at. I did not give him a detailed listing or accountability [sic] as to what I reviewed.

Q: What’s your best recollection of what you told Agent Wagner that you had discovered in your trip to Baltimore in February of '88?

A: The only thing I specifically recall is we were in the hallway outside of his office approximately three-or-four days after I returned. He inquired as to how it went. I do specifically recall mentioning to him that Mr. Mozer's name had turned up in either a telephone listing or a rolodex listing from an individual who was a known crime figure out of Buffalo, New York, who had been assassinated.

Q: Do you recall telling Agent Wagner anything else you had learned in Baltimore?

*693 A: That is the only thing I can specifically recall.

Cross-Examination

Q: To pick up where the government left off for a moment, when you testified that you talked to Mr. Wagner about your review of the F.B.I. case files in Baltimore, you stated that you don't recall discussing more information with him than what you have already testified to, correct?

A: Yes, sir.

Q: Now, you're not testifying that you didn't discuss more information. That you just don't recall discussing more information, is that correct?

A: That is correct.

Q: So you can't actually testify as to what information he received from you of your review of the F.B.I. files in Baltimore?

A: Within limitations I can, yes, sir.

. To say nothing of the fact that both Pitt and Benner knew that Kilroy had actually.confessed in Las Vegas to embezzling funds from NSCC before they made the formal decision to investigate Kilroy.

. The majority — by approving the district court’s finding that no taint occurred — rejects, without explanation, the government's concession that Wagner had been at least partially tainted by Pitt prior to the former's 1990 grand jury appearances.