Messrs. Lawrence D. Hollman and Sherwood B. Smith, Jr. (both appointed by this court) for appellant in No. 21,249.

Messrs. James A. Strazzella, Lawrence Lippe, and James E. Kelley, Jr., Asst. U. S. Attys., with whom Messrs. David G. Bress, U. S. Atty., and Frank Q. Nebeker, Asst. U. S. Atty., were on the briefs, for appellee. Mr. Seymour Glanzer, Asst. U. S. Atty., was on the brief for appellee in No. 21,001.

Miss Carol Garfiel, Asst. U. S. Atty., also entered an appearance for appellee in No. 19,846.

Mr. Foy R. Devine, Washington, D. C., with whom Mr. William W. Greenhalgh, Washington, D. C. (both appointed by this court) was on the brief, for appellant in No. 19,846.

Before BAZELON, Chief Judge, and DANAHER, BURGER, WRIGHT, McGOWAN, TAMM, LEVENTHAL and ROBINSON, Circuit Judges.

McGOWAN, Circuit Judge:

These three criminal appeals were, prior to final decision by the panels which initially heard them, placed en banc for hearing and disposition by the full court. This was because each case involved an identification issue growing out of the circumstances under which pre-trial identification had been made. Since several other appeals pending in this court involve the same kind of question, and since it is recurring in the District Court with great frequency, we thought that en banc consideration was desirable in the interest of sound judicial administration. Recognizing that no case is ever quite like another on its facts, we selected three cases typical of the class. Our disposition in each instance appears below,1 after a discussion of the legal principles lately emerging in this field and an examination of the factual context for their application in each case.

* In three cases decided June 12, 1967, the Supreme Court brought into focus its concern with the manner in which pre-trial identifications are frequently made. United States v. Wade, 388 U.S. 218, 87 S.Ct. 1926, 18 L.Ed.2d 1027; Gilbert v. California, 388 U.S. 263, 87 S.Ct. 1951, 18 L.Ed.2d 1178; Stovall v. Denno, 388 U.S. 293, 87 S.Ct. 1967, 18 L.Ed.2d 1199. In the first two of these cases, which involved lineups held at substantial intervals after arrest, indictment, and appointment of counsel, but without counsel present, the Court found a violation of the Sixth Amendment right to counsel. It said in each instance that the lineup was a critical stage in the criminal proceedings at which the accused were constitutionally entitled to have their lawyers present. In rejecting the argument that pre-trial confrontations for identification are under no circumstances significant parts of the process of criminal prosecution, the Court adverted at length to the chronic uncertainties of eyewitness testimony. It prefaced its rehearsal of these problems with the statement that "the confrontation compelled by the State between the accused and the victim or witnesses to a crime to elicit identification evidence is peculiarly riddled with innumerable dangers and variable factors which might seriously, even crucially, derogate from a fair trial"; and it went on the remark that "[a] major factor contributing to the high incidence of miscarriage of justice from mistaken identification has been the degree of suggestion inherent in the manner in which the prosecution presents the suspect to witnesses for pretrial identification." 388 U.S. at 228, 87 S.Ct. at 1933.

On such premises the Court concluded that the presence of counsel at lineups was necessary in order (1) to minimize the likelihood of an unduly suggestive confrontation and (2) to enable an informed challenge to be made at trial to either the admissibility or the credibility of identification evidence. In Wade, a federal prosecution, the Government's case at trial had included only in-court identifications, and the lineup identifications had been brought out by defense cross-examination. The Court, saying that "[o]n the record now before us we cannot make the determination whether the in-court identifications had an independent origin," sent the case back for a resolution by the District Court of that question, or for a determination of whether, in any event, the introduction of the evidence was harmless error under Chapman v. California, 386 U.S. 18, 87 S.Ct. 824, 17 L.Ed.2d 705 (1967).

In Gilbert, a state case, where the in-court identifications had been unsuccessfully objected to as tainted by the uncounseled lineup confrontation, the Court said that a disposition similar to that in Wade would normally be appropriate since "as in Wade, the record does not permit an informed judgment whether the in-court identifications at the two stages of the trial [guilt and penalty] had an independent source." But, since testimony describing the lineup identifications was given by certain prosecution witnesses as part of their direct examination in both stages of the trial, the Court said that the prosecution is "not entitled to an opportunity to show that the testimony had an independent source," and that only a per se exclusionary rule can effectively assure respect for the right to counsel. Therefore, said the Court, unless the California Supreme Court on remand could declare a belief under Chapman that the violation was harmless beyond a reasonable doubt, Gilbert must have a new trial.

Stovall was a federal habeas corpus attack upon a state conviction. The defendant, arrested within a day after a brutal assault committed in the course of a robbery, was the following day taken in police custody to the hospital room of the victim who was in critical condition after surgery. The only Negro in the room, manacled to one of five white policemen and accompanied by two prosecutors, and not yet represented by counsel although his preliminary hearing had been continued for that purpose, the defendant was identified after the police asked the victim if he "was the man." The victim recovered, and at trial made an in-court identification of Stovall and also testified on direct examination in the presence of the jury that she had identified the defendant at the hospital.

The Court ruled in Stovall that the Sixth Amendment right to counsel recognized on direct appeal in Wade and Gilbert would be given only prospective effect. Thus confined to pre-trial confrontations occurring after June 12, 1967, Stovall's conviction was unassailable on this ground. But the Court then, in the brief compass of one page, (1) identified, as a different and independent ground of attack, a claim that the pre-trial confrontation "was so unnecessarily suggestive and conducive to irreparable mistaken identification that [defendant] was denied due process of law,"2 (2) observed that "[t]he practice of showing suspects singly to persons for the purpose of identification, and not as part of a lineup, has been widely condemned,"3 and (3) held that, since "a claimed violation of due process of law in the conduct of a confrontation depends on the totality of the circumstances surrounding it," the Court of Appeals had been right in finding no violation because a police station lineup had not been feasible under the circumstances. The opinion for the Court made no reference to the fact that the State at trial had buttressed the victim's in-court identification by eliciting the pre-trial confrontation on direct examination in the jury's presence.

Stovall is of primary importance to the cases before us for two reasons. First, its holding on the full prospectivity of the Sixth Amendment right to counsel eliminates that issue, and, second, it is, in strict contemplation, the sole source of Supreme Court doctrine on the contours of due process in the area of pre-trial confrontations for purposes of identification. As such, the illumination it provides is meager indeed, partly because of its brevity and partly because of the result reached. On two occasions since Stovall, however, the Supreme Court has addressed itself to due process in pre-trial identification, and we look to those cases for such help as they may give.

The more significant of these is Simmons v. United States, 390 U.S. 377, 88 S.Ct. 967, 19 L.Ed.2d 1247 (1968). At trial the Government's case on identification consisted solely of in-court identifications made by five employees who witnessed the robbery of a bank. Simmons urged that this testimony was tainted by pre-trial identifications made through photographs under circumstances amounting to a denial of due process or at least requiring reversal in the exercise of the Court's supervisory powers. The Court characterized Simmons' claim, and disposed of it, as follows (at p. 383, 88 S.Ct. at p. 970):

"[H]e asserts simply that in the circumstances the identification procedure was so unduly prejudicial as fatally to taint his conviction. This is a claim which must be evaluated in light of the totality of surrounding circumstances. See Stovall v. Denno, 388 U.S. 293, at 302 [87 S.Ct. 1967]; Palmer v. Peyton, 359 F.2d 199. Viewed in that context, we find the claim untenable."

The Court went on to recognize "that improper employment of photographs * * * may sometimes cause witnesses to err in identifying criminals." It identified various methods which it thought undesirable, but it concluded that it was "unwilling" to prohibit the use of pre-trial photographic identification "either in the exercise of our supervisory power, or, still less, as a matter of constitutional requirement."

"Instead, we hold that each case must be considered on its own facts, and that convictions based on eyewitness identification at trial following a pre-trial identification by photograph will be set aside on that ground only if the photographic identification procedure was so impermissibly suggestive as to give rise to a very substantial likelihood of irreparable misidentification. This standard accords with our resolution of a similar issue in Stovall v. Denno, 388 U.S. 293, 301-302, 87 S. Ct. 1967, 1972-1973, and with decisions of other courts on the question of identification by photograph." 390 U.S. at 384, 88 S.Ct. at 971.

The Court alluded to the urgent character of the circumstances in which the photographs were used. A serious felony had been committed, and the bandits were still at large. There were signs pointing to Simmons, but they were inconclusive. The help of identification was greatly needed. The Court noted that each employee had had a good opportunity to observe Simmons at the time of the crime; the photographs were shown to them only a day later when recollections were fresh; there was no evidence of verbal guidance by the police; and, lastly, the employees' in-court identifications were positively made and firmly adhered to under cross-examination. The Court concluded in these terms (at pp. 385-386, 88 S.Ct. at p. 972):

"Taken together, these circumstances leave little room for doubt that the identification of Simmons was correct, even though the identification procedure employed may have in some respects fallen short of the ideal. We hold that in the factual surroundings of this case the identification procedure used was not such as to deny Simmons due process of law or to call for reversal under our supervisory authority."

