dissenting:
The majority holds today that a criminal defendant’s right to call a witness in his own defense is trumped by the prospective witness’ privilege against self-incrimination even where the judge has found both that the witness’ testimony is essential to the defendant’s case and that there is no practical possibility that the witness will in fact be prosecuted. In so holding, the court overrules a line of cases which balanced the interests involved reasonably and sensitively, to the apparent satisfaction of defendants, witnesses, and prosecutors. The result of the majority’s ruling will be that, to a substantially greater degree than before, the opportunity of some defendants to present exculpatory testimony, and thus to receive a fair trial, will depend not on an objective determination by the judge, but rather on the tender mercies of a prosecutor who has previously been instructed to grant use immunity to a defense witness only in “exceptional circumstances.” In my view, the new rule will deny some defendants the opportunity to call witnesses whom they should have the right to call, and whom they could have called under prior law. As a result, innocent defendants may be found guilty. Because I am aware of no legitimate interest that will be served by this unnecessary departure from fair and equitable procedure, I respectfully dissent.
I.
IF IT AIN’T BROKE, DON’T FIX IT
A majority of this court apparently decided to “go en banc” in this case in order to overrule Jaggers v. United States, 482 A.2d 786 (D.C.1984) (per curiam), on account of that decision’s perceived doctrinal flaws. My colleagues’ dissatisfaction with prior law is purely theoretical, for I know of no practical harm that Joggers was doing to the rights of any participant in the judicial process.
Joggers was decided in 1984, but it was not an especially Orwellian or otherwise ominous case. The error which critics of Joggers perceive in that decision is supposed to be that the court construed too narrowly the Fifth Amendment rights of defense witnesses. So far as I am aware, however, there has not been a single ease, in the dozen years since Joggers came down, in which a defense witness has complained to this court that he was compelled to incriminate himself, or that he was held in criminal contempt for refusing to incriminate himself, as a result of the rule in Joggers. If any witness in this jurisdiction had been convicted of an offense on the basis of a disclosure which a court had required him or her to make because of Joggers, or if such a witness had been harmed in some other way, then the resourceful attorneys for the government, well aware that proof of an actual injury would have a good deal more juridical sex appeal than a showing of purely theoretical harm, would surely have told us about it. I therefore think it is fair to infer that the government has not identified such a case because there has been no such case.1
*353My thesis that the perceived shortcomings of the Joggers doctrine were of an entirely theoretical character, and that they did not interfere with any party’s legitimate interest, find further support in the government’s effective acquiescence in the doctrine for more than a decade. Judge Kern, dissenting in Joggers, 482 A2d at 800, and Judge Gallagher, dissenting from the division opinion in the present case, Carter v. United States, 643 A.2d 348, 361-62 (D.C.1994) (Carter I ),2 both argued that the majority in Joggers had im-permissibly ignored this court’s earlier decision in Alston v. United States, 383 A.2d 307, 312 (D.C.1978), allegedly in violation of the rule of MAP. v. Ryan, 285 A.2d 310, 312 (D.C.1971). Whether this thesis is right or wrong, it provided the government with a theory on which it could have attacked Joggers’ reasoning before a division of the court without asking the division to exceed its authority. Nevertheless, for a dozen years, the government has repeatedly assumed and accepted the correctness of the J aggers’ decision.
In the present case, in which the prosecution’s prospects of victory would obviously be substantially better under Alston alone than under Alston, Joggers, and Joggers’ progeny combined, the government never challenged the correctness of Joggers before the division. On the contrary, the government got into the act only after Judge Gallagher, dissenting in Carter I, 643 A.2d at 362, explicitly invited reconsideration of Joggers by the court sitting en banc.3 If Joggers and the eases that have followed Joggers had impaired any legitimate right of the prosecution, the government would surely have complained long ago.
The theoretical issue now before the en banc court is undoubtedly an interesting one with respect to which reasonable judges can and do disagree. Nevertheless, where, as here, a practice has been in effect for more than a decade without anyone claiming to have been injured by it, I think we should have left well enough alone, in conformity with the grammatically flawed but substantively wise slang aphorism which I have used as the heading of this portion of my dissent.
II.
THE FLAWS OF CARTER II
Unfortunately, the Joggers approach, to which no witness or other affected party has apparently objected for a dozen years, is being replaced by the new Carter II approach, to which many defendants will most assuredly object, and with good reason.4 The effect of the majority’s- decision will be to deny a defendant the opportunity to call a witness who has a theoretical Fifth Amendment problem, even if the judge is of the opinion that, in the real world, the practical possibility that the witness will be prosecuted is nil. The harm to the defendant’s rights under the Sixth Amendment is palpable; the benefit to the witness is illusory.
