United States v. North, 920 F.2d 940 (D.C. Cir. 1990)

ORDER

PER CURIAM.

On consideration of appellee’s petition for rehearing filed September 4, 1990 and appellant’s opposition thereto, it is

ORDERED, by the Court, that the petition for rehearing be granted in part and denied in part, and that Section II of our prior opinion be withdrawn and replaced by Section IV of the opinion filed herein, and point (2) of the Summary relating thereto be withdrawn and replaced by the following: “(2) The District Court’s jury instructions on Count 9 did not, taken as a whole, pose a genuine risk that the jury would be confused.” The reasons are set forth more fully in the opinion of the court filed herein this date.

Opinion dissenting as to Parts I, II & III filed by Chief Judge WALD.

PER CURIAM:

In its petition for rehearing, the Independent Counsel (“IC”) has raised several new issues regarding our original disposition. As we explain below, we believe that all but one of the IC’s claims lack merit. We therefore grant in part and deny in part the petition for rehearing and modify our original opinion, 910 F.2d 843, accordingly.

I. Immunized Testimony at Trial

The IC claims that we misapplied United States v. Rinaldi, 808 F.2d 1579 (D.C.Cir.1987), in remanding “for a massive inquiry into ‘the taint of the testimony and the derivation of the testimony.’ ” Petition for Rehearing at 7-8 (“Pet. for Reh’g”) (quoting Majority Opinion 910 F.2d at 866 (“Maj. Op.”) (emphasis in original)). The IC’s argument rests on the ipse dixit that “the prosecution’s freedom from taint establishes that its evidence was necessarily derived independently” and therefore that the inquiry mandated by Rinaldi would be “superfluous.” Pet. for Reh’g at 8. This bold proposition, however, would convert Kastigar’s total prohibition on use, Kastigar v. United States, 406 U.S. 441, 453, 92 S.Ct. 1653, 1661, 32 L.Ed.2d 212 (1973), to a mere ban on significant prosecutorial expo

sure to the immunized testimony. It simply does not follow that insulating prosecutors from exposure automatically proves that immunized testimony was not used against the defendant. Kastigar is instead violated whenever the prosecution puts on a witness whose testimony is shaped, directly or indirectly, by compelled testimony, regardless of how or by whom he was exposed to that compelled testimony. Were the rule otherwise, a private lawyer for a witness sympathetic to the government could listen to the compelled testimony and use it to prepare the witness for trial. The government would presumably thereby gain the advantage of use of the immunized testimony so long as it did not actually cooperate in that effort. This interpretation of Kastigar (“Look ma, no hands”) pressed by the IC, if accepted, would enormously increase the risk of providing immunized testimony. To reject it, it is unnecessary to decide whether, as North asserts, particular significance should be placed on the fact that other government personnel in the legislative and executive branches outside the Independent Counsel’s office were, after exposure to immunized testimony, actively involved in preparing witnesses.

Indeed, Rinaldi explicitly recognizes that witnesses’ exposure to immunized testimony can taint their trial testimony irrespective of the prosecution’s role in the exposure and that an inquiry is therefore necessary into whether the content of witnesses’ testimony was derived from or motivated by the immunized testimony. It specifically mandates an inquiry into what a witness knew prior to exposure to the immunized testimony and what information she gleaned from that exposure: “[tjhere is accordingly a question of fact as to how much Reardon [the witness] knew prior to that meeting [at which she was exposed to the immunized testimony] and what additional knowledge, if any, she may have gained as a result of listening to Rinaldi’s testimony.” 808 F.2d at 1583. And even where the witness testifies from personal knowledge, use within the meaning of Kastigar may occur, as Rinaldi points out, if the immunized testimony influenced the witness’ decision to testify. Rather than recognizing, as the IC claims, that a Kastigar hearing may not focus on the “psychological processes of witnesses,” Pet. for Reh’g at 7 n. 4, Rinaldi directed inquiry into whether a witness’ testimony “was motivated by, and therefore indirectly derived from, Rinaldi’s immunized statements to the police.” 808 F.2d at 1584 n. 7 (emphasis added).1 Our opinion is thus entirely consistent with Rinaldi in calling for an inquiry on remand into the content and circumstances of witnesses’ testimony.

Our dissenting colleague does not disagree with us on this central point so vigorously disputed by the IC — that the content and circumstances of testimony given by a witness exposed to the defendant’s immunized testimony may constitute “use” of the immunized testimony in violation of a defendant’s constitutional rights regardless of the prosecutor’s “fault.” But she does contend that we have extended Rinaldi by insisting that the testimony of any witness exposed to the immunized testimony be “pre-recorded” in much the same way as prosecutors memorialize their investigative material, including witnesses’ statements, so as to be able to prove in a Kastigar hearing that the government has obtained no leads from the immunized testimony. We did not, however, set forth such a requirement; we only said this burden “may” be met by “cannpng]” the testimony beforehand, Maj.Op. at 872-873, just as wise prosecutors meet their burden of showing independent investigation by “canning” the results of the investigation before the defendant gives immunized testi

mony.2

To be sure, if such steps are not taken, it may well be extremely difficult for the prosecutor to sustain its burden of proof that a witness exposed to immunized testimony has not shaped his or her testimony in light of the exposure, or as the Rinaldi court observed, been motivated to come forward and testify in light of the immunized testimony. But we surely did not mean to preclude the use of any techniques of which we are not aware, nor did we mean to even suggest that the prosecutor was barred from trying to show in any fashion that a witness’ testimony was not influenced by the immunized testimony.3

What we did insist upon, however— and here we quite definitely part company with the Chief Judge — is that the prosecutor has to prove that witnesses who testified against the defendant did not draw upon the immunized testimony to use it against the defendant; the burden of disproving use cannot, under Kastigar, be shifted onto the defendant, nor can the defendant be required to assume the burden of going forward with evidence that puts in issue the question of use. Most important, the defendant is entitled to a hearing at which he would be able to challenge the prosecution’s case for non-use.