On the same day Simmons came down, the Court affirmed, by an equally divided court, Justice Marshall not sitting, a rape conviction before it on certiorari to the Tennessee Supreme Court. Biggers v. Tennessee, 390 U.S. 404, 88 S.Ct. 979, 19 L.Ed.2d 1267 (1968). The facts, as stated in the dissenting opinion of Mr. Justice Douglas, are very similar to those in Palmer v. Peyton, note 2 supra, and indicate that (1) Biggers was arrested for another rape seven months after the crime in issue, (2) the victim was brought down to the police station where her identification was made by listening to Biggers speak in another room, (3) there was no in-court identification but only the victim's testimony (and that of police officers) about her voice identification at the station house, and (4) the victim's opportunity to observe her attacker at the time of the assault was limited. Justice Douglas stated that: "Of course, due process is not always violated when the police fail to assemble a lineup but conduct a one-man showup." But it was his view that, on the facts before him, an evaluation of the due process challenge to Biggers' conviction by reference to the governing Simmons standard of "the totality of surrounding circumstances" dictated the conclusion that the identification processes employed were "so unduly prejudicial as fatally to taint his conviction."

With higher authority in this slender state of development, generalization is more hazardous than usual. It would appear, however, that the Supreme Court has, at the least, cast an unmistakable shadow across those post-arrest single confrontations at the police station where formal lineups are a feasible alternative. For the future, of course, all such post-incarceration exposures, singly or in lineup, are subject to the protections of the Sixth Amendment right to counsel, and many of the problems we face today will be nonexistent. However, the resolution of due process claims, both past and future, remains for us, as it was for the Supreme Court in Stovall, Simmons, and Biggers, an inescapable duty. That affirmances were the result in all those cases may or may not be significant, but, in any event, we think the Court has formulated a broad standard of review which focuses upon the distinctive facts of each case in their totality, and which relies very heavily upon the special capacity and experience of judges, trial and appellate, to discriminate between real and fancied dangers of the miscarriage of justice.

In the conduct of trials involving post-Wade and Gilbert identifications, the mode of proceeding would seem to be as follows: Whenever the prosecution proposes to make eyewitness identification a part of its case, the defense is entitled to know, through disclosure by the prosecution or by evidentiary hearing outside the presence of the jury, the circumstances of any pre-trial identification.4 If it was one where the court finds that the Sixth Amendment right to counsel existed but was not observed, the prosecution may not, under the per se exclusionary rule enunciated by the Supreme Court in Gilbert, offer such identification as part of its case; and the same rule would appear to be applicable with respect to prosecution evidence of post-Stovall pre-trial identifications found by the court to be violative of due process.

Where the prosecution intends to offer only an in-court identification, the defense may challenge its admissibility. The court should then, on facts elicited outside the presence of the jury, rule upon whether a pre-trial identification by the same eyewitness is violative of due process or the right to counsel. If a violation is found, the court should then decide whether the in-court identification is still admissible because it has an independent source; indeed, it would appear in the interest of expeditious judicial administration for such a ruling to be made in any event. If the judge regards only the in-court identification as admissible, in the trial to the jury thereafter, the defense may, as a matter of trial tactics, decide to bring out the pre-trial confrontation itself, hoping that it can thus detract from the weight the jury might otherwise accord the in-court identification.

There are, of course, many questions thus far left unanswered by the Supreme Court decisions as to the precise scope of the constitutional rights to counsel and to due process in pre-trial identification. In the case of photographs, Simmons makes clear that only a due process (or, in the case of federal prosecutions, an appellate supervisory power) issue can be involved where the police problem is one of getting leads to possible suspects. A due process violation in this area will largely be a function of the heed given by the police to the considerations alluded to in Simmons as relevant to the fair use of photographs. Any well-run department will presumably prepare ? and enforce ? careful regulations in this regard for the guidance of its personnel, as it will also do with respect to police station lineups.5 Apart from photographic leads and the familiar post-arrest police station confrontation, there are possible variations in the circumstances of arrest and detention where the right to counsel, as well as the demands of due process, will have to be defined and measured from case to case by reference to the reasonableness of the police conduct under the particular circumstances.

We have referred above to the guidance which we think the Supreme Court has thus far afforded for determining whether a conviction following upon eye-witness testimony is or is not below the levels of tolerance of the due process concept. Equipped with that instruction, we turn to the cases immediately before us.

II

No. 21,249 ? Hines v. United States

Hines was convicted by a jury on six counts of robbery. 22 D.C.CODE § 2901. The charges arose out of a holdup of the operators and customers of a beauty parlor at about 7:00 P.M., July 29, 1966.6 Hines was arrested at about 1:00 A.M., July 30, 1966, with $98.00 in his pocket, an amount which he and a corroborating witness claimed he had won at a "crap" game the same evening. This was the only evidence brought out at trial, other than the identification of Hines by various of the ladies in the beauty parlor.

The Government has confessed error on this appeal, and has asked that the case be remanded for a new trial.7 We accede to that request. Nevertheless, both parties have agreed that important issues concerning the admissibility into evidence of identification testimony were raised in the District Court, and should now be considered by this court. Hines has also raised a serious question as to whether he has been denied his Sixth Amendment right to a speedy trial and has asked that, whatever the court's disposition of the identification issues may be, the indictment be dismissed with prejudice because of the long delays in bringing him to trial and in processing his appeal.

We recognize that appellate consideration of what may possibly become moot questions is generally undesirable. This is the only case of the three before us, however, in which the trial was held after the Supreme Court's opinion in Stovall, and it is, thus, the only one in which the District Court ruled at trial on the propriety of certain out-of-court and in-court identifications with a due process standard explicitly in mind.8 It is, moreover, a case in which there was a great variety of identification testimony. The ladies differed not only in their opportunities to observe the robber before and during the robbery, but also in the manner in which, and the positiveness with which, they identified Hines after the robbery. Therefore, we think that, in the interest of efficient judicial administration, we are warranted in discussing the merits of the identification issues raised on this appeal.9

Nine ladies were in the beauty parlor when the robbery took place. Three of them testified that they had seen the same man enter the shop earlier in the day, and two of these heard him ask for his mother when questioned about what he was doing in the beauty parlor. Eight of them had the opportunity to see the robber during the robbery (one hid in the bathroom throughout). Seven of them testified at trial. Six said that the robber had stolen property from them. Of the seven who testified, five saw photographs, four of them the same night as the robbery, one the next morning. Three identified Hines as the robber the following morning when he appeared before a magistrate at a preliminary hearing; the other four saw him on August 16, 1966, the day he was indicted by the grand jury, in a cellblock at the court house.

At the trial, which began June 15, 1967, after a hearing before the judge alone in which one of the three ladies who had seen the defendant at the preliminary hearing testified, all three were permitted to testify both that they had identified Hines as the robber at the preliminary hearing, and that they presently recognized Hines as the robber in the courtroom. Of the four who had viewed defendant in the cellblock, all of whom testified before the judge with the jury absent, two were not permitted to identify Hines at all, and two were not allowed to relate the circumstances of the cellblock identification on direct examination, but were permitted to make in-court identifications. Defense counsel, however, cross-examined both of the latter witnesses with respect to the cellblock confrontation.

Hines has raised three issues with respect to the propriety of the District Court's rulings on the admissibility of identification testimony, and the Government has raised two others. The three questions posed by Hines are:

  (1) Was the trial court correct in holding that the confrontations at the preliminary hearing were not a violation of due process?

  (2) Was the trial court correct in holding that two of the witnesses who had viewed defendant in the cellblock could nevertheless identify him in court?

  (3) Was it error for the trial court to permit pre-trial identifications to be admitted where in-court identifications already made had not been impeached?

The two issues raised by the Government are:

  (1) Was the trial court correct in holding that the cellblock confrontations were a violation of due process? (2) If they were, was the trial court correct in excluding all identification testimony by two of the ladies who had viewed Hines in the cellblock?

The District Court, pursuant to Hines' motion to suppress identification testimony heard the testimony of four witnesses before the jury was called. They were: (1) Mrs. Aggrey, an operator of the beauty salon, who had talked to the robber when he entered the shop earlier, who had examined photographs the night of the robbery and identified Hines as the robber from them,10 and who was taken to the cellblock two and a half weeks after the robbery and identified Hines there as well; (2) Mrs. Parker, also an operator, who had not seen, but only heard, the robber before 7:00 P.M., who had looked at photographs but had not been able to identify anyone, and who was taken to the cellblock and was able to identify Hines as the robber positively only after she had heard his voice; (3) Mrs. Grandy, a customer, who had seen the photographs and was unable definitely to pick out Hines from them other than to say that he was "one of two," but who had identified him as the robber at the preliminary hearing immediately11 and positively; and (4) Detective Simmons, who described the happenings (a) when the ladies looked at the photographs,12 (b) at the preliminary hearing, and (c) at the cellblock viewing.

