Appellant Webster was convicted on multiple counts of possessing and distributing cocaine. The evidence that he did so was overwhelming; indeed, he took the stand and admitted his involvement in the larger transaction charged. His defense at trial was founded entirely on a theory of entrapment: that he was seduced into committing the offense by the importunings of a female government informant with whom he was having an affair and who supplied him with the contraband. He asserts three points on this appeal.

The first and most troubling point relates to rebuttal evidence put on by the government to impeach Webster's entrapment defense. This took the form of testimony from a Drug Enforcement agent that some months before the transactions charged, a confidential informant who had proved reliable and trustworthy in the past told him that Webster had sold him cocaine on about five occasions. The testifying agent did not know Webster personally but gave evidence that the informant called Webster by his first and last name, correctly described him as a frequenter of Hialeah Race Track, and gave an accurate physical description of him as a large, white male weighing about 260 pounds. Objections to this evidence on the ground that it was hearsay were overruled, as was a motion for mistrial grounded in the agent's admission that he did not himself know Webster and was therefore necessarily guessing at whether the informant's Keith Webster and the defendant were one and the same.

We characterize this point as troubling, not because we are in doubt that on binding authority it must be rejected but because Webster's attack on the authority that binds us is strong. Strong or weak, however, it cannot carry, since our cases clearly hold that such evidence is admissible. Indeed, a recent opinion of our court goes so far as to itself criticize our rule but concludes that only the court sitting en banc can alter it. United States v. Daniels, 572 F.2d 535, 539 (5th Cir. 1978) (concerning hearsay within hearsay). Thus, as to the admission of this rebuttal evidence, there is no decision for us, only obedience. Nevertheless, because of the attack's force, we shall briefly discuss it and the development of our existing rule admitting such evidence.

Defendant's attack is grounded both in logic and in the rejection of our rule by other authorities. In particular, he relies on the reasoning of United States v. McClain, 531 F.2d 431 (9th Cir. 1976), a Ninth Circuit opinion that, having spent several pages excoriating this court's rule as "astonishing," etc., concludes that the error was harmless to McClain. The logic of McClain is appealing: that our present rule represents the uncritical transfer of a principle from a context in which it was valid into one where it is not. The first context identified by McClain was that in which the entrapment defense focused largely, or at least significantly, on the character of the Government's conduct, where it was therefore appropriate to show the basis of a good-faith belief by the policeman that the defendant was engaged in criminal activity. Thus, the "hearsay" about defendant's activities and reputation was admissible, not for the truth of its assertions, but to show a reasonable basis for the policeman's belief and consequent actions, rebutting the notion that he set out to entrap an innocent. The second context arrived, according to McClain, with the decision in United States v. Russell, 411 U.S. 423, 93 S. Ct. 1637, 36 L. Ed. 2d 366 (1972), where the Supreme Court, reversing the Ninth Circuit, held that it is the defendant's state of mind, not that of the policeman, that counts. With this McClain opines, the logical basis of our circuit's rule was destroyed, and such matter became out-and-out hearsay, improperly admitted for the truth of the matter asserted.

Though we see much force in this analysis and might well follow it were we unfettered by binding precedent, it is not without serious flaws, flaws that result from its over-simplified view of the historical development and present status of the entrapment defense. McClain's analysis to the contrary notwithstanding, it was not at the time we adopted our rule, nor has it ever been, the law that the focus in entrapment decisions is primarily on the character of the government conduct involved. The quality of that conduct has always been viewed by the high Court as a secondary consideration. As the Supreme Court teaches in Russell, supra at 428-32, 93 S. Ct. 1637, the notion that government conduct was the primary or sole consideration has from the first since the Court recognized that defense in Sorrells v. United States, 287 U.S. 435, 53 S. Ct. 210, 77 L. Ed. 413 (1932) been a minority view on the Court. Predisposition of the defendant is now and always has been the first consideration.

Nor is it necessarily so that with Russell predisposition, having once been a secondary consideration, became the sole one. The Russell Court's opinion speaks of nonpredisposition as "the principal element in the defense," Id. 411 U.S. at 433, 93 S. Ct. at 1643, which seems to imply that there are subsidiary ones. And elsewhere, the Russell Court hypothesizes government conduct so outrageous that it may constitute a defense. Id. at 431, 93 S. Ct. 1637. Our circuit's authorities have long considered both the predisposition of the defendant and the character of the conduct of government agents to be material to the entrapment defense. Washington v. United States, 275 F.2d 687, 689 (5th Cir. 1960); Accardi v. United States, 257 F.2d 168, 172-73 (5th Cir. 1959). In the latter opinion cited, our panel "weighed carefully the conduct of the government agents," correctly recognizing that in Sherman v. United States, 356 U.S. 369, 78 S. Ct. 819, 2 L. Ed. 2d 848 (1958), Sorrells' follow-on companion, "the language of the majority of the court, shows that they attach almost as much importance as the minority to the Conduct of the government agents." Accardi, supra at 173 (emphasis in original). And though the 1973 decision in Russell clearly requires us to give preeminence to the predisposition factor, neither it nor reason requires us to abandon all consideration of the character of police conduct in a given affair.