Although our dissenting colleague expresses concern over the special institutional interests of the Independent Counsel, she does not suggest, nor could it plausibly be suggested, that the defendant’s constitutional rights are somehow lessened because he was prosecuted by the Independent Counsel rather than a United States attorney or the criminal division of the Justice Department. Therefore, the dissent’s rationalization (truly post hoc) of the district judge’s refusal to put the IC to its proof at a hearing must apply to any case in which the defendant gave immunized public testimony to a congressional committee. The dissent’s focus is on executive branch misconduct, but congressional committees have held highly publicized hearings into a number of other perceived problems, like the influence of organized crime, corruption in the labor movement, dubious practices on Wall Street, and even, as some will recall, communist influence in Hollywood. We cannot imagine what subjects will be of public and thus congressional interest in the future. We must assume that any American could be compelled to testify in return for use immunity under 18 U.S.C. § 6005 (1988), which authorizes Congress to grant that immunity, even over protests of the prosecutor, independent or otherwise. And thus the novel approach suggested by our colleague would apply to any person who finds him or herself in the position of being prosecuted after having testified publicly before Congress.

The dissent argues that the district judge was not obliged to hold a hearing to determine whether or not the witnesses changed or shaped their testimony (or, for that matter, offered to testify) because of the immunized testimony. It is not because the question is irrelevant, according to the dissent, but rather because the IC produced something like “a prima facie case” that the witnesses’ testimony was not tainted. Dissent on Pet. for Reh’g at 954. We are rather puzzled because we do not perceive that the IC put on any evidence whatsoever

directed to this issue. Indeed, the district judge (as did our colleague) relied on an in-chambers review, a sort of “self-directed ... inquiry,” Carducci v. Regan, 714 F.2d 171, 177 (D.C.Cir.1983), to satisfy himself that there was no taint. Of course, this “review” neatly avoided any cross-examination of witnesses who were admittedly exposed to the immunized testimony (which, coupled with the switch in testimony of one key witness, may be at the core of this dispute). Even if such a proceeding could substitute for an adversary hearing (which we think quite a remarkable proposition), it could not possibly do so in this case because the district judge believed that independent exposure can never violate Kasti-gar. See United States v. Poindexter, 698 F.Supp. 300, 307, 313-14 (D.D.C.1988). Therefore, his “preliminary” Kastigar finding, which the dissent describes as based on “a thorough reading of preliminary witness interviews and the grand jury record,” Dissent on Pet. for Reh’g at 955, seems not only inadequate, but quite beside the point as well.

If the prosecutor were to demonstrate through testimony that a particular witness exposed to the immunized testimony had not been affected by the exposure, for example, by showing that the witness had set down his story before exposure, then the burden of going forward would shift to the defendant to challenge that version. But the dissent constructs the “prima facie case” out of whole cloth by relying merely on the district court’s admonition to witnesses to testify based on personal knowledge,4 as if that satisfied the prosecutor’s Kastigar burden.

The IC (and the district court) obviously wished to avoid cross-examination of the exposed witnesses. Some might convincingly testify that their exposure had no effect on their trial or grand jury testimony. Others might well testify that they simply were unable to determine just how much exposure affected their testimony, in which case that uncertainty would surely be a grave problem for the party with the burden of proof — the prosecutor. In this case, however, we know that at least one crucial witness changed his testimony before Congress after hearing the immunized testimony, and then presumably told the modified version to the trial jury. If by some alchemy the burden of going forward at the outset were to be shifted to the defendant (which we think plainly impermissible), we find it impossible to understand why that undisputed showing alone would not satisfy that burden. Why does it not meet even our dissenting colleague’s requirement for a contradiction between testimony prior to exposure and testimony subsequent to exposure?5

When all is said and done, the district judge denied the defendant a hearing to which the Constitution entitled him. The Chief Judge’s new formulation of constitutional law does appear to justify that denial, but we think it does so only by subordinating the defendant’s constitutional rights to various other interests.

The premise of the Chief Judge’s position is that the federal courts must treat the issue of witness exposure quite differently from prosecution staff exposure because a federal prosecutor can control only the latter. We think it is a mistaken premise because, as we have noted, it does not

matter to a defendant if his immunized testimony is used against him by a key witness (and his lawyer) rather than by the prosecutor. The damage to the defendant — which is the focus of Kastigar — is the same in either case.

In any event, although it is not an insignificant problem, we think the dissent exaggerates a prosecutor’s difficulties. We can think of a number of ways that a federal prosecutor could seek to prevent exposure of witnesses he or she may intend to use, and even more ways to memorialize a witness’ testimony before exposure. The dissent is apparently concerned about a witness hostile to the prosecution who deliberately exposes himself to the immunized testimony in order to destroy the value of his testimony. There is, of course, not a shred of evidence that the IC encountered this problem in this case. Indeed, only the Chief Judge raises this problem — not the IC and not the district judge. If a prosecutor had such a concern about a particular witness, that witness would be at the very top of the list of those whose testimony should be prerecorded. And if at a Kasti-gar hearing such a witness should seek to undermine the case, the prosecutor would be entitled to bring out such a motivation in his examination of the witness.

Our dissenting colleague points also to the particular problem faced by an Independent Counsel whose investigation is targeted against one or more executive branch officials. She suggests that other officials in the executive branch who could be called as witnesses might seek to frustrate the conviction of a target by exposing themselves to immunized testimony. That is actually less of a problem than it would be for an ordinary United States attorney targeting a non-government official. Government officials are subject to greater restraints on their behavior than private individuals. The Ethics in Government Act requires the Department of Justice to cooperate with an Independent Counsel. See 28 U.S.C. § 594(d) (1988). Other executive departments are expected to cooperate with the Department of Justice, the chief law enforcement arm of the executive. Moreover, the IC presumably has the power to bring charges of obstruction of justice against anyone who attempts to sabotage the investigation. We are not aware that the Independent Counsel made any efforts to prevent government officials who were to testify or who had already testified from exposing themselves to immunized testimony. The IC’s position, unlike the dissent’s, seems to have been that the matter was irrelevant or perhaps, after the exposure of the key witness and his modification of his testimony, that the horse was out of the barn.