After hearing argument from counsel, the District Court concluded that (1) the identifications at the preliminary hearing were not a violation of due process, (2) the cellblock viewings were a violation of due process, (3) Mrs. Aggrey would be permitted to identify Hines in court, but (4) Mrs. Parker would not. In upholding the preliminary hearing identifications, the trial court appears to have relied on the evidence that over one hundred people had been milling about the courtroom, that various people had come out of the same door as Hines, and that the ladies were not influenced in any way to identify Hines, were not even told the suspect's name, and had identified him even though he was wearing clothing different from that which the robber had worn. The court also apparently believed Mrs. Grandy's testimony that she had recognized the defendant instantly, and that she had not been influenced in her determination by the two ladies with whom she was sitting. Her testimony in this last respect was that: "We were sitting together and we said it ["that's the fellow"] together."

We are not prepared to reject the District Court's conclusion that the preliminary hearing confrontation survives due process challenge. We note in this regard that the presentment occurred promptly the morning after the arrest and that, in contrast with the cellblock confrontations, there was no lengthy intervening period of time in which a normal police station lineup could have been arranged. It is, of course, possible for a non-lineup confrontation to be held under conditions which assimilate it very closely to a lineup, and this is what the trial court essentially found was the case here. It must be evident, however, that the conditions of such a confrontation are much harder to control than those of a formal lineup, and that it is also much more difficult to establish by clear and undisputed testimony exactly what those conditions were. It is, at the least, a practice fraught with perils to a degree suggesting its sparing use as the part of prudence.

In finding that a violation of due process did occur at the cellblock viewing, the District Court stated that "in taking the situation in its totality" the confrontations were "unnecessarily suggestive." It reached its conclusion on the basis of these facts: (1) Hines was alone in custody, and standing in a screened-off section, when presented to the witnesses; (2) the witnesses knew that Hines had been arrested for the robbery and was considered by the authorities to be the culprit; (3) previous to testifying before the grand jury, the witnesses were asked to make certain that this was the man; (4) they were taken to the cellblock as a group and viewed defendant as a group; (5) at least one witness nodded her head affirmatively during the confrontation; and (6) after hearing Hines' voice, Mrs. Parker announced to the others that she believed him to be the man.

Again we cannot say that the trial court erred in its appraisal of the cellblock confrontation. It found in effect that the conditions which obtained on that occasion (which occurred weeks after Hines had been in custody and available for normal lineup viewing) did not approximate a lineup, and that, contrarily, they brought Hines into a focus unacceptably antithetical to the lineup principle. Thus, we do not disturb the trial court's determination that the Government's direct case could not profit from the cellblock confrontation.

Having found no fault with either of the trial court's holdings on due process, it remains for us to consider its conclusions that Mrs. Aggrey could identify defendant in court, and that Mrs. Parker could not. In finding that Mrs. Aggrey's in-court identification "was not the fruit of the illegal confrontation," the District Court relied on the circumstances that (1) she had seen the robber once prior to the robbery, as well as during the robbery itself, (2) she testified that she had paid particular attention to his appearance, and (3) she had been able to pick out Hines' photograph the night of the robbery. Mrs. Parker, on the other hand, was unable to give the police a description of the robber or to identify Hines' picture after the robbery, was told before going to the cellblock that the man to be viewed was the man identified earlier by the ladies at the preliminary hearing (Mrs. Aggrey testified that she did not know this), and was able to make an identification only on the basis of having heard Hines speak in the cellblock. Thus, it appeared to the trial court "that the witness' testimony certainly is a product of the illegal confrontation of August 16th."

The Supreme Court has, as noted above, expressly contemplated that in-court identifications may be found capable of standing on their own feet, even though preceded by deficient pre-trial confrontations. It has also emphasized the key role which the trial court, because of its direct exposure to the witnesses, plays in any such determination. Examined against this background, we think the District Court's rulings with respect to these two witnesses are not properly the subject of appellate reversal.

Later in the course of the trial, the court, out of the presence of the jury, heard the testimony of two more of the ladies who had attended the cellblock viewing. In the case of Mrs. Steele, who testified positively that she "certainly would" have remembered Hines even if she had not seen him in the cellblock and who gave a detailed description both of Hines' features and of his clothing, the court permitted an in-court identification. In doing so, it apparently relied rather heavily on her credibility in saying that she would have identified Hines at trial even without the prior meeting. This is borne out by the court's contrary ruling in the case of Mrs. Potite, who had seen the robber twice previous to the robbery, but who testified that she probably would not have been able to identify Hines at trial had she not seen him earlier in the cellblock. In granting Hines' motion to suppress Mrs. Potite's identification, the court relied primarily on her testimony that she was uncertain that she would have remembered him without the cellblock confrontation, and only secondarily on the fact that she did not remember anything "particular" about the robber that helped her make the identification.

Again the remoteness of our vantage point disinclines us to reject these rulings, resting so heavily as they do on the immediate impact of the two witnesses' testimony upon the trial court. It may seem somewhat anomalous that the in-court identification of Mrs. Potite was excluded, although she saw the robber twice before the robbery as well as during the robbery itself, whereas that of Mrs. Steele was admitted, despite the fact that she nearly fainted on that occasion. But Mrs. Potite herself attributed great significance to her cellblock viewing and was vague about her recollections of Hines during the robbery, whereas Mrs. Steele was quite positive on the latter score. This does suggest that, although the positiveness of the witness about an independent base for an in-court identification is a relevant factor, it is to be weighed warily and in the realization that the most assertive witness is not invariably the most reliable one. But this is a phenomenon of which experienced trial judges need no reminder from us, and we have no reason to doubt that the rulings in question were made with ample awareness of it.

We turn, finally, to the third identification issue raised by Hines, namely, the claim that it was error to permit a witness whose in-court identification was unimpeached also to relate a pre-trial identification. Among the in-court identifications made at trial were those by witnesses Reid, K. Thomas, and Grandy, who had identified Hines at the time of his preliminary hearing, but were not in the group taken to the cellblock. Their in-court identifications were supplemented by their testimony of the preliminary hearing identifications. Hines argues that this was error, either because the preliminary hearing confrontation was a violation of due process or, alternatively, because testimony regarding the prior identifications, offered as independent support for the in-court declarations, was hearsay. What we have said above concludes Hines on the first of these grounds. We are also of the view that the second is insupportable.

It is true, as Hines asserts and the Government recognizes, that there is conflicting authority on this question in the state courts and that there has been no authoritative resolution of it for this jurisdiction by this court.13 The Second Circuit, speaking through Judge Augustus N. Hand, has ruled in favor of admissibility. United States v. Forzano, 190 F.2d 687, 689 (1951); see also the continuing approval of this approach expressed in United States v. Miller, 381 F.2d 529, 538 (2d Cir. 1967) (Friendly, J.). Other federal circuits have inclined in the same direction. See Bolling v. United States, 18 F.2d 863 (4th Cir. 1927); Eidson v. United States, 272 F. 2d 684 (10th Cir. 1959).

The Supreme Court specifically took note of the problem in Gilbert. It said there that:

There is a split among the States concerning the admissibility of prior extra judicial identifications, as independent evidence of identity, both by the witness and third parties present at the prior identification. See 71 A.L.R.2d 449. It has been held that the prior identification is hearsay, and, when admitted through the testimony of the identifier, is merely a prior consistent statement. The recent trend, however, is to admit the prior identification under the exception that admits as substantive evidence a prior communication by a witness who is available for cross-examination at trial. See 5 A.L.R.2d Later Case Service 1225-1228. That is the California rule. In People v. Gould, 54 Cal.2d 621, 626, 7 Cal.Rptr. 273, 275, 354 P.2d 865, 867, the Court said:

"Evidence of an extra-judicial identification is admissible, not only to corroborate an identification made at the trial (People v. Slobodion, 31 Cal. 2d 555, 560, 191 P.2d 1), but as independent evidence of identity. Unlike other testimony that cannot be corroborated by proof of prior consistent statements unless it is first impeached * * * evidence of an extra-judicial identification is admitted regardless of whether the testimonial identification is impeached, because the earlier identification has greater probative value than an identification made in the courtroom after the suggestions of others and the circumstances of the trial may have intervened to create a fancied recognition in the witness' mind. * * * The failure of the witness to repeat the extra-judicial identification in court does not destroy its probative value, for such failure may be explained by loss of memory or other circumstances. The extra-judicial identification tends to connect the defendant with the crime, and the principal danger of admitting hearsay evidence is not present since the witness is available at the trial for cross-examination." New York deals with the subject in a statute. See N.Y.Code Crim.Proc. § 393-b.