I recognize that the new rule does not leave the defendant without any hope at all in the kind of situation which arose here. The government may, in an abundance of generosity, grant a defense witness use immunity under the new Carter II standard. As the majority recognizes, however, the United States Attorney’s Manual itself permits such immunization of a defense witness only in “exceptional circumstances.” With a directive like that, a prosecutor cannot reasonably be expected to go out of his way very often on behalf of his adversary in a vigorously contested criminal ease.
The government has suggested a “debriefing” procedure for prospective defense wit*354nesses, and the majority finds merit in the suggestion. Given the majority’s basic disposition of the constitutional issue, I suppose that the proposal is a constructive one, and Judge Gallagher has done his best to make it as fair as it can be in a situation in which the prosecutor will be accorded such wide discretion. Realistically, however, I do not think that this procedure will work very well. There is little incentive for a witness to submit to such “debriefing,” and I suspect that most attorneys will advise their clients to pass such an opportunity by.
From the defense point of view, the government’s proposal is intrinsically objectionable because it makes the defendant’s right to secure the testimony of a witness — even of a witness who is not in any realistic danger of prosecution — contingent upon the willingness of that witness to cooperate in the debriefing procedure. Under Jaggers, if there was no real prospect that the witness would be prosecuted, the witness would have to testify, whether he wanted to or not. Defense counsel’s subpoena, in other words, would not have to say please. Under Carter II, the unwillingness of the witness to submit to debriefing ends the defendant’s hope of obtaining the witness’ testimony, no matter how exculpatory the testimony may be, and regardless of how unlikely it is that the witness will actually be prosecuted.
Let us compare the operation of Jaggers with the operation of Carter II in a situation where the choice between the two approaches makes a real difference. Suppose that the judge, after hearing from all concerned, determines that the testimony of a prospective defense witness could legally facilitate his prosecution. The judge further finds, however, after examining the government’s uniform practice over a period of years, that the prospects of prosecution are trifling and fanciful, rather than real and substantial. Under Jaggers,
1. the defendant was allowed to present the testimony of the witness;
2. that testimony, having been compelled by the judge, could not be used against the witness;5 and
3.the government was thus powerless to prosecute the witness on the basis of his testimony.
Under Carter II, on the other hand,
1. the defendant can call the witness only if the government discerns “exceptional circumstances” and grants immunity, or if the court finds misconduct by the prosecution, neither of which is an event likely to happen often;
2. there will be no testimony for the government to use against the witness;
3. the government will thus have no basis for prosecuting the witness.
The results of our comparison are apparent. For the witness and for the government, the consequences under the two procedures are essentially the same. For the defendant, however, they are dramatically different. Under Jaggers, the defendant had the opportunity to present the testimony of the witness. Under Carter II, he ordinarily does not. From the perspective of the public interest, the judge and the jury were in most cases in a position to hear relevant evidence under Jaggers, but they will usually be denied access to that evidence under Carter II. Moreover, in what will doubtless be the rare case in which the court concludes that immunity was wrongfully withheld, the remedy may well be the dismissal of the prosecution, which will preclude a disposition of the case on the merits.
In one respect, the Carter II rule will be more favorable to the government than the Jaggers rule was. The prosecution will have more leeway in preventing the defense from calling witnesses, simply by refusing to grant use immunity even to those who, if the government’s conventional procedures were followed, would not be prosecuted at all. The judge may interfere, under the majority’s approach, only if he or she finds prosecutorial “over-reaching” or deliberate “distortion of the fact-finding process.” Although the majority’s stated test is whether the government has “submit[ted] to the court a reasonable basis for not affording use immunity,” *355unreasonableness in this context apparently requires bad intent or its equivalent.6
Under Joggers, on the other hand, it was not necessary to explore the depths of the prosecutorial soul for the presence of impermissible motivation. Rather, the judge made an objective determination as to whether the possibility of prosecution was real or illusory. Joggers had it right, for “[i]t is of no consolation to an individual denied [his constitutional rights] that it was done in good faith.” Burton v. Wilmington Parking Auth., 365 U.S. 715, 725, 81 S.Ct. 856, 862, 6 L.Ed.2d 45 (1961).7
III.
TREATING DIFFERENT SITUATIONS DIFFERENTLY
As I have noted above, this entire hornet’s nest could have been avoided if we had left Joggers and its progeny alone, at least until someone came to us with a legally cognizable injury attributable to Joggers. Now that we have gone en banc, however, it cannot be gainsaid that Judge Gallagher has assembled an impressive array of authorities for two propositions on which the government relies:
1. that it is not the province of the judge to predict what action the prosecution will or will not take vis-a-vis a prospective defense witness;8 and
2. that the validity of a witness’ invocation of the privilege against self-incrimination turns on the legal possibility of prosecution, and not on the question whether, as a practical matter, the witness is likely to be prosecuted.