Finally, and perhaps at the heart of the dissent’s concerns, is the argument that a straightforward application of Kastigar in cases where a witness testifies before Congress, after Congress grants immunity under section 6005, unduly restricts Congress’ role in exposing wrongdoing in the nation — including wrongdoing in the executive branch'. She even contends that witnesses who wished to frustrate prosecutors would “line up to testify before Congress[ ] in exchange for ... immunity.” Dissent on Pet. for Reh’g at 953. We do not think Congress would be so naive as lightly to grant use immunity to such prospective defendants. Surely Congress does so only when its perception of the national interest justifies this extraordinary step. When Congress grants immunity before the prosecution has completed preparing its “case,” the prosecutor, whoever that may be, can warn that the grant of immunity has its institutional costs; in this case, the IC indeed warned Congress that “any grant of use and derivative use immunity would create serious — and perhaps insurmountable— barriers to the prosecution of the immunized witness.” Memorandum of the Independent Counsel Concerning Use Immunity 1 (Jan. 13, 1987) (Submitted to the Joint Congressional Iran/Contra Committees) (J.A. at 2502). The decision as to whether the national interest justifies that institutional cost in the enforcement of the criminal laws is, of course, a political one to be made by Congress. Once made, however, that cost cannot be paid in the coin of a defendant’s constitutional rights. That is simply not the way our system works. The

political needs of the majority, or Congress, or the President never, never, never, should trump an individual’s explicit constitutional protections. Indeed, the government may not even use immunized testimony to prevent a defendant’s subsequent perjury. See New Jersey v. Portash, 440 U.S. 450, 459, 99 S.Ct. 1292, 1297, 59 L.Ed.2d 501 (1979) (“Testimony given in response to a grant of legislative immunity is the essence of coerced testimony.... [In the case of immunized testimony,] we deal with the constitutional privilege against compulsory self-incrimination in its most pristine form. Balancing, therefore, is not simply unnecessary. It is impermissible.”).6

The Chief Judge tentatively suggests a post hoe doctrinal justification for the district court’s breach of the Kastigar “any use” wall, but, in doing so, we think she muddles two distinct concepts: (1) how one traces a causal connection between immunized testimony and grand jury or trial testimony to determine whether there is in fact “use,” and (2) who, if anyone, may “use” immunized testimony against a defendant. Of course, a ban on any use of immunized testimony requires the prosecution to prove that proffered evidence, which does not appear on its face to be drawn from immunized testimony, is not “directly or indirectly derived from such testimony.” 18 U.S.C. § 6002 (1988) (emphasis added). And that process also, of course, is focused on the presence or absence of causal links between the immunized testimony and the proffered testimony. That is a factual inquiry; indeed, it is the central purpose of a Kastigar hearing. In our case, for instance, the district judge might find that a particular witness’ exposure to immunized testimony was so peripheral — let us say he only saw a newspaper headline — that the judge could safely find that the witness’ testimony was not tainted by the exposure, and therefore there is, in that situation, no use of the immunized testimony against the defendant. It is in that context that prior Fourth Amendment and Fifth Amendment eases on causation discussed under the rubric “fruit of a poisonous tree” are arguably relevant.7 As should be quite apparent, however, the exploration of these causation issues demands a Kastigar hearing. Instead, the dissent asks us to permit the district court to wave a wand of benedic

tion over boxes of documents and proclaim a factual finding. But absolutely nothing in the Chief Judge’s discussion of Fourth and Fifth Amendment causation cases provides support for assuming non-causation as a matter of law or for affirming on appeal findings made or implied without a hearing, on the mere supposition that these actual or implicit findings must have been based on the totality of the evidence before the district judge.

In any event, causation is not the central issue in this case, and therein lies the Chief Judge’s muddle. It would seem that witnesses who appeared before the grand jury and trial jury actually studied North’s testimony. It is hard to imagine a more direct use of immunized testimony than in such a situation, and therefore the dissent’s use of the phrase “attenuated use” seems quite inappropriate. We take it that the Chief Judge means to imply that such direct use is somehow “attenuated” if it does not bear the prosecutor’s fingerprints, but that is simply another way of putting her main argument, that Kastigar’s protection can be easily evaded so long as an actor, independent of the prosecutor, does the damage to the defendant. None of the cases cited and discussed by the Chief Judge provides a shred of support for that proposition.

II. Immunized Testimony Before the Grand Jury

The IC renews its argument that presentation of immunized testimony to the grand jury is permissible and that no inquiry into whether the grand jury considered evi-denee based upon North’s congressional testimony is therefore appropriate. The IC cites a long line of cases from both the Supreme Court and this court as having decided that a “defendant who has been compelled to incriminate himself” is entitled only to have that evidence suppressed at trial and not to have it excluded from consideration by the grand jury. Pet. for Reh’g at 10 & n. 9. But the IC fails to note that we have previously held that the grand jury may not consider immunized testimony or evidence derived from it. See Rinaldi, 808 F.2d at 1582-83. We therefore adhere to our original disposition; however, because neither our original opinion nor our opinion in Rinaldi discussed the line of authority now relied upon by the IC, we offer the following additional thoughts.

Under the IC’s view, these cases stand for the proposition that a prosecutor could, consistent with Kastigar, read compelled testimony verbatim to the grand jury. All of the cases the IC cites, however, involve the presentation to the grand jury of evidence previously obtained in violation of a defendant’s privilege against self-incrimination;8 none involves grand jury consideration of testimony obtained from the defendant through a grant of immunity. The IC thus continues to miss the fundamental distinction between the presentation to the grand jury of evidence that has previously been unconstitutionally obtained and that of constitutionally-obtained evidence whose exposure to the grand jury amounts to a constitutional violation in and of itself.

As we recently recognized, “[gjrand jury inquiries grounded on information obtained in violation of a constitutional provision do not themselves work an additional wrong; they ‘are only a derivative use of the product of a past unlawful [action].’ ” In re Sealed Case, 877 F.2d 976, 982 (D.C.Cir.1989) (quoting United States v. Calandra, 414 U.S. 338, 354, 94 S.Ct. 613, 623, 38 L.Ed.2d 561 (1974)). And “[w]hether such derivative use of illegally obtained evidence by a grand jury should be proscribed presents a question, not of rights, but of remedies.” Calandra, 414 U.S. at 354, 94 S.Ct. at 623. By contrast, the grand jury and the grand jury process may never “itself violate a valid privilege”; such action presents a question not of remedies but of rights. Id. at 346, 94 S.Ct. at 619 (“Although ... an indictment based on evidence obtained in violation of defendant’s Fifth Amendment privilege is nevertheless valid [citation omitted], the grand jury may not force a witness to answer questions in violation of that constitutional guarantee.”).