Surely it is possible that what Webster sought to show in this case that he was an utter innocent, corrupted and seduced by the sexual favors of a government agent into selling back to the government cocaine supplied him by that agent at the government's behest might be seen as "outrageous" by some courts. And though we are not connoisseurs of the degrees of outrage, we are willing to hazard that the showing of an honest and well-founded belief by the government agents involved that Webster was no innocent at all, but rather a criminal who dealt routinely in large amounts of contraband drugs, would remove considerable of the blush from Webster's entrapment-outrage rose. We therefore cannot say that the showing made here was irrelevant to the character of government conduct that constitutes an element of the entrapment defense. Nor do either logic or clear authority teach that, when the character of that conduct is attacked as it was here, it may not be defended by proof that it was motivated by a desire to trap the unwary criminal, not the unwary innocent. See Sherman v. United States, 356 U.S. 369, 372, 78 S. Ct. 819, 2 L. Ed. 2d 848 (1958). If so, the evidence in question was properly admitted in rebuttal for this defensive purpose.

Webster seeks to distinguish his case by asserting that here, unlike in other cases, the agent testifying to the "hearsay" did not know and therefore could not identify him. If our above analysis is correct, this does not matter. What matters is that the DEA agents might reasonably have believed that the Webster to whom they furnished an opportunity to deal in cocaine was the same Webster who was already routinely dealing in it, not the victimized na if he now claims to be. In short, we recognize the force of the attack that appellant delivers, mounted on McClain. We are not, however, persuaded that the matter is so simple or our circuit's rule so irrational as appellant and McClain would have it, nor are we free to depart from our rule did we wish to do so. We therefore reject this contention.

Webster's other two points do not require so much discussion. Both relate to the government's refusal to disclose the true identity of its female informant who, in Webster's version of the facts, beguiled him to commit his crimes. We, like the Supreme Court, enjoy an enforced familiarity with the drug scene, derived from our duty to peruse the transcripts of many trials. This vicarious experience of a dark side of the contemporary American landscape teaches that the "faceless informant" is one of its standard inhabitants. Moved by a desire for vengeance or for money, by the hope of lenience for his own misdeeds, by revulsion at drug trafficking, or by any combination of the above, he is a resource of police intelligence information apparently indispensable to their penetration of this criminal subculture. See Rovario v. United States, 353 U.S. 53, 77 S. Ct. 623, 1 L. Ed. 2d 639 (1957) (especially Clark, J., dissenting). To at least the same degree, his exposure and elimination is a matter of the greatest continuing concern to drug operators; he is the spy at their councils, the mole in their operational table. Ibid. Few prosecutions for drug dealing proceed without his involvement to one degree or another; and of these, few do not include a demand by the defense that his identity be disclosed. In so observing, we in no sense imply that such demands by counsel are motivated by any invidious or general desire to expose informers; making such a demand places counsel in the happy tactical position of being able either to pillory the often all too vulnerable informant before the jury or to rail at his client's faceless accuser and at the prosecution's use of such measures. To the contrary, we do no more than recognize that whenever such a demand is made by counsel, pursuing the legitimate tactical demands of his client's defense, other values and considerations outside the case in hand are implicated.

Roviaro, supra, recognizes these competing interests and, rather than any fixed rule, enunciates a balancing test to be applied case by case:

The problem is one that calls for balancing the public interest in protecting the flow of information against the individual's right to prepare his defense. Whether a proper balance renders nondisclosure erroneous must depend on the particular circumstances of each case, taking into consideration the crime charged, the possible defenses, the possible significance of the informer's testimony, and other relevant factors.

353 U.S. at 62, 77 S. Ct. at 629. In applying this rule, our cases have necessarily arrived at somewhat disparate results, which do not lend themselves to easy generalization because of the multiple factors to be considered in each case. It is safe to say, however, that the more closely the informant is connected to the criminal transaction itself as opposed to being a mere tipster or introducer of persons the more likely a disclosure of identity should be required. Here she was, as to Webster and the agents, a mere introducer; no action or statement of hers figured in the transaction proper, and Webster was free in the field to describe to his disparagement his version of how she led him down the primrose path. It is hard to see how her testimony, which might be expected to characterize her actions somewhat more favorably, could have aided him in this endeavor.

Moreover, Webster made no serious effort to discover her identity. Though he knew she was not present at the drug transactions, his request for particulars asked for her name only if "she was present at the time of the alleged offense." The government responded that she was not, and no further request for disclosure of her identity was made until cross-examination of a government agent during trial; nor was any specific reason ever advanced why Webster needed her name or testimony. The other reference to her at trial occurred when, after Webster testified, defense counsel bewailed before the jury his inability to call her and rested. As might be imagined, her absence and anonymity figured largely in counsel's summation. In view of these things, it is hard to see what help she could have afforded Webster's case and harder still to avoid the conclusion that counsel preferred, as a tactical choice, to denounce her absence than to secure her presence. The refusal of the trial court to require disclosure of her identity was not error in this case.