Formidable as these authorities may seem, however, most of them deal with a situation quite different from the one presented here, and none is dispositive of our ease. As the division majority explained in Carter I,
Judge Gallagher relies on United States v. Miranti 253 F.2d 135, 139 (2d Cir.1958), and on a substantial number of other decisions from various jurisdictions, which would limit the trial judge’s inquiry, in assessing a witness’ Fifth Amendment claim, to the question whether prosecution of the witness would be legally possible (rather than reasonably possible). The claims of privilege in Miranti (and in most of the other cases relied on by Judge Gallagher) were asserted by witnesses whose testimony the government was seeking to compel, and not by prospective defense witnesses. There was thus no actual or potential collision between the rights of an accused under the Sixth Amendment and a witness’ privilege against self-incrimination, and there was no occasion for the courts to attempt to balance or reconcile these basic constitutional protections. Cf. Wilson [v. United States], 558 A2d [1135], 1140 [(D.C.1989)] (because forced election between Fifth and Sixth Amendment rights is so painful, courts must attempt to preserve them both to a reasonable extent).
A few courts have followed the analysis utilized in Miranti even where the privilege has been asserted by a defense witness whose testimony a criminal defendant has sought to compel. See, e.g., Commonwealth v. Francis, 375 Mass. 211, 375 N.E.2d 1221, 1224-25 , cert. denied, 439 U.S. 872, 99 S.Ct. 205, 58 L.Ed.2d 185 (1978); In re Keijam T., 226 Conn. 497, 628 A.2d 562, 565-66 (1993). In Francis, the court explicitly “rejected] the defendant’s contention that we should attempt to “balance’ his rights under the Sixth Amendment against his [witness’] decision *356to invoke the Fifth Amendment.” 375 N.E.2d at 1224. This approach apparently permits a purely theoretical possibility that a witness could be prosecuted to deprive a criminal defendant of potentially exculpatory testimony which may be essential to his defense. See Wilson, supra, 558 A.2d at 1140, and authorities there cited; (James) Harris v. United States, supra, 614 A.2d [1277,] 1283 n. 10 [ (D.C.1992) ].
643 A.2d at 358 n. 17.
Some of the decisions on which Judge Gallagher relies are evidently animated by a civil libertarian spirit which would rarely be generated by a case in which the defendant has a palpable need for the witness’ testimony, and in which the practical likelihood of the witness’ prosecution is nil. The Supreme Court has aptly cautioned that the
words of our opinions are to be read in the light of the facts of the order under discussion. To keep opinions within reasonable bounds precludes writing into them every limitation or variation which might be suggested by the circumstances of cases not before the Court. General expressions transposed to other facts are often misleading.
Armour & Co. v. Wantock, 323 U.S. 126, 132-33, 65 S.Ct. 165, 168, 89 L.Ed. 118 (1944) (emphasis added); see also Khiem v. United States, 612 A.2d 160, 164 (D.C.1992), cert. denied, 507 U.S. 924, 113 S.Ct. 1293, 122 L.Ed.2d 684 (1993). I am not at all sure that the courts which decided most of the cases on which the majority relies would rule in the government’s favor on facts such as these, in which the defendant’s right to a fair trial is implicated, and where the actual likelihood of self-incrimination is remote or even non-existent.
Where the privilege against self-incrimination is the only constitutional protection at issue, it should be accorded the broadest possible scope. If, however, one person’s Fifth Amendment protection comes into conflict with another individual’s Sixth Amendment right to present a defense, then some reasonable accommodation must be made. “In the crunch, when all else fails, the Fifth Amendment privilege of the witness prevails over the defendant’s right to compel him to testify.” Wilson, supra, 558 A.2d at 1140. Courts should, however, take reasonable steps to avert the crunch.9 There ought surely to be some “give” at the outer edges of the Fifth Amendment right where the actual danger of self-incrimination is negligible,10 and where the defendant genuinely needs the evidence in order to receive the fair trial to which he is entitled under the Sixth Amendment.
When a defendant confronts the awesome power and resources of the government in a criminal trial, courts should jealously guard his right to present his defense. He should be denied the right to present relevant exculpatory evidence only upon a showing that his need for that evidence is outweighed by some even more compelling interest. Recently, in Winfield v. United States, 676 A.2d 1 (D.C.1996) (en banc), this court removed one unreasonable impediment — the “clearly link” doctrine — to the vindication of the accused’s Sixth Amendment rights. I fear that in the present case, the court has constructed a new such impediment to take the place of the old one.