We believe — as do five of our sister circuits, see Maj.Op. at 869-870 — that grand jury consideration of evidence already unconstitutionally compelled and grand jury consideration of immunized testimony fall on different sides of this fence. In the former context, such as where Miranda warnings are not given, the constitutional violation occurs independent of the grand jury. Whether the resulting confession can be used derivatively by the grand jury or in subsequent proceedings is a matter of the reach of the exclusionary rule, which is a function not of any rights of the defendant but of a remedial balance factoring in the possible unreliability of the confession and the need to deter the government from future violations of Fifth Amendment rights. See, e.g., Oregon v. Elstad, 470 U.S. 298, 304-08, 105 S.Ct. 1285, 1290-93, 84 L.Ed.2d 222 (1985); New York v. Quarles, 467 U.S. 649, 654, 104 S.Ct. 2626, 2630, 81 L.Ed.2d 550 (1984); Michigan v. Tucker, 417 U.S. 433, 445-47, 94 S.Ct. 2357, 2364-65, 41 L.Ed.2d 182 (1974). The exclusionary rule is, of course, by no means talismanic — where the goals of trustworthiness and deterrence are not implicated, the confession may be used. For example, a confession obtained in violation of Miranda may (provided it is trustworthy) be used to impeach the defendant’s contrary testimony at trial because “sufficient deterrence flows when the evidence in question is made unavailable to the prosecution in its case in chief.” Harris v. New York, 401 U.S. 222, 224-25, 91 S.Ct. 643, 645-46, 28 L.Ed.2d 1 (1971). The cases cited by the IC thus reflect only the familiar proposition that, while evidence obtained previously in violation of the defendant’s Fifth Amendment privilege may not be used by the prosecution in its case in chief at trial, it may be used for other purposes, including grand jury proceedings.

In contrast, as the Supreme Court has recognized, use of immunized testimony is never justified by such balancing considerations. See Portash, 440 U.S. at 459, 99 S.Ct. at 1297. The prosecution obtains the immunized testimony legally, but only by promising that neither the testimony or information itself nor any information directly or indirectly derived from it will “be used against [the defendant] in any criminal case.” 18 U.S.C. § 6002. And it is only this promise that compels the defendant to testify in spite of his constitutional privilege: “immunity from use and derivative use is coextensive with the scope of the privilege against self-incrimination.... [because i]t prohibits the prosecutorial authorities from using the compelled testimony in any respect.” Kastigar, 406 U.S. at 453, 92 S.Ct. at 1661 (emphasis in original). When the prosecution reneges on this constitutionally-mandated bargain and presents the immunized testimony to the grand jury, the constitutional violation is part and parcel of the grand jury process. The presentation — “use”—of the testimony is precisely the proscribed act. The issue is thus not one of "derivative use ... by the grand jury” and of the exclusionary rule (indeed, any use of the testimony is per se excluded under the statute and Kastigar). Rather, the situation is no different than if the grand jury had itself forced the defen

dant to give incriminating answers and any indictment based upon immunized evidence is no less tainted.9

III. Count 9 Authorization Instructions

The IC is barred from arguing that we erred in construing 18 U.S.C. § 2071(b) to require that a defendant possess “knowledge of unlawfulness.” Maj.Op. at 886. Despite its current protestations to the contrary, the IC never made, and thus waived, this argument. At no point in its brief did the IC dispute North’s contention that “knowledge of unlawfulness” is a required element of this statute. North's reply brief emphasized that knowledge of unlawfulness meant actual knowledge, citing two cases in support — including United States v. Cullen, 454 F.2d 386 (7th Cir.1971), on which the IC now relies.10 As a result, at oral argument both the court and North’s counsel understood the question to be uncontested. The following dialogue took place during the opening argument of North’s counsel.

Tr. at 37. If the IC had remained silent on this issue at oral argument we would surely have been entitled to assume that it had waived the contention it now raises, but the IC, in response, expressly conceded the point.

Tr. at 73 (emphasis added).11

The revisionist nature of the IC’s current claim is further illustrated by its new found reliance in its petition for rehearing on dictum in Cullen, 454 F.2d at 392, which, according to the IC, conflicts with our view of the requisite mens rea. The IC did not cite Cullen in its brief or mention it at argument, as it surely would have done had it meant to raise the point it now argues. Thus, the IC never asserted in these extensive proceedings — until its petition for rehearing — that the test of 18 U.S.C. § 2071(b), the statutory section alleged to be violated in Count 9, could be satisfied by the government if it showed that a defendant should have known, in accordance with an objective test, that his conduct was unlawful. Indeed, Chief Judge Wald, who dissented from our origi

nal opinion, did not then dispute our characterization of the IC’s argument as having agreed (or conceded) that actual knowledge of unlawfulness was required under Count 9. Compare Maj.Op. at 884-886 with Diss.Op. at 929, 930-932.

To be sure, Judge Ruth B. Ginsburg, dissenting from the denial of rehearing en banc on this Count alone, asserts that, notwithstanding all of the above, the IC did preserve in its brief the argument it now makes. She quotes from page 39 of the IC’s brief: “But an unreasonable belief that an order to lie to Congress or alter official records made such conduct legal no more negates guilt than would an unreasonable belief that the conduct itself had not been prohibited.” The difficulty with this sentence, however, which is not specifically directed at Count 9, is that the same conduct was also alleged in Count 6 as violating § 1505 and the IC did argue that it was not obliged to prove knowledge of unlawfulness as to Count 6. At no point in the IC’s brief does the IC claim that it was not required to prove actual knowledge of unlawfulness to make out the violation of § 2071, charged in Count 9. The closest the IC comes to that argument is its carefully crafted footnote 73 in which it claims that “North cites no case holding that the state of mind required for violation of either § 1505 or § 2071 is precluded by an unreasonable belief that the conduct was authorized.” We note that the sentence purposefully does not affirmatively argue the contrary. Authorization, moreover, is not necessarily the equivalent of lawfulness. The petition for rehearing clearly makes the new and broader argument that even if North believed (reasonably or not) that his conduct was authorized, he can still be convicted under Count 9 — requiring the government to prove knowledge of unlawfulness — if his reliance on authorization to avoid illegality was unreasonable. The IC’s brief avoided any analysis of the mental state required under § 2071; footnote 73 glosses over that issue, eliding it with the mental state required by § 1505.