Webster's final point complains of the prosecutor's response to a portion of the summation of the defense that castigated the government's failure to produce the informant as a witness and inquired rhetorically why she was being hidden. In reply, the government observed that she was a young girl, while Webster was a big man, propounding its own rhetorical question about what might happen to her if Webster learned her identity and whereabouts. An objection by the defense was sustained, but no curative instruction was either requested or given. Assuming that the prosecution's response was not a fair one in the circumstances presented, we cannot say that a proper instruction would not have cured whatever unfair prejudice it may have occasioned. And we certainly cannot say that the failure of the trial judge to deliver such an instruction sua sponte was plain error.

For the above reasons, the judgment of the trial court is

AFFIRMED.

I concur in the result but only because I am bound to do so. I have always thought that this circuit is wrong in permitting the prosecution, where the defendant pleads entrapment, to introduce hearsay testimony as tending to prove the defendant's predisposition to commit the crime or the reasonableness of the government's actions, or both. The criticism leveled at us by the Ninth Circuit's opinion in U. S. v. McClain, 531 F.2d 431 (CA9, 1976) is justified.

To understand considerations that are involved one needs to examine the sub-issues that make up an entrapment defense, the shifting evidentiary burdens, the quantum of evidence required to satisfy each burden, and the respective roles of trial judge and jury.

An entrapment defense comes to life when the defendant presents a prima facie case that the government's conduct has created " 'a substantial risk that the offense would be committed by a person other than one ready to commit it.' " U. S. v. Gomez-Rojas, 507 F.2d 1213 at 1218 (CA5, 1975); Pierce v. U. S., 414 F.2d 163 at 168 (CA5, 1969).1 Whether the defendant makes out a prima facie case of substantial governmental involvement is a question of law for the court. Pierce, supra, 411 F.2d at 168.

Once the court rules that the defendant has satisfied his burden of evidence sufficient to raise a question for the jury of whether there is governmental conduct creating a substantial risk, then the government must come forward with evidence sufficient to prove beyond a reasonable doubt that the defendant was predisposed to commit the offense. U. S. v. Dickens, 524 F.2d 441 (CA5, 1975). If the government adduces enough evidence to get to the jury on predisposition, then the ultimate issue of entrapment is submitted to the jury. Pierce v. U. S., supra.

It is against this background that we must balance the prejudice and probative value of the hearsay admitted in this case against the defendant. The camel's nose used by the prosecution, and permitted by our cases, is that the evidence gave the government reasonable probability to suspect that the defendant might be engaged in drug activities, and, therefore, tended to prove that the government's activities with respect to defendant did not create " 'a substantial risk that the offense would be committed by a person other than one ready to commit it.' " I. e., the risk was not "substantial" because the government had sufficient reason to believe that defendant was not " 'a person other than one ready to commit it.' "2 Thus the hearsay evidence comes in on an issue of law reserved to the court.3 Once it is in evidence, the real impact is on the issue of predisposition, which is an issue reserved to the jury. In addition, predisposition is not a reasonable probability issue but a "yes" or "no" issue. In truth, all know that usually the real purpose of the evidence's being offered is its impact on the jury on the issue of predisposition. The final result is that on the ultimate issue of guilt or innocence the defendant is sorely prejudiced by "bad man" evidence. If the offer of "bad man" evidence was offered in some context other than the Byzantine intricacies of entrapment, we would not even consider admitting hearsay. But, hypnotized by the miniscule camel's nose I have described, we permit the destructive evidence to come in full blast.

Our decisions simply have not addressed these problems but, in a broad-brush fashion, have said that entrapment involves a charge that the government has done bad things to the defendant and therefore it is all right for the government to show that it had good reason to think that the defendant was a bad person (whether he really is or not). This simplistic approach overlooks the real considerations that careful scrutiny brings to the surface.

The fallacy of this circuit's approach is demonstrated by the rationale of our early cases addressing this problem. We permitted hearsay evidence of the defendant's Reputation which, though essentially a hearsay concept, bore upon the government's "reasonable cause to believe" that I have described above. E. g., U. S. v. Robinson, 446 F.2d 562, 564 (CA5, 1971). But, as the present case demonstrates, we have left the limits of reputation evidence and now roam free, permitting evidence of prior specific acts of some person never even properly identified as the defendant. If predisposition to commit an offense were an issue considered in isolation we would not seriously entertain the notion that the fact of predisposition could be proved by this kind of amorphous report from a faceless informer describing specific incidents committed by an unidentified person.

U. S. v. Russell, 411 U.S. 423, 93 S. Ct. 1637, 36 L. Ed. 2d 366 (1973), does not change the picture but reinforces what I have said. In Russell the defendant acknowledged that the evidence supported the jury finding that he was predisposed and he sought to escape this by urging that protection against over-zealous governmental actions was of constitutional dimension, I. e., that he was unconstitutionally "entrapped" even though he was predisposed. In declining to accede to defendant's effort, the Court described predisposition as the principal element of the defense and governmental conduct as less significant. This being so, it would seem all the more important that hearsay not slip into the case on the dubious ground that it is somehow relevant to the secondary issue (governmental conduct) when its main prejudicial force strikes at the principal issue (predisposition).