By the time this case went to trial, the government knew or should have known for years that Craig Carter had been using drugs. In fact, Craig’s work release had been revoked on account of such use. His urine tests had been “dirty.” The Chief of the Felony Trial Division of the United States Attorney’s office candidly advised the trial judge that it was “extremely unusual for the government to ever prosecute misdemeanor drug possessions based on historical testimony.” He was right; in more than sixteen years as a judge in the District of Columbia court system, I do not recall hearing of a single case in which the government *357did so. Yet, because the theoretical legal possibility of prosecution existed, George Carter was precluded from presenting testimony from Craig Carter which might have cast serious doubt on the credibility of one of George Carter’s accusers. A statistically negligible chance of harm to Craig was used to cause real and palpable harm to George.
Somewhere in our hard-nosed judicial landscape of “bright line” rules and “no exceptions” — of “a rose is a rose is a rose, and that, Mr. Defendant, is that” — there ought to be room for a measure of proportionality and for the saving grace of common sense — for a decision, I suggest, like Jaggers. Perhaps Mark Antony was right. To paraphrase his famous oration,
The evil that [cases] do lives after them
The good is oft interred with their [prose]
So let it be with [,Jaggers ].
William ShakespeaRE, Julius Caesar, Act III, Scene 2. All the same, I count myself among those who would like to place a bouquet of forget-me-nots on the Jaggers decision’s grave.11
. Even the "worst case” scenario under Jag-gers — one that evidently has not yet occurred— would not really endanger the rights of the witness. Suppose that the judge concluded that the possibility of prosecution was imaginary and trivial rather than real and substantial, but that the United States Attorney subsequently disagreed and attempted to use the compelled testimony in a subsequent prosecution of the witness. The Supreme Court has stated that “if [the witness] is nevertheless compelled to answer, his answers are inadmissible against him in a later criminal prosecution.” Lefkowitz v. Turley, 414 U.S. 70, 78, 94 S.Ct. 316, 322, 38 L.Ed.2d 274 (1973). Thus, as a practical matter, a judge’s ruling that the witness may not invoke his privilege against self-incrimination would protect the witness even in the event that the prosecutor subsequently acted in a manner inconsistent with the judge's expectations.
. I will refer to today’s en banc decision as Carter II.
. In general, a party is not entitled to consideration on rehearing of an issue which it failed to raise before the division. See, e.g., U.S. v. Smith, 781 F.2d 184, 185 (10th Cir.1986); Manon Steam Shovel Co. v. Bertino, 82 F.2d 945, 948 (8th Cir.), cert. denied, 299 U.S. 556, 57 S.Ct. 17, 81 L.Ed. 409 (1936). In light of Alston, the govemment was free to urge the division to reject daggers, and such a request would not necessarily have been futile, for the division had the power to grant it.
.Ironically, the time may come when the government also objects, for the remedy suggested by the majority for extreme cases — dismissal of the indictment — is a good deal more drastic and intrusive than the J aggers approach.
. See Lefkowitz, supra note 1, 414 U.S. at 78, 94 S.Ct. at 322.
. The majority states, for example, that if the government denies us immunity because the witness declines to be debriefed, then "prosecutorial bad faith would hardly be present_" Id. at 343-344 (emphasis added).
. To the extent that the government might object to the loggers procedure because it contemplated the judge’s questioning of prosecutors as to their charging practices, I believe that the court’s Carter II methodology is likely to prove at least as intrusive, if not more so.
.The Supreme Court has recently commented on the difficulties in determining, in the context of an allegedly pretextual traffic arrest, what a police officer "would have” done but for his assertedly improper motive. See Whren v. United States, — U.S. —, —, 116 S.Ct. 1769, 1774-76, 135 L.Ed.2d 89 (1996). The present situation is at least somewhat comparable.
. For measures that could have been invoked in this case, see Carter I, 643 A.2d at 353-54 ("The Road Not Taken”) — a passage which is endorsed by Judge Ruiz in her separate statement, and which apparently is acceptable to the majorily. See maj.op. at 342 n. 7.
. Actually, in light of Lefkowitz v. Turley, supra note 1, 414 U.S. at 78, 94 S.Ct. at 322, the danger is evidently not merely negligible, but nonexistent.
. If I could set aside my belief that we should not abandon Jaggers — a belief that Judge Ruiz unfortunately does not share — I would be sorely tempted to join much of her thoughtful and constructive separate opinion.