In short, Judge Ruth B. Ginsburg’s differing reading notwithstanding, it appears to us that for whatever reason (perhaps the lack of any supporting authority), the IC deliberately chose not to make the argument it now raises for the first time. We therefore believe that the IC’s contention upon which it rests its petition for rehearing — that the panel “[erroneously claim[ed] that the Government agreed that ... 18 U.S.C. § 2071(b) ‘requires subjective knowledge of unlawfulness for conviction,’ ” Pet. for Reh’g at 12 (quoting Maj.Op. at 886) — is a misstatement. In any event, we cannot imagine that this issue is en banc worthy in accordance with any of the principles that govern such determinations. We disagree with Judge Ruth B. Ginsburg’s characterization of our reasoning in the case that was before us, which, after all, relies in part on a prior decision of this court. See United States v. Rhone, 864 F.2d 832, 835 (D.C.Cir.1989). The petition for rehearing, however, is based, not on a disagreement with the reasoning we employed in the case we decided, but rather on the new theory, previously disavowed, presented in the petition. Although we decline to entertain this revisionist argument at this late stage in the proceeding (whatever its merit or lack thereof), we do not see how the government (or the IC) could be barred from raising the argument in another case. So, on the assumption that the IC should not be treated any differently from other litigants before the court, we do not understand how it can be thought that its new claim is en banc worthy.

IV. Jury Unanimity Instruction

As to the IC’s final point — that no specific unanimity instruction was required as to Count 9 — upon reviewing the petition, the response of North, and the instructions of the district court to the jury, we conclude that this point is well taken. This Count of the indictment “charged that North, having custody of NSC documents, ‘willfully and knowingly did conceal, remove, mutilate, obliterate, falsify and destroy and did cause to be concealed, removed, mutilated, obliterated, falsified and destroyed records, papers and documents

filed and deposited in a public office....”’ United States v. North, Maj.Op. at 876, quoting Joint Appendix at 260-61. As originally conceived, Count 9 concerned North’s handling of at least three groups of documents in several distinct incidents. At the time of the issuance of our original opinion, both the majority opinion (Maj.Op. at 876-877) and the dissent (Dissent at 925) failed to note that the district court had instructed the jury to limit its consideration to a single incident involving five specific documents. Therefore, each opinion dealt with the question as if all documents and incidents originally charged in the indictment were still before the jury.

The IC in his motion for reconsideration called our attention to the trial court’s limiting instruction. We now review the instruction including the limitation under the principle that these instructions must be considered “as a whole.” United States v. Mangieri, 694 F.2d 1270, 1280 (D.C.Cir.1982). Having done so, we conclude that the instructions on Count 9 as a whole did not pose “a genuine risk that the jury [would be] confused.” United States v. Duncan, 850 F.2d 1104, 1114 (6th Cir.1988), cert. denied, _ U.S. _, 110 S.Ct. 732, 107 L.Ed.2d 751 (1990). As the IC argues in his petition for rehearing, “[t]he evidence was identical as to all five documents ... [t]hus, the jury was restricted to one continuous course of events. United States v. North, 716 F.Supp. 644, 649 (D.D.C.1989).” Thus, the omission of a specific unanimity instruction did not constitute reversible error. We will therefore allow the petition of the Independent Counsel to this limited extent, and withdraw Section II of our prior opinion herein.

WALD, Chief Judge,

dissenting as to Parts I, II, & III:

In his petition for rehearing, the Independent Counsel (“IC”) argues that the prohibitions of the use-immunity statute, 18 U.S.C. § 6002, do not extend to the government’s use of witnesses who have independently exposed themselves to immunized testimony. Although the claim that the statute does not cover any such witness exposure is problematical, I do agree, as my earlier dissent reflects, that the statute does not require that independent witness exposure and prosecutorial exposure be treated identically for prophylactic purposes. Since, as the IC’s petition forcefully points out, my colleagues’ original opinion effectively transformed a limited use immunity into a sweeping transaction immunity, I would grant the IC’s petition for rehearing on the Kastigar issue. By exalting form over substance, the original per curiam eviscerates both the use-immunity and independent-counsel statutes; its consequences for future cases of public import are ominous.

I. Kastigar Requirements

A. The Problem with the Original Opinion

The Supreme Court has recognized that use immunity is coextensive with the fifth amendment privilege. Accordingly, restrictions on the use of immunized testimony are exacting.

Kastigar v. United States, 406 U.S. 441, 460, 92 S.Ct. 1653, 1665, 32 L.Ed.2d 212 (1972).

This “very substantial protection,” id. at 461, 92 S.Ct. at 1665, while reflecting the importance of the constitutional values at stake, was not meant to make the prosecution’s burden an impossible one. The use-immunity statute makes a precise accommodation between the privilege against self-incrimination and the public’s legitimate interest in securing testimony; use immunity thus exists in a delicate tension with the fifth amendment.

By mandating additional — and practically unattainable — requirements not found in Kastigar itself, my colleagues have upset this tension. They have rendered impossible in virtually all cases the prosecution of persons whose immunized testimony is of such national significance as to be the subject of congressional hearings and media

coverage. In their opinions, my colleagues have ruled that Kastigar requires at least four distinct showings, only the first two of which can be derived from Kastigar itself. First, the prosecutors must demonstrate that they avoided “significant exposure” to the immunized testimony. Majority Opinion (“Maj. op.”) at 859. Second, the prosecution must demonstrate that its identification and questioning of witnesses was based solely on “independent leads” — without the use of immunized testimony. Maj. op. at 863. Third — a new requirement, appearing for the first time in the opinion denying rehearing — the prosecution must demonstrate that the immunized testimony did not “motivate[]” its witnesses to testify. Opinion on Petition for Rehearing at 942. And fourth, the prosecution must demonstrate that the testimony of witnesses “exposed” to immunized matter has been “ ‘canned’ by the prosecution before such exposure.” Maj. op. at 872.

The last and most stringent of these requirements — that witness testimony be pre-recorded — is certainly an unwarranted departure from current law. For this reason, it deserves quotation at some length.

Maj. op. at 872-873 (latter emphasis supplied). Although my colleagues now maintain that they did not intend to rule that pre-recording was the only way the prosecution could demonstrate that the testimony of a witness exposed to immunized testimony was admissible, the words of the per curiam are clear: either can the witness or can his testimony.1

My colleagues invoke United States v. Rinaldi, 808 F.2d 1579 (D.C.Cir.1987), to support that radical extension of current law. Yet Rinaldi does not even suggest that the witness’ original knowledge need be or was pre-recorded. Instead, the Ri-naldi court indicated a far more lenient rule of inevitable discovery — the prosecution need only show that “the police would inevitably have learned the [facts] from [the witness].” 808 F.2d at 1583 (emphasis supplied). As set forth in my original dissent and outlined below, I believe that Judge Gesell’s findings clearly established such inevitable discovery. See Dissenting Opinion (“Diss. op.”) at 917, 922-923.