388 U.S. at 272, 87 S.Ct. at 1956, n. 3.

The "recent trend" so remarked received powerful impetus many years ago from Professor Wigmore. See 4 WIGMORE, EVIDENCE § 1130 (3d ed. 1940). We think that the rationale behind the exclusion of hearsay evidence has little force in the case of witnesses, such as those here involved, who are available for cross-examination. We also think that juries in criminal cases, before being called upon to decide the awe-some question of guilt or innocence, are entitled to know more of the circumstances which culminate in a courtroom identification ? an event which, standing alone, often means very little to a conscientious and intelligent juror, who routinely expects the witnesses to identify the defendant in court and who may not attach great weight to such an identification in the absence of corroboration. We, therefore, find no error in the trial court's handling of this matter.

Hines' conviction is, as requested by the Government, reversed, and the case is remanded.

No. 21,001 ? Clark v. United States

Clark was convicted by a jury of the armed robbery of a liquor store. 22 D.C. CODE § 2901. In the store at the time (shortly before 9:00 P.M., February 14, 1966) were the owner, Alberstadt, and an employee, Jones. The robber placed a paper bag on the counter in front of Alberstadt and demanded that it be filled with money. Alberstadt pretended not to understand for a time, and the demand was repeated. As Alberstadt started to comply, Jones approached the robber, but was confronted with a sawed-off shotgun and ordered to stand still. The police arrived promptly after the robber had fled and, upon the basis of descriptions given by Alberstadt and Jones, the following message went out on the police teletype:

"Lookout for white male, 35 years of age, 6'2", 185 to 195 lbs., blond hair, light greenish-gray eyes, wearing a red checkered snap-brim cap, dark corduroy, three-quarter length coat, armed with a single-barreled sawed-off shotgun."

Later that evening both Alberstadt and Jones went to the police station and were shown a number of photographs. Each tentatively identified Clark. A few weeks later, on March 6, Clark was invited by a detective whom he knew to come down to the police station. While there, the detective called Alberstadt and asked him to come down. When Alberstadt entered the Robbery Squad room, he saw Clark seated at a desk, with his feet up on the table and talking on the telephone. Alberstadt immediately jumped over the railing to get at Clark, shouting that he was the robber. Clark was then arrested and charged with the crime.

The next morning Jones was asked to come down to the police station. He was taken to a detention area where there were several prisoners, and asked if he could identify anyone. He asked that Clark, who was seated, arise and come closer. When he did, Jones identified him as the robber.

On appeal of his subsequent conviction to this court, Clark argued, first, that the eyewitness testimony against him was the product of an illegal arrest, and, second that, in any event, his pre-trial identifications were violative of due process and rendered all identification at trial invalid. The panel of this court which heard the appeal in the first instance remanded the case to the District Court for augmentation of the record.14 The District Court held an evidentiary hearing, after which it made detailed findings of fact and conclusions of law. United States v. Clark, 294 F.Supp. 44 (D.D.C. June 17, 1968).15

With respect to Alberstadt, the District Court found that, when Detective Hannon telephoned him to come down to the police station on Sunday afternoon to view a suspect, it was not suggested that the suspect was the one tentatively identified earlier from photographs nor was anything else said about the suspect he was to see. Detective Hannon did not see him arrive, and no one said anything to him before he rushed at Clark immediately upon his entry into the room. Some six to eight detectives were at work at desks in the room, or standing about the room in conversation, a majority of whom were white. They were dressed in normal business clothes, some having their coats off and ties loosened, while Clark was casually dressed as well. At least one white detective was nearly as tall as Clark, and others were over six feet.

The District Court, correctly we think, characterized Stovall as directed "primarily toward single suspect confrontations wherein a witness is presented with one individual and asked if he can make an identification;" and it observed "that the possibility of such a procedure amounting to unfairness violative of due process must be examined in light of the totality of the surrounding circumstances." It enumerated a number of tests it believed relevant to this end,16 and, applying these to the facts it had found, it concluded that Alberstadt's identification of Clark was not violative of due process. We accept that conclusion. It is one that carries with it the consequences that (1) evidence of the pre-trial identification by Alberstadt is admissible as part of the Government's case and (2) there can be no question of the propriety of his in-court identification.

The District Court appeared to think that, as to (1), the Government had not sought at trial "to bolster [its] case by placing before the jury the witness' prior identification;" and that it was "only after cross-examination had begun that the jury became cognizant of the fact that the witness had been able to identify defendant on a prior occasion." This is not quite so clear to us since Alberstadt did testify on direct examination that he had seen Clark prior to trial. In any event, the finding of no violation makes the matter academic. As to Alberstadt's in-court identification, the District Court found that, even if a pre-trial violation be assumed, there was clear and convincing evidence of an independent source. In this regard the court noted the good opportunity Alberstadt had to observe Clark at the time of the robbery, the accurate description he gave to the police immediately thereafter and his tentative identification from photographs later that same evening. Although we do not need to deal with the merits of this conclusion in view of our upholding of the determination of no violation, we think it was in the interest of efficient judicial administration of the District Court to proceed as it did. See p. 1237, supra.

With respect to Jones, the District Court found that on Monday, March 7, Jones was told by Alberstadt to report to police headquarters. There he was met by Detective Hannon and taken to the detention area behind the room where lineups are held. There were about twelve prisoners in the cell in addition to Clark, only about two of whom appear to have been white. Jones was not told by Hannon which man in the group was the suspect.

The District Court, applying the tests referred to above in note 16, concluded that the cellblock confrontation was unnecessarily suggestive. In reaching this result, the court, although professedly unable to perceive that any "affirmative suggestive influences" had been brought to bear, was at a loss to understand why a lineup was not held, since Clark was already in custody and, indeed, in the lineup area. In view of these latter facts, said the court, "Det. Hannon should have been able to delay the confrontation in order to fashion a better test of the witness' ability to identify defendant." Since, in the court's view, a jury inevitably regards evidence of a prior identification as probative of guilt, "[I]f the challenged confrontation is suggestive, and as such not a good test of the witness, the jury should not be permitted to hear of the confrontation and draw this inference."

The court thus decided that the cellblock identification could not come in as a part of the Government's case, but it noted that this had not happened at trial. The fact of the cellblock confrontation was brought out initially by the defense, and it was the court's view that the Government was entitled, by rebuttal testimony, to give its version of that confrontation.17 The court then addressed itself to the question of whether Jones' in-court identification could be said to have been shown by clear and convincing evidence to have had a foundation independent of the cellblock confrontation. It concluded that the Government had met its burden in this regard, noting that (1) Jones had observed the unmasked robber under good lighting conditions, (2) he recalled clearly the robber's distinctive physical appearance, (3) he had picked out his photograph from a display shown to him the night of the robbery, and (4) he had testified firmly and positively that his in-court identification was not influenced by the cellblock confrontation.

We are far from being able to say that the District Court erred in these conclusions. The District Court has, thus, discharged the task assigned to it in the panel's remand by careful and comprehensive fact-finding, including credibility determinations, and by the application of legal principles which we conceive to be compatible with the legal principles thus far enumerated by the Supreme Court in the area of eyewitness identification. Clark's conviction is, accordingly, affirmed.

No. 19,846 ? Clemons v. United States

Clemons was convicted by a jury of four counts of robbery and four of assault with a dangerous weapon.18 22 D.C.CODE §§ 502, 2901. It was alleged that, together with a companion, he boarded a D. C. Transit bus early on the morning of January 18, 1965, and held up the driver and three of the passengers.

Immediately after the robbery, the driver (Darby) drove the bus to the nearest police precinct and reported the crime. Detectives from the Robbery Squad were called. They interviewed the witnesses and showed some of them photographs. Three of the passengers (Raniecky, Charles, and Wilson) identified Clemons as one of the robbers from his photograph. One of these three (Raniecky) saw a collection of five photographs before making his identification. The two others were handed only the photograph of Clemons, although a police officer testified that he intended to pass along all the photographs, but the witnesses identified the first one. The bus driver, according to the testimony of a police officer, stated: "He would rather not see the pictures, it might get him confused. He would rather see the subject in person."

On February 4, less than three weeks after the robbery, Clemons was arrested and charged in a completely separate case. On February 11, the date set for the preliminary hearing on that charge,19 the police asked Darby to come to the U. S. Commissioner's office to see whether he could identify Clemons. Darby sat alone in the hearing room and was told to observe everyone who came in and out of the room. A detective testified as to the conditions in the Commissioner's office in this fashion:

"Was Mr. Darby sitting with you when he was brought in?