My colleagues’ extension of Kastigar is infirm not only because, as the IC suggests, it is unrooted in existing law, but also because it effectively emasculates both use immunity and the Office of the Independent Counsel.

A uniform requirement of pre-recording witness knowledge in exquisite detail is unworkable. As even the greenest trial lawyer knows, the accrual of evidence is interactive — the statements of one witness often suggest new questions for earlier witnesses. Pre-recording of every line of every witness’ trial testimony in every prosecution in which a defendant might publicly offer immunized testimony would ultimately prove unfeasible.

The consequences of a pre-recording requirement are both predictable and troubling. Prospective targets of grand juries

in national scandals would line up to testify before Congress, in exchange for what is effectively transaction immunity. A requirement of “nonuse” would be converted into a guarantee of nonprosecution. Moreover, as the IC notes, with regard to crimes involving government corruption, the IC and Congress have parallel responsibilities. Petition at n. 1. A pre-recording requirement compels Congress to delay congressional hearings until all of the testimony of all potential witnesses has been pre-recorded; in doing so, this requirement unduly burdens Congress’ exercise of its legitimate authority to hold important public hearings.

The Supreme Court has said that use immunity “grants neither pardon nor amnesty.” Kastigar, 406 U.S. at 461, 92 S.Ct. at 1665. But the majority’s excruciatingly heightened Kastigar requirements run counter to Congress' express assertion that “the [use-immunity] provision is not an ‘immunity bath.’ ” H.Rep. No. 91-1188, 91st Cong., 2d Sess. at 12 (1970) (citing United States v. Monia, 317 U.S. 424, 63 S.Ct. 409, 87 L.Ed. 376 (1943)).

The interpretation of Kastigar offered in the original per curiam opinion also conflicts with the congressionally-bestowed responsibilities of the IC. Congress created the IC Office pursuant to its Article II authority “to enact laws to guard against the evils of massive conflicts of interest involved in the enforcement of federal criminal law against the highest officials of government.” In re Olson, 818 F.2d 34, 43 (D.C.Cir.Indep.Couns.Div.1987) (emphasis in original). Congress acted on the basis of

818 F.2d at 42. Accordingly, the IC Office serves a critical role in our constitutional system of checks and balances. See generally Morrison v. Olson, 487 U.S. 654, 108 S.Ct. 2597, 101 L.Ed.2d 569 (1988). It safeguards the primary precondition for government under law: that no one sit in judgment of his or her own case.

My colleagues’ extension of Kastigar thwarts Congress’ central purpose in creating the Office of the IC. Congress required the appointed counsel to be truly independent and its authority exclusive; Congress thus required that “the Department of Justice, the Attorney General, and all other officers and employees of the Department ... suspend all investigations and proceedings” on matters within the IC’s jurisdiction. 28 U.S.C. § 597. Yet in its expansive interpretation of Kastigar, the majority provides an easy out for conflicted government officials; they can immunize their colleagues from prosecution by exposing themselves to immunized testimony. In this case it was insiders — high government officials — who “soaked” themselves in North’s immunized testimony. To raise an insurmountable bar against the admission of the testimony of these officials, however necessary it might be to the prosecution, critically undermines the authority and independence of the IC in cases of self-defined conflict on the part of government officials.

The majority is correct in noting that, if Kastigar is read only to apply to exposure of prosecutors, then “a private lawyer for a witness sympathetic to the government could listen to the compelled testimony and use it to prepare the witness.” Opinion on Petition for Rehearing at 942. But it is also true that if Kastigar is read to require pre-recording of all government-witness testimony, then a witness hostile to the government could “listen to the compelled testimony and use it” to insulate himself from testifying.

There must be a middle ground, and I believe the case should be reheard (by the panel or en banc) to find it. The significance of these issues for the prosecution of future governmental scandals and for the effective functioning of separation of powers is too great to let the overreach of the

original opinion (or my colleagues’ undefined backtracking) stand.

B. Prosecutorial Exposure v. Independent Witness Exposure

No prior decision has articulated the proceedings Kastigar requires in cases involving widely disseminated immunized testimony. The trial judge here labored in uncharted territory. In cases like North’s several goals and values must be accommodated: the immunized person’s fifth amendment privilege; Congress’ purposes in enacting the use-immunity statute; the function and integrity of the Office of the IC; and the government’s need for workable guidelines.

As the IC’s petition highlights, the central conceptual weakness of the majority’s analysis is its failure to recognize that a prosecutor’s exposure to immunized testimony and a witness’ independent exposure to such testimony raise related but distinct issues. By holding to the same standard witnesses who have “thoroughly soaked” themselves in immunized testimony and prosecutors who have assiduously avoided the slightest taint, the majority renders virtually impossible any prosecution of an immunized defendant who testifies publicly-

Prosecutorial use of immunized testimony is the paradigmatic violation of the use-immunity statute and prosecutorial exposure is, thus, at the heart of use immunity. Indeed, as the IC’s petition points out, virtually every prior ease concerning use immunity involved prosecutorial knowledge of immunized testimony. We all agree that prosecutors must avoid “significant exposure” to immunized testimony, develop independent leads to witnesses, and refrain from using immunized testimony to lead or coach those witnesses. In this regard, the prosecution bears the continuous and uninterrupted burden of persuasion, and in this case, no one disputes that the IC has met that burden. Maj. op. at 859; Diss. op. at 921.

Independent witness exposure poses a different problem. Despite the prosecution’s diligent, and often extraordinary, prophylactic measures, witnesses may, for a variety of reasons, “soak” themselves or be soaked by others in immunized testimony. For such witnesses, the prosecutor’s burden to show nonuse should arguably be quite different. Initially, the prosecutor must establish a prima facie case that a witness’ testimony does not constitute a prohibited “use.” This initial burden should, however, be met, as it was in the North case, by a five-factored showing:

Collectively, these requirements should be sufficient to establish a prima facie case that any testimony offered at trial by the IC is untainted.

Upon establishment of a prima facie case, the defense then bears a burden to produce some specific evidence that the testimony — either in source or content — is tainted; upon such production, the court is required to hold a hearing on the alleged taint. The defense need only produce specific, not conclusive evidence, for as Hasti-

gar suggests, the prosecution always bears the burden of ultimate persuasion. Specific evidence might include a pattern of recollection or specificity that suggests taint, or contradictions between recorded and unrecorded testimony. As the Supreme Court stated in another allocation-of-burdens context, “[i]t is sufficient [that] the defendant’s evidence raises a genuine issue of fact” as to whether the witness’ testimony is tainted. Texas Department of Community Affairs v. Burdine, 450 U.S. 248, 254, 101 S.Ct. 1089, 1094, 67 L.Ed.2d 207 (1981).

This allocation of burdens would be one way to reconcile the several constitutional values and statutory commands implicated by prosecutions like North’s. As the Supreme Court has noted

Keyes v. School District No. 1, 413 U.S. 189, 209, 93 S.Ct. 2686, 2698, 37 L.Ed.2d 548 (1973) (citations omitted).

Whether or not the best resolution of these values lies in burden shifting, the important thing is that the use-immunity statute be construed so as to accommodate the real difference between prosecutorial exposure and use — the central danger posed by use immunity — and witness exposure, which is more diffuse and beyond the immediate control of the prosecution. Respecting this critical difference, this court should seek to provide a workable and attainable standard that guarantees the integrity of use immunity while allowing for a cautious and untainted prosecution.

C. Kastigar Proceedings in This Case

My earlier dissent cataloged the thorough, if tedious, Kastigar procedures employed by the trial court. Diss. op. at 914-917. These efforts — unrebutted by any showing of taint by the defense — ensured that there was no “use” of North’s immunized testimony and thus no reversible error.

Based on knowledge of the preliminary interviews and grand jury testimony, the court was able to determine that no tainted testimony was introduced at trial.

Taken together, all of these procedural protections and the continued diligence of the trial court made clear that no testimony was tainted by North’s immunized testimony and thus that no illegal “use” of immunized testimony was made. The majority’s remand for a “line-by-line” re-examination of the source (and perhaps the motivation) of the testimony of every prosecution witness and its mandate that the prosecution prove — by pre-testimony canning or some undefined functional equivalent — that all testimony was untainted represents an unneeded and unprecedented incursion into the trial court’s discretion in managing a fair trial.5

II. Constitutional Questions

To hear my colleagues tell it, the Kasti-gar analysis I have outlined “lessen[s]” and “subordinares North’s] constitutional rights.” The majority reasons that because use immunity is “coextensive” with the scope of the fifth amendment privilege, Kastigar, 406 U.S. at 453, 92 S.Ct. at 1661, any violation of the immunity statute is a violation of the Constitution. Assuming this is correct, and that it is the Constitution we are expounding, then we should begin by looking at relevant interpretations of what the fifth amendment requires.

Although in theory and rhetoric the fifth amendment privilege against self-incrimination is often cast as absolute, the bar on the admission of evidence acquired pursuant to6 a fifth amendment violation is less than absolute.7 Even though a confession is coerced, evidence secured pursuant to that confession will only be suppressed if it is deemed to be “fruit of the poisonous tree.” Harrison v. United States, 392 U.S. 219, 222-24, 88 S.Ct. 2008, 2010-11, 20 L.Ed.2d 1047 (1968); cf. Wong Sun v. United States, 371 U.S. 471, 83 S.Ct. 407, 9 L.Ed.2d 441 (1963). “The exclusionary rule ... does not apply if the causal connection between the information obtained illegally and the proof offered by the prosecution has ‘become so attenuated as to dissipate the taint.’ ” United States v. Davis, 617 F.2d 677, 688 (D.C.Cir.1979) (quoting Nardone v. United States, 308 U.S. 338, 341, 60 S.Ct. 266, 268, 84 L.Ed. 307 (1939)), cert. denied, 445 U.S. 967, 100 S.Ct. 1659, 64 L.Ed.2d 244 (1980).

For example, the Supreme Court has recognized that a coerced first confession inevitably influences a second confession, but has ruled that the later confession is admissible if there is a “ ‘break in the stream of events ... sufficient to insulate’ ” the sec

ond confession from the first.8 Darwin v. Connecticut, 391 U.S. 346, 349, 88 S.Ct. 1488, 1490, 20 L.Ed.2d 630 (1968) (quoting Clewis v. Texas, 386 U.S. 707, 710, 87 S.Ct. 1338, 1340, 18 L.Ed.2d 423 (1967)). In such rulings, the Court has not held that successive confessions are not causally related; indeed the Court has expressly indicated its recognition that they are.

United States v. Bayer, 331 U.S. 532, 540-41, 67 S.Ct. 1394, 1398-99, 91 L.Ed. 1654 (1947); see also Davis, 617 F.2d at 689.9

If evidence connected in some fashion to a fifth amendment violation may be admissible if its connection to the violation is sufficiently remote, and use immunity is coextensive with the fifth amendment privilege, then it would stand to reason that the exclusionary mandate of the use-immunity statute may also be limited as the proffered evidence becomes more and more remote from the original immunized statement. That is, at some point the link between the initial utterance (whether a coerced confession or immunized testimony) and the ultimate evidentiary “use” becomes so attenuated as to be constitutionally irrelevant. In such an eventuality, the government’s burden of disproving taint would certainly be modified accordingly.10

Indeed this proposition is well supported in the legislative history of the use-immunity statute. Explaining § 6002’s prohibition on the use of immunized testimony or “any information directly or indirectly derived from such testimony,” Congress, citing Wong Sun, indicated that some “use” of immunized testimony may be so attenuated as to be outside the scope of the Constitution and the statute. H.Rep. No. 91-1188, 91st Cong., 2d Sess. 12 (1970). As Justice Blackmun has noted in concluding his review of the statute, “References to the ‘fruits’ doctrine are scattered throughout the legislative history.” Pillsbury Co. v. Conboy, 459 U.S. 248, 277, 103 S.Ct. 608, 624, 74 L.Ed.2d 430 (1983) (Blackmun, J., concurring) (collecting authorities).

Thus the nexus of use immunity and the fifth amendment privilege raises many problematic issues. Perhaps, at some point, testimony by exposed witnesses is so far removed from the immunized source as to no longer be “fruit of the poisonous tree.” Perhaps independent witness exposure is so tangentially related to the immunized

statement as to warrant the sort of eviden-tiary scheme I have outlined.11

I raise these points not to answer them, but to emphasize that we tread on virgin ground here. And, if as my colleagues maintain, this is constitutional territory, then there is all the more reason for the most careful consideration before we lay down absolute and intolerable burdens on the prosecution in a media-dominated age. My colleagues are certainly correct that “the political needs of the majority ...' never, never, never should trump an individual’s explicit constitutional protection.” But this is only to beg the question: How far does the fifth amendment’s “explicit constitutional protection” extend? The question is a deep and unsettled one in current constitutional law. Our answer to date is unsatisfying and cries out for thoughtful reconsideration.