"He was not sitting with me. He was sitting in the hearing room.

* * * * * *

"I told Mr. Darby to sit in the hearing room and observe everyone that came in and out of the hearing room in front of the Commissioner.

"Were there a number of people coming in and out of the hearing room on that morning?

"Yes, sir, there was.

"Was it what you would call a rather busy place?

"I could say a normal day.

* * * * * *

"I would say it was, yes, sir, a little more than normal. It was quite crowded in there, with spectators and all."

Darby identified Clemons as the robber at that time.

On the day first set for trial, April 12, 1965,20 apparently after defense counsel had departed and without his knowledge, the prosecutor took the three witnesses who had identified Clemons' photograph the night of the robbery down to the courthouse cellblock where Clemons was being held. One of the witnesses testified that "they brought him to the screen," and that the prosecutor "asked me was it the same man." It is clear that the witnesses were shown only one man, and that they knew that this was the man the Government was proposing to try for the robbery. The three witnesses identified Clemons at this April cellblock show-up.

The case finally came on for trial on October 5, 1965, but only after the prosecutor had taken the opportunity during a one-day continuance of the trial from October 4, and again without the knowledge of defense counsel, to bring a fourth passenger witness (Redman) to the cellblock. This witness, who at the time of the robbery had been "looking at the gun more than at him," could not identify Clemons in the cellblock or at trial.

At trial, the Government's case in chief included the testimony of the six persons on the bus and that of two police officers who had investigated the crime. The bus driver, who had not seen photographs and who had been to the preliminary hearing, and the three passengers who had identified Clemons' photographs and who had attended the April cellblock viewing, positively identified Clemons in court. The passenger who had attended the October cellblock viewing was unable to identify Clemons. The fifth passenger, whom the robber had apparently ignored in going through the bus and who saw neither the photographs nor Clemons in person before trial, was also unsure Clemons was the robber, commenting that: "It sort of looks like him. I mean."

The bus driver also testified on direct examination with respect to the pre-trial identification made by him at the preliminary hearing. In cross-examining the first passenger witness (Redman) presented by the Government, the defense brought out the fact of the October cellblock confrontation. In presenting the succeeding witnesses, the Government brought out on direct examination the cellblock identifications made in April.

Elicitation by the defense of the October cellblock incident does not, in our view, constitute an opening up by it of the April viewing. Different witnesses were involved, and the October witness was unable to identify Clemons, in court or out. In addition, the defense before trial had, by reference to both incidents, challenged all eyewitness testimony by the persons involved therein, and the trial judge had ruled that evidence of both incidents would be admissible. It therefore cannot now be contended by the Government that, because the defense first introduced the October confrontation, the defense must be taken as having opened up the April identifications as well.

Thus, it could be argued that the Supreme Court's per se rule in Gilbert excluding out-of-court identifications violative of the Sixth Amendment right to counsel also applies to "due process" violations, and that this requires a reversal in this case absent a Chapman determination that the out-of-court identifications admitted at the Government's instance were harmless error. Stovall announced that the Sixth Amendment right to counsel was to apply only prospectively, and the related per se exclusionary rule, devised expressly to deter denial of the right to counsel in the future, does not appear to have been consciously formulated with reference to the due process problem of the non-lineup confrontation. Where, as here, the Government was not on notice either at the time of the confrontation or at trial that its conduct might be considered improper, it makes little sense to reverse a conviction on the basis of a per se exclusionary rule expressly intended to operate as a deterrent.21 Certainly if a determination can be made that the out-of-court identifications admitted into evidence here were justified by sources other than the cellblock confrontations and thus can be deemed to be reasonably reliable indicators of guilt, we do not feel bound, in the present state of higher authority, to give the Gilbert per se exclusionary rule retrospective operation in the due process area in a case where the pre-trial confrontation occurred before Stovall.

The record indicates that the in-court identifications made were positive and complete.22 The bus, at the time of the robbery, was well-lighted. The robber did not wear a mask or cover his face in any way, and spent an appreciable amount of time in the bus. Thus, the opportunity which the witnesses had to observe the robber was good. After one of the two police officers had testified as to the circumstances of the photographic identification23 and the aftermath of the robbery, and the other had corroborated Darby's identification at the preliminary hearing, the Government rested.

The defense attempted to establish an alibi based on the testimony of Clemons and a friend that they had spent the relevant morning in the company of a lady far from the scene of the crime. Clemons also claimed that his right hand had been severely injured on January 1, that he had been hospitalized January 9-13 for this injury, and that he would not have been able to handle a gun on the day of the robbery. However, the Government on rebuttal produced the lady in whose company Clemons claimed he had been on January 18, and she testified that she had not been with him at the time. The Government also produced hospital records to show that Clemons had not returned to the hospital to have his hand treated or examined after his original treatment, even though he had been directed to return for further care.24

The Stovall problems presented by this record seem to us to derive from (1) the pre-trial identification made by the bus driver at the preliminary hearing of Clemons on another charge and (2) the pre-trial identifications made by the three passengers at the April cellblock confrontation.25 As to the former, we have today sustained in Hines the finding by the District Court in that case that the preliminary hearing confrontation there involved did not violate due process. The facts with respect to Darby's pre-trial identification in this case are reasonably well developed in the record before us and are not in dispute. They show many similarities to those involved in Hines, especially with respect to the large numbers of people going in and out of the hearing room and the absence of suggestive influences brought to bear upon Darby. We do not think that this pre-trial confrontation, viewed in the totality of the circumstances, was a violation of due process, or that the matter need be pursued further. In reaching this result, we reiterate our observations in Hines about the risks inherent in the use of this identifying device as a standard substitute for the lineup.

In the matter of the April cellblock confrontation, we have today in both Clark and Hines refused to disturb District Court findings that the cellblock confrontations there involved were defective. If our rulings in those instances are sound, they point ineluctably, on the facts of this record, to the same result, again without the necessity of further inquiry in the trial court. The Government's own testimony establishes that, more than two months after Clemons was taken into custody and well after his indictment, the April cellblock confrontation was arranged by the prosecution; Clemons was singled out from the other prisoners for exhibition, and the question put to the witnesses in a group was whether he was the man; and the witnesses knew that he was the defendant in the trial, just continued, against whom they had been summoned to testify. These uncontested facts make the April cellblock viewing an a fortiori case as compared with Clark and Hines, and we treat it, accordingly, as constituting an improper procedure.

This does not, however, conclude the question of whether Clemons is entitled to reversal and a new trial. That question turns upon whether, in the language of Simmons, "the identification procedure was so unduly prejudicial as fatally to taint [Clemons'] conviction," a matter which, so said the Court, "must be evaluated in the light of the totality of surrounding circumstances." We note in this regard that he was identified in court by one witness, the bus driver, whose opportunity to observe the robber during the commission of the crime was excellent, and whose pre-trial identification we have found to be compatible with due process. But the same good opportunity to observe the robber at the time of the crime was also afforded to the three passenger witnesses who were involved in the defective April cellblock confrontation. One of these three, within minutes after the robbery (as contrasted with the more than two months intervening before the cellblock viewing), picked Clemons' picture out of a collection of photographs, and the others concurred in that choice.

We think that, certainly as to the first of these passenger witnesses, the record provides an independent source for his identification, and that we can justifiably regard that identification as not fatally tainted by the cellblock exposure. As for the other two passengers, although their photograph identifications, like those in Simmons, were made under "less than ideal" conditions, these identifications were positive and were based on an excellent opportunity to observe the robber during the robbery.

Thus, we conclude that the identifications of these three witnesses, in-court and out-of-court, were based on sources independent of the improper confrontations at the cellblock, and did not therefore create, in the language of Simmons, a "very substantial likelihood of irreparable misidentification." Accordingly, the conviction is affirmed.

Meanwhile, we note that Hines was continuously in custody from July 30, 1966 to June 3, 1968, the date on which the Government confessed error. Since his sentence was 18 to 54 months, he has served 22 months already ? 4 months more than his minimum term. Ten and one-half months elapsed from indictment to trial, some of which appears chargeable to him but not all. Appointed counsel in this court strove valiantly to get the transcript. The error committed by the prosecution in the trial makes further delay inevitable. This suggests that, when a case is already infected with delay, the consequences of prosecution error at trial may be greater than when new trials can be reached with reasonable dispatch. All these considerations will, we presume, be taken into account by the prosecution in deciding whether to press for a new trial on this indictment.

10 Hines has not questioned the legality of the photograph viewing underSimmons. Apparently the police, still trying to ascertain the identity of the robber, showed the ladies a variety of pictures. Mrs. Aggrey's testimony is somewhat weakened, however, by a police officer's testimony that only Mrs. Reid positively identified the defendant from the photographs.

11 Mrs. Grandy, although she said she recognized Hines immediately, did not tell the detective about her identification for some time thereafter

12 See note 10 supra. He testified that he had brought six photographs and had shown them to the ladies.

13 See Smith v. United States, 119 U.S. App.D.C. 272, 340 F.2d 797 (1964). In Mack v. United States, 150 A.2d 477 (1959), the D.C. Municipal Court of Appeals, however, expressed the opinion that prior out-of-court identifications should be admissible even where in-court identifications have not been impeached.

14 The remand order requested the District Court to hold an evidentiary hearing and to make findings of fact and conclusions of law with respect to the following:

  (1) Time, place, circumstances and legality of appellant's arrest.

  (2) Circumstances surrounding the pre-trial identification of appellant by the witness Alberstadt. Compare Stovall v. Denno, 388 U.S. 293 [87 S.Ct. 1967] (1967). See also United States v. Wade, 388 U.S. 218 [87 S.Ct. 1926] (1967); Gilbert v. California, 388 U.S. 263 [87 S.Ct. 1951] (1967).

  (3) Admissibility of evidence of pre-trial identification by the witness Alberstadt.

  (4) Admissibility of courtroom identification by the witness Alberstadt.

  (5) Circumstances surrounding the pre-trial identification of appellant by the witness Jones.

  (6) Admissibility of evidence of pre-trial identification by the witness Jones.

  (7) Admissibility of courtroom identification by the witness Jones.

15 We do not pursue the question of the legality of the arrest. Clark's argument essentially is that he was taken into custody for investigation by means of a ruse, and that identifications made during that custody are illegal fruits of that ruse. The District Court, however, explored in detail the somewhat unusual nature of Clark's relationship with the police, and concluded that Clark's presence in the police station prior to Alberstadt's appearance was wholly voluntary and unrestrained. Our reading of the record does not warrant any disturbance of the District Court's findings, or a rejection of its conclusions of law drawn from them, that Clark's arrest validly took place after Alberstadt's identification of him at the police station

16 1. Was the defendant the only individual that could possibly be identified as the guilty party by the complaining witness, or were there others near him at the time of the confrontation so as to negate the assertion that he was shown alone to the witness?

2 Where did the confrontation take place?3

3 Wise v. United States, 127 U.S. App.D.C. 279, 383 F.2d 206, 210 (1967). See also footnote 2 of Judge Bazelon's dissent in Wright, supra, at p. 1262 [Wright v. United States, 404 F.2d 1256 (D.C.Cir. Jan. 31, 1968)].

3 Were there any compelling reasons for a prompt confrontation so as to deprive the police of the opportunity of securing other similar individuals for the purpose of holding a lineup?4

4 See Stovall, supra, note 1, 388 U.S. at 302 [87 S.Ct. 1967].

4 Was the witness aware of any observation by another or other evidence indicating the guilt of the suspect at the time of the confrontation?

5 Were any tangible objects related to the offense placed before the witness that would encourage identification?5

5 See Palmer v. Peyton, 359 F.2d 199, 201 (4th Cir. 1966).

6 Was the witness' identification based on only part of the suspect's total personality?6

6 Ibid.

7 Was the identification a product of mutual reinforcement of opinion among witnesses simultaneously viewing the suspect?

8 Was the emotional state of the witness such as to preclude objective identification?

9 Were any statements made to the witness prior to the identification indicating to him that the police were sure of the suspect's guilt?

10 Was the witness' observation of the offender so limited as to render him particularly amenable to suggestion, or was his observation and recollection of the offender so clear as to insulate him from a tendency to identify on less than a positive basis?

17 It could be argued, although it has not been, that defense counsel might, as a matter of trial tactics, have elected not to inquire into the circumstances of the cellblock confrontation if the District Court had already ruled, in the course of a hearing out of the presence of the jury, that Jones' in-court identification was not tainted by the cellblock incident. This is, however, a very speculative basis for requiring a new trial, and we do not do so. There was in this case no pre-trial motion to suppress eyewitness testimony, and no request during trial that the jury be excused for an evidentiary inquiry into the circumstances of pre-trial identification

18 He appealed his conviction to this court, and a panel affirmed the conviction by order. Thereafter the panelsua sponte vacated its judgment and ordered that the case be reheard in light of the Supreme Court's decision in Stovall. After rehearing by the panel, but before a decision had issued, the case was placed en banc for consideration by the full court.

19 Clemons would have us reverse his conviction because he never received a preliminary hearing for the offense here charged. Instead, while he was being held on the first charge for which he had received a hearing, he was indicted by the grand jury for the offense with which we are here concerned. The mere failure to hold a preliminary hearing where an indictment is brought by the grand jury, however, is not error. Walker v. Rodgers, 128 U.S.App.D.C. 420, 389 F.2d 961 (1968)See generally Ross v. Sirica, 127 U.S.App.D.C. 10, 380 F.2d 557 (1967); Crump v. Anderson, 122 U.S. App.D.C. 173, 352 F.2d 649 (1965).

20 The trial was continued from April to October, on defense counsel's motion, pending a mental examination of defendant

21 It is, of course, true that a conviction obtained by means which fail "to meet `those canons of decency and fairness' established by the fundamental law of the land" has always been subject to due process attack. But Justice Douglas as recently as inBiggers has said that "[o]f course, due process is not always violated when the police fail to assemble a lineup but conduct a one-man showup." And it is blinking reality to regard Stovall, in language if not in result, as doing other than giving a significantly new dimension to the due process concept in relation to some police practices which have been long and widely engaged in without constitutional challenge.

22 There was, however, a minor inconsistency in the testimony as to whether the robber had worn a sweater or a shirt

23 No claim is made that the display which the officer showed one passenger was unfair. The only challenge posed is to the fact that he then passed along Clemons' picture (which the original viewer had picked out) first and, after the other witnesses had recognized the picture as well, never passed along the others

24 Clemons complains of the refusal of the trial court to allow the jury to hear the testimony of a taxicab driver that had identified Clemons as the man who had robbed him in his cab on a specific date in 1964. This testimony ? which was heard by the court out of the presence of the jury ? when coupled with the defense showing that Clemons was serving a 30-day sentence in Occoquan for public intoxication on the day of that robbery, was offered as illustrating the general fallibility of eyewitness identification. The court ruled that this evidence of an unrelated matter would only confuse or prolong the proceedings. The question of admissibility in this instance fell within the area of the trial judge's discretion, and we find no abuse warranting reversal

25 The October cellblock viewing is of no consequence because the Government reaped no advantage from it

LEVENTHAL, Circuit Judge (concurring):

I join in the opinion of the court. Because the Clemons case is one that has prompted particular reflection on my part ? I have been pondering it since I first joined, with unexpressed reservation, in the July 1967 order affirming without opinion ? I hope I may be permitted to try my hand at a personal statement of what I regard as the appropriate doctrinal context.

My opinion is not intended to undercut the court's opinion in Clemons, though I recognize that the other judges concurring in Clemons may not agree with all or most of what I say here.

A "totality of the circumstances" test is appropriate in these pre-Wade, pre-Stovall identifications where we are dealing with a "due process" claim, and not a claim of denial of effective assistance of counsel or of other specifically enumerated rights in the Bill of Rights.

I further think it appropriate ? particularly in cases where the trial as well as the identification preceded Wade, Gilbert and Stovall ? that the court assess not only the "totality of circumstances" surrounding the challenged identification, but also the totality of circumstances at trial concerning the issue of identification, i. e. the totality of proof and other factors bearing on the likelihood of misidentification. The mere fact that some "unreliable" identification testimony was received does not establish a denial of the due process right to a fair trial.1 That right is to be tested by assessing the totality of proof on the identification issue. The soundness of making a due process right depend on a review of evidence becomes clear when one focuses on the interests protected retroactively by the Simmons-Stovall due process right.2 Unlike the interests preserved by the guarantee of counsel or some of the other specific guarantees, the interests guarded by Stovall cannot so easily be factored out and assessed independently of the evidence.

In essence what the Stovall due process right protects is an evidentiary interest. Identification testimony is a particularly important kind of evidence, so important as to occasion exclusionary rulings that are sweeping in prospective application.3 But we are dealing here with the right to a retroactive condemnation of a pre-Wade admission of the identification evidence.

It is part of our adversary system that we accept at trial much evidence that has strong elements of untrustworthiness ? an obvious example being the testimony of witnesses with a bias. While identification testimony is significant evidence, such testimony is still only evidence, and, unlike the presence of counsel, is not a factor that goes to the very heart ? the "integrity" ? of the adversary process.

Counsel can both cross-examine the identification witnesses4 and argue in summation as to factors causing doubts as to the accuracy of the identification ? including reference to both any suggestibility in the identification procedure and any countervailing testimony such as alibi.