When all is said and done, the issue comes down to whether the Constitution requires that the trial judge conduct a full-scale hearing as to the source of every line of testimony emanating from any witness who has been exposed to immunized testimony, wholly independent of any action by the prosecutor, and even when the defense can provide no evidence whatsoever that the witness was influenced by the exposure. I conclude that such an absolute requirement is not self-evident from past interpretations of the fifth amendment. Indeed, the plethora of arguments about the scope of the Kastigar requirements spawned in my colleagues’ North I and II opinions only underscores the uncertain legal basis of the original decision.

III. Authorization and Intent

The IC also challenges the per curiam holding that 18 U.S.C. § 2071(b) “requires subjective knowledge of unlawfulness for conviction.”12 Maj. op. at 886 (emphasis in original). As the IC suggests, it seems difficult to reconcile this requirement with the majority’s express rejection of the infamous “Nuremberg defense” that “following orders ... can transform an illegal act into a legal one.” Maj. op. at 881. Indeed, I can think of no way to harmonize the rejection of the “following-orders” defense with the claim that a jury may acquit a defendant who relied on an admittedly unreasonable belief that he was “authorized” to commit what he otherwise knew was an unlawful act. Unfortunately, this novel holding has broad implications; as the petition notes, twenty-six federal criminal statutes contain wording similar to that in 18 U.S.C. § 2071(b). Petition at 13. Under any one of them, a defendant can now be found to lack the requisite intent, even though he knew an act was otherwise unlawful, if he unreasonably relied on a vague instruction from a superior official. This cannot be the law and the majority’s ruling that it is clearly warrants rethinking.13

Conclusion

The implications of the original per cu-riam are profound and pervasive, as to

both the use-immunity and the authorization-intent issues.14 It is difficult to think of a case of more exceptional importance, or, consequently, a case more worthy of rehearing by the panel or by the court en banc.

Before WALD, Chief Judge, MIKVA, EDWARDS, RUTH B. GINSBURG, SILBERMAN, BUCKLEY, WILLIAMS, D.H. GINSBURG, SENTELLE, THOMAS, HENDERSON, and RANDOLPH, Circuit Judges.

ORDER

PER CURIAM.

Appellee’s Suggestion for Rehearing En Banc has been circulated to the full court. The taking of a vote was requested. Thereafter, a majority of the judges of the court in regular active service did not vote in favor of the suggestion. Upon consideration of the foregoing, it is

ORDERED, by the Court en banc, that the suggestion is denied.

Circuit Judges MIKVA and EDWARDS did not participate in this matter.

Chief Judge WALD would grant the suggestion for rehearing en banc on the grounds set forth in her dissent from the original panel decision and from the denial of rehearing by the panel.

Circuit Judge RUTH BADER GINSBURG would grant the suggestion for rehearing en banc with respect to the Count 9 Authorization Instructions only. Her statement is attached.

Statement of RUTH BADER GINSBURG, Circuit Judge,

dissenting from the denial of the suggestion for rehearing en banc with respect to Count 9 Authorization Instructions:

A majority of the panel held that a government official may not be convicted under 18 U.S.C. § 2071 for altering, concealing, or shredding official documents if he believed, however unreasonably, that his actions were legal. Section 2071, the panel held, requires the prosecution to show that the official had no belief, not even a patently unreasonable belief, in his actions’ legality. The panel’s conclusion does not follow obviously from the statutory terms “willfully and unlawfully.” Nor did the panel justify its interpretation on the basis of prior constructions of § 2071, the statute’s legislative history, or — unlike the dissent— compelling policy reasons. Rather, the panel’s interpretation of § 2071 relies on the Independent Counsel’s purportedly clear concession that the statute requires proof of “subjective knowledge” that the action was illegal. I see no such concession, either in the Independent Counsel’s brief or at oral argument.

North’s opening brief contended that the statute requires that “[a] defendant’s intent must be judged by a subjective standard,” citing Morissette v. United States, 342 U.S. 246, 276, 72 S.Ct. 240, 256, 96 L.Ed. 288 (1952), and United States v. Rhone, 864 F.2d 832 (D.C.Cir.1989). Brief of Appellant at 37. North concluded that the trial judge had committed reversible error by instructing the jury on Count 9 that only a reasonable belief in his actions’ proper authorization could negate North’s criminal intent under § 2071. Id. The Independent Counsel’s brief disputes this conclusion directly, arguing that “an unreasonable belief that an order to lie to Congress or alter official records made such conduct legal no more negates guilt than would an unreasonable belief that the conduct itself had not been prohibited.” Brief for Appellee at 39. The Independent Counsel notes that North cited no case interpreting § 2071 that sustains the defense position, and then directly disputes the relevance of Morissette and Rhone — North’s

support for his claim that § 2071 requires proof of subjective knowledge. Id. at n. 73. I see no basis for deriving from this concise, precisely-trained discussion any waiver of the prosecution’s right to dispute North’s contention (swiftly accepted by the panel) that § 2071 requires proof of subjective knowledge.

Nor would I extract a dispositive concession from the colloquy at oral argument on which the per curiam opinion on rehearing ultimately relies. As the transcript of oral argument indicates, the topic of discussion immediately before and directly after the Independent Counsel’s purported concession was Count 6, not Count 9. The question to which Mr. Lynch responded “Yes”— the moment at which he is said to have made his “express” concession — was a question about Count 6, not Count 9. The sole reference to the Independent Counsel’s position on “subjective knowledge” and Count 9, however, was parenthetical, placed between a statement about Count 6 and the question about Count 6.* At no point during Mr. Lynch’s presentation did he or the panel pause to focus singularly on the relation of North’s authorization evidence to Count 9. Given the number and complexity of issues Mr. Lynch had to discuss, and 'the barrage of questions he faced from the bench, the court should be loath to find a concession, either express or implied.

I agree with Chief Judge Wald that the panel’s conclusion on this issue is troubling and warrants further consideration. The panel majority has declared its ruling on the merits good for this day and case alone, and so deems the matter not worthy of en banc consideration. See per curiam opinion on rehearing at 950. I think it vital, however, to dispel the impression that, in responding to questions before the D.C. Circuit, diligent counsel are best advised to just say “No.”