Since the interest protected is in essence an evidentiary one, denial of the retroactive right recognized in Stovall and explicated in Simmons5 must depend on the assessment of all the factors and evidence bearing on the identification issue. The presence of other untainted identification testimony, counsel's opportunity to inquire into the circumstances of the challenged identification, and to bring out the facts, are all relevant to the consideration of overall due process fairness to the accused.

In post-Wade trials the situation may be different since the alerted court and prosecutor could take steps to confine the trial to unquestionable identification testimony. But when a trial, as well as the identification, has taken place prior to Wade and Stovall, the fact that some of the testimony would raise problems, if independently considered, should not be required to undo the trial, provided that essential fairness is pointed to by identification of defendant by evidence that is substantial and reasonable, and that the defense had an adequate opportunity to probe the relevant factors.

I am not certain whether or to what extent the foregoing differs in net effect from the "traditional" "harmless error approach" proposed by the dissent. I suspect any differences in result would reflect not differences of standards so much as differences in judges applying them. But differences in the formulation of standards are sometimes significant in clarifying the role and direction of the trial courts called upon to apply the standards. That leads me to an observation that these matters are properly to be resolved by the trial judge, at least in the first instance, and that appellate resolution marks the rare exception.

I am authorized to say that Judge BURGER joins in this opinion.

If the retroactive due process right condemned all trials at which the court received "tainted" identification testimony, it would then be necessary to focus on the case in terms of the "harmless error" test ? including whether there is any basis under the Constitution for relaxing the harmless error approach set forth in Chapman v. California, 386 U.S. 18, 87 S.Ct. 824, 17 L.Ed.2d 705 (1967).

2 It is important to bear in mind that when we deal with problems of identification evidence in pre-Wade trials, there is no point in being concerned with undesirable police conduct unless it produces evidence which is likely to result in an injustice or is so egregious that it is unseemly to allow a conviction to stand. Cf. Miller v. Pate, 386 U.S. 1, 87 S.Ct. 785, 17 L.Ed.2d 690 (1967); Napue v. Illinois, 360 U.S. 264, 79 S.Ct. 1173, 3 L.Ed.2d 1217 (1959).

The Constitution has always been concerned with the quality of evidence and other evidentiary matters. We do not allow convictions to stand when there is no reasonable evidence at all. Yet that determination must essentially be made by assessing the record and the testimony. Cf. Thompson v. City of Louisville, 362 U.S. 199, 80 S.Ct. 624, 4 L.Ed.2d 654 (1960).

3 See Gilbert v. California, 388 U.S. 263, 272-274, 87 S.Ct. 1951, 18 L.Ed.2d 1178 (1967); Stovall v. Denno, 388 U.S. 293, 296, 87 S.Ct. 1967, 18 L.Ed.2d 1199 (1967).

4 In Simmons v. United States, 390 U.S. 377, 385, 88 S.Ct. 967, 972, 19 L.Ed.2d 1247 (1968), the Court pointed out: "Notwithstanding cross-examination, none of the witnesses displayed any doubt about their respective identifications of Simmons."

5 Simmons v. United States, 390 U.S. 377, 88 S.Ct. 967 (1968)

BAZELON, Chief Judge:

I concur in Judge McGOWAN'S opinion except as to its treatment of the witness Mrs. Steele in Hines v. United States. On the latter point I agree with Judge WRIGHT that the record cannot support a finding of an independent source for her in-court identification.

J. SKELLY WRIGHT, Circuit Judge (concurring in part and dissenting in part):

I agree with much of Judge McGOWAN'S well reasoned and well written opinion. The general statement of the issues certainly delineates the contours of the Stovall1 problem, and the treatment of the circumstances surrounding each case is admirable. In my view, however, the discussion in Clemons casts the analysis of pre-Stovall cases into an awkward and somewhat nebulous mold. I think that analysis, and the result reached in the case, give too little weight to the fact that unnecessarily suggestive confrontations are constitutional violations, and testimony concerning them is constitutional error; accordingly I dissent in that case. I concur fully in Clark, and I concur in the treatment of Hines with the one exception hereinafter noted.

Clemons v. United States

The evidence against Malcus Clemons included his identification by three witnesses ? Wilson, Charles and Raniecky. Wilson and Charles identified Clemons' picture immediately after the robbery; they were shown only Clemons' picture. All three subsequently viewed Clemons alone in his cellblock. At trial all three testified on direct examination to the cellblock confrontation, and each made an in-court identification of Clemons. I would:

  (1) Hold as a matter of law that the cellblock viewing by all three and the picture identification by Charles and Wilson were unnecessarily suggestive confrontations, violative of due process. Therefore, the admission of testimony relating to the cellblock viewing and the admission of any in-court identifications tainted by the pre-trial confrontations was constitutional error.

  (2) Remand to the trial court to determine whether the testimony about the cellblock viewing was harmless error under the traditional harmless error standard, as enunciated in Kotteakos v. United States, 328 U.S. 750, 66 S.Ct. 1239, 90 L.Ed. 1557 (1946). That standard is whether the judge can say with fair assurance (reasonable probability) that the verdict of the jury was not substantially swayed by the error.

  (3) Remand to the trial court to determine whether the in-court identifications were tainted by the suggestive pre-trial confrontations. The burden here is on the Government to show by clear and convincing evidence that the in-court identifications had an independent source.

  (4) Remand to the trial court to determine whether, if any of the in-court identifications were tainted, their admission was nonetheless harmless error under the traditional Kotteakos test.

If the trial court cannot find that the admission of testimony about the cellblock viewing and the admission of any tainted in-court identification was harmless, then a new trial is required.

This case arose before the Wade2-Gilbert3-Stovall trilogy. Had it come afterward, the majority would apply an analysis similar to the one just outlined. The majority recognizes that the sequence outlined in Gilbert for dealing with violations of the right to counsel will apply in the future for Stovall violations of due process as well. Specifically, this would mean: First, the circumstances of each pre-trial confrontation would be examined to determine whether it was unnecessarily suggestive. The majority goes over many of the relevant factors to be considered (see, e. g., footnote 16). Second, if a confrontation is found to be suggestive, the burden shifts to the Government to show by clear and convincing evidence that an in-court identification was the product of an independent source, untainted by the invalid confrontation. And last, if the tainted identification was allowed, or testimony elicited as to an invalid confrontation, the reviewing court would determine whether the error was harmless. Here, for future cases, the very strict standard of harmless error used in Chapman v. California, 386 U.S. 18, 24, 87 S.Ct. 824, 828, 17 L.Ed.2d 705 (1967), would apply ? the conviction is reversed unless the court is "able to declare a belief that it was harmless [error] beyond a reasonable doubt."4

For pre-Stovall cases I would follow this approach as well, with one exception. Because prior to Wade the police and prosecutors were on less notice of the dangers inherent in lineups and other confrontations, and because there may be large numbers of people who can raise a colorable claim that they were subjected to unnecessarily suggestive procedures. I would not apply the strict Chapman test. Even though the error is constitutional, out of recognition of the police notice consideration, and in the interest of not unduly disrupting the efficient administration of the criminal process, I would apply the less stringent traditional Kotteakos harmless error test.5

The majority opinion goes a different route. It abandons the three-step analysis in pre-Stovall cases and applies instead an "independent source" test for all confrontations, pre-trial as well as in-court. Thus the majority holds it was not error to have admitted even direct testimony about a "defective" confrontation (such as the cellblock viewing in Clemons) if the court can find that the witness had other, non-suggestive, chances to view the defendant. This approach makes it unnecessary, in Clemons, to reach the question of harmless error.

Conceptually, the majority approach is sound. However, I believe that it suffers from two drawbacks. First, there is obviously a need here for guidelines for the trial courts. I think it is more clear to keep the same analysis for all cases, pre- and post-Stovall, changing only the harmless error test. This breaks down the inquiry into discrete steps. The majority, by introducing a different approach for pre-Stovall cases, introduces a needless confusion in this area.

Second, and much more important, I think the majority approach sells short the gravity of the problem. The approach I have taken first focuses on the confrontation. If it was unnecessarily suggestive, then it is a violation of due process, and the introduction of testimony about that confrontation is error ? constitutional error. Recognizing it as such, I would then have the court determine whether it permeated the trial to an extent sufficient to warrant reversal. The majority cuts off the inquiry at its source, because even as suggestive a procedure as the cellblock viewing in Clemons is redeemed if the court can find an independent source for the identification made at it. If this is so, I can see no reason why the same should not be true in post-Stovall cases; an "independent source" means that it is likely that they got the right man, and if the independent source is all the inquiry that needs to be made in a Stovall problem then the same approach should apply in all cases. Yet the majority will not apply the same approach in post-Stovall cases, no doubt because it feels that the Supreme Court in Stovall had in mind a more searching inquiry along the lines set out in Gilbert, one which views with a more jaundiced eye a suggestive pre-trial confrontation. This, in my judgment, is the approach that should be taken in pre-Stovall cases as well.

I recognize the elusiveness of capturing an attitude in words. However, my difference with the majority's approach is more than a matter of semantics. In its application to this case the difference is substantial. For I cannot see how, on this record, we can rest easy with the conviction of Malcus Clemons without further inquiry into the disturbing identification problems so clearly raised. I think that, by applying my approach to the facts in Clemons in more detail, the import of the test I have formulated will become more clear.

A. Suggestibility of the pre-trial confrontations.

Subsequent to a robbery of the driver (Darby) and passengers (including Charles, Wilson and Raniecky) of a bus, these four witnesses identified Clemons as one of the robbers at various pre-trial proceedings and at the trial itself. Raniecky, Wilson and Charles viewed Clemons at a one-man showup which the court finds, on the Government's own testimony, to be an a fortiori violation of Stovall, given the treatment of similar showups in Hines and Clark. In addition, each identified Clemons' picture immediately after the robbery, Raniecky from a group of five pictures, and Charles and Wilson after being shown only Clemons' picture. As with one-man lineups, one-man picture viewings, especially where the officers had other pictures readily available, are also to be viewed suspiciously. I would find that the picture identifications by Charles and Wilson were also unnecessarily suggestive. See Simmons v. United States, 390 U.S. 377, 88 S.Ct. 967, 19 L.Ed.2d 1247 (1968).

Darby did not participate in the picture viewing or the cellblock confrontation. He did pick Clemons out of a room in which Clemons was being presented for a preliminary hearing. I agree with the court's cautious affirmance of the validity of that procedure in this case. Because of the invalid cellblock confrontation and the suggestive picture viewing, however, the burden shifts to the Government to prove an independent source for the in-court identifications by Raniecky, Charles and Wilson.

B. The issue of taint.

For Charles and Wilson the only independent source for their identifications could be the robbery itself. As the court notes, the bus was well lighted, and I agree that the opportunity which these witnesses had to observe the robber may have been good. However, we know very little of what observations each of them actually made. I note that two passengers, despite the good opportunity, were unable to identify Clemons. Nor do we know much about the description of the robber that Charles and Wilson were able to give the police. For example, the scant testimony bears the inference that Charles did not notice much about the robber except that he was definitely wearing either a sweater or a shirt. Under these circumstances,6 there should be a remand to determine whether either Charles' or Wilson's observations at the time of the robbery were sufficiently clear to provide an independent source for their in-court identifications. The trial court has not made this determination and I doubt that we should make it, particularly on a record where the issue has not been fully developed.

I would be unwilling to order a remand solely on the basis of the testimony of Raniecky, because he was able to pick Clemons out of five pictures. However, since there should be a remand for taint as to Charles and Wilson, and since we know equally little about Raniecky's observations at the time of the robbery (we do know that he sat in the front of the bus and did not look at the robber as he passed down toward the rear of the bus), the judge on remand should make an "independent source" determination for Raniecky as well.

C. Harmless error.

Raniecky, Charles and Wilson all testified on direct examination to the invalid cellblock showup. In addition, as noted above, the in-court identifications by Charles and Wilson, and to a lesser extent Raniecky, may have been tainted by their pre-trial confrontations. Thus in a case based solely on identification, three of four witnesses may have given tainted and impermissible testimony. To me this raises a grave and close question whether the conviction can stand. But despite the testimony as to the cellblock showing, and despite the lack of knowledge as to taint, the majority, applying its taint test on an incomplete record, affirms Clemons' conviction. I would leave it to the trial judge in the first instance to develop and weigh the factors going to taint, and to consider the issue of harmless error under the standard: was it reasonably probable that the jury was swayed by the invalid testimony? Accordingly, I would remand.

Hines v. United States

I concur in the court's disposition of Hines with the exception of one witness, Mrs. Florence Steele. I agree, of course, that for each witness:

"* * * The court should then, on facts elicited outside the presence of the jury, rule upon whether a pre-trial identification by the same eyewitness is violative of due process. If a violation is found, the court should then decide whether the in-court identification is still admissible because it has an independent source * * *."

In the case of the witness Florence Steele, the facts as developed by the trial court in the Stovall hearing7 in my judgment require the conclusion that her testimony identifying the defendant should not have been admitted. Therefore, if a new trial is held, she should not be allowed to so testify.

The first and only time Mrs. Steele identified the defendant before the trial was at the cellblock confrontation where Hines, locked alone in his cell, was displayed by the prosecutor to four of the ladies who knew he had been charged with the crime. The trial court, and this court, aptly characterize this proceeding as bringing Hines "into a focus unacceptably antithetical to the lineup principle." This confrontation was thus ruled invalid under Stovall.

Since Mrs. Steele did not see any pictures, or have any other opportunity to identify Hines, the only "independent source" for her identification could be the robbery itself. She testified that the robbery lasted three to four minutes, and that she saw Hines enter the shop and also looked at him as she handed him some money. However, her testimony is severely undercut by her statements that:

"I had a slight blackout.

* * * * * *

"I was looking at his face when I was sitting in the chair. I felt this seizure coming on ? this blackout coming on and I just slumped into the chair."

She did pass out, and was revived by Mrs. Reid, who rubbed her face with a wet towel.

The description of the robber which Mrs. Steele gave the police does not demonstrate that she had an adequate independent source for her identification. Following the robbery she described him to the police as a "small faced, dark fellow" wearing "khaki pants or something and this dark straw hat," either "dark blue or dark gray"; further, the robber had on a "soiled sweat-shirt," later described as a soiled "T shirt." From this very general description it is hard to conclude that she carefully observed and remembered what the robber looked like. And even in this description she failed to tell the police, as she later testified, that the robber was wearing sunglasses.

Finally, at the Stovall hearing she testified that she remembered the robber because "[h]e has a very unusual face," "I mean the structure of his nose and lips and eyes and hairline." However, no one else testified that Hines' face was unusual. And whether Mrs. Steele actually noticed the eyes and hairline of a robber who wore sunglasses and a hat is doubtful.

At the Stovall hearing Mrs. Steele stated that she "certainly" could have recognized the robber even without having seen him at the cellblock confrontation. The trial court relied on her credibility in allowing her testimony. In upholding the trial court, this court states:

"* * * [A]lthough the positiveness of the witness about an independent base for an in-court identification is a relevant factor, it is to be weighed warily and in the realization that the most assertive witness is not invariably the most reliable one. * * *"

I would agree that the positiveness of a witness is a factor to be considered, and weighed warily, by the trial judge, and his consideration is to be given weight by this court on review. Further, since it is relevant, in a close case it can be controlling. We should not, though, give the trial court unreviewable discretion in the matter. However little guidance Stovall gives us, it does place upon us a responsibility to review an identification as a matter of law. Here the case was not close with regard to Mrs. Steele. Her sole pre-trial identification was based on an unnecessarily suggestive confrontation, and her independent source for identification in the robbery itself was at best limited. Thus however much the trial judge believed her, as a matter of law her testimony should not have been admitted. Since the Government's confession requires a new trial, we do not have to decide whether her erroneously admitted testimony would require reversal; if a new trial is held, however, in my judgment her testimony should be excluded.

"* * * We are not concerned here with whether there was sufficient evidence on which the petitioner could have been convicted without the evidence complained of. The question is whether there is a reasonable possibility that the evidence complained of might have contributed to the conviction. * * *"

The Chapman Court then equated the "reasonable possibility" with a belief "beyond a reasonable doubt."

5 This standard was described by the Court as follows:

"* * * [T]he question is, not were they [the jury] right in their judgment, regardless of the error or its effect upon the verdict. It is rather what effect the error had or reasonably may be taken to have had upon the jury's decision. The crucial thing is the impact of the thing done wrong on the minds of other men, not on one's own, in the total setting. * * *

* * * * *

"If, when all is said and done, the conviction is sure that the error did not influence the jury, or had but very slight effect, the verdict and the judgment should stand * * *. But if one cannot say, with fair assurance, after pondering all that happened without stripping the erroneous action from the whole, that the judgment was not substantially swayed by the error, it is impossible to conclude that substantial rights were not affected. The inquiry cannot be merely whether there was enough to support the result, apart from the phase affected by the error. It is rather, even so, whether the error itself had substantial influence. If so, or if one is left in grave doubt, the conviction cannot stand." 328 U.S. at 764-765, 66 S.Ct. at 1248. (Citations omitted.)

6 It should be borne in mind that the trial inClemons was conducted before Stovall and that consequently counsel and the trial court understandably did not focus with particularity on Stovall considerations.

7 UnlikeClemons, the trial court in Hines did conduct a Stovall hearing.