United States v. Franklin Perry
449 F.2d 1026
D.C. Cir.
1971
Check Treatment

UNITED STATES of America v. Franklin PERRY, Appellant.

No. 22469

United States Court of Appeals, District of Columbia Circuit.

Argued Sept. 25, 1969. Decided June 1, 1971.

448 F.2d 1026

those who were covered by the original Sexual Psychopath Act and who were “not insane” but were “mentally ill.” To the extent that the “mentally ill” standard of the 1965 Hospitalization Act is applied to the Sexual Psychopath Act and limits the application of that law by excluding persons originally defined as sexual psychopaths, I would not release any person within that group from his commitment as a sexual psychopath (not insane) unless he were found to be not mentally ill and commitable under the new definition. I believe that to be the true intent of Congress. Certainly Congress in passing the Hospitalization Act did not intend to bring about a wholesale release of sexual psychopaths. In this connection it should also be pointed out that while both standards encompass persons likely to injure others and authorize their confinement until cured, the “injury” standards of the two statutes are not completely complementary. The sexual psychopath definition is broader in that in addition to persons who are likely to injure others it also includes persons who are “likely to attack or otherwise inflict * * * loss, pain, or other evil on the objects of his desire” (emphasis added) and provides for the mandatory commitment of such persons (D.C.Code § 22-3508). However, commitment of those mentally ill under the Hospitalization Act is discretionary (D.C.Code § 21-545(b)).

This portion of the wider definition in the Sexual Psychopath Act is particularly applicable to that very small group of persons who have an uncontrollable propensity to commit sexual attacks and otherwise engage in perverse and other sexual type assaults that some people consider do not cause visual physical “injury.” Some people argue that many of such sick type acts do not cause “injury.” In this respect, they argue for a very narrow definition of “injury” which in many instances would exempt them from the Sexual Psychopath Act because they are mentally ill, and would also not confine them under the discretionary authority conferred by the Hospitalization of the Mentally Ill Act. To release this group of people from their commitment as sexual psychopaths because they may be said to be mentally ill (and possibly not considered commitable because they were not likely to cause actual physical injury to others in the more limited sense of the mental illness definition) would be a complete negation of the true intent of Congress in the Sexual Psychopath Act. Consequently, those persons who meet the definition of being mentally ill and who under the discretion provided by the hospitalization statute are not to be confined, if they are likely to cause injury under the broader standard of the original Sexual Psychopath Act, I would continue their commitment under the Sexual Psychopath Act. Otherwise, they would be relegated to prosecution as minor criminals and the sentences for the normal offenses in such cases are all too frequently in the misdemeanor category. These misdemeanor statutes, and even some felony statutes, do not adequately consider or treat the exceedingly dangerous propensity of the sexual psychopath in conferring punishment. Such persons should not be considered as criminals but as patients and as such they should be confined until cured. I believe that to have been the intent of Congress in legislating in this area.

Mr. Peter C. Jenkins (appointed by this court) for appellant. Mr. John A. Terry (appointed by this court at the time the brief was filed) was on the brief for appellant. Mr. William W. Greenhalgh (appointed by this court) also entered an appearance for appellant.

Mr. Allan C. Fork, Atty., Department of Justice, with whom Messrs. Thomas A. Flannery, U. S. Atty., and Roger E. Zuckerman, Asst. U. S. Atty., were on the brief, for appellee. Mr. David G. Bress, U. S. Atty., at the time the record was filed, also entered an appearance for appellee.

Before BAZELON, Chief Judge, FAHY, Senior Circuit Judge, and ROBINSON, Circuit Judge.

SPOTTSWOOD W. ROBINSON, III, Circuit Judge:

The sole issue on this appeal, from convictions of robbery1 and simple assault,2 is whether a near-the-scene identification confrontation between a suspect without a lawyer and the victim, within an hour to an hour and a half after commission of the offenses,3 violated the suspect‘s Sixth Amendment right to counsel4 or his Fifth Amendment right to due process of law.5 The District Judge, after a hearing, held that no such violation was established, and permitted an in-court identification of the suspect, our appellant, and testimony as to the out-of-court identification at the trial. After careful study of the problem—one residing in a difficult and gray area of the law—we reach the same conclusion, and accordingly affirm appellant‘s convictions.

I

The offenses in suit occurred on a Sunday morning between 11:30 o‘clock and 12:00 noon as William Cook was returning home from a grocery store.6 Two men approached him from behind; one man put his arm around Cook‘s neck, and the other came around in front to face him with a knife. The first, whom Cook later identified as appellant,7 then threw Cook to the ground and the other rifled his pockets, removing approximately three dollars. At this point, a car stopped nearby and an occupant inquired as to whether a robbery was in progress, whereupon the two men fled. Cook then returned home and called the police.

In reporting the robbery, Cook gave full descriptions of both men, including such details as ages, heights, weights and the fact that one wore a blue zippered waist-length jacket and blue trousers. The descriptions were flashed over the police radio network, and about 12:20 p. m. were noted by Officer Willie L. Polk while on duty in the robbery area. Officer Polk recalled that earlier that day he had seen a man—appellant, as it turned out—matching one of the descriptions. The officer then began a cruise of the area, and shortly thereafter spotted appellant and another man fitting the other description standing among a group only a half-block from the robbery scene. Appellant was wearing a blue jacket and blue trousers, and his physical features approximate those of the man described in the broadcast as having been so attired.

Officer Polk took the two men into custody8 about 12:45 p. m. and immediately transported them to Cook‘s house, a distance of about four blocks. Cook was asked to look at the men, the officer stating that they matched the descriptions Cook had given; aside from this statement, there was no suggestion to Cook that the police suspected either man. Cook came outside, viewed both men, and identified appellant as one of the robbers.9 It was then about 1:00 p. m., no lawyer was present, and the record is fuzzy as to the extent to which appellant may have been advised of a right to counsel.10

It is this confrontation that appellant attacks and, since it occurred after the Supreme Court‘s Wade-Gilbert-Stovall pronouncements,11 the principles delineated in those decisions were fully operative.12 Upon a defense motion to suppress, the District Judge conducted a hearing, at which Cook and Officer Polk both testified, and thereafter rejected appellant‘s Wade-Gilbert and Stovall approaches.13 This ruling opened the door to Cook‘s in-trial identification of appellant as one of the robbers, and testimony by Cook and Officer Polk as to the identification at Cook‘s home.

Appellant‘s defense rested on his testimony that, while walking along the street, he came upon Cook arising from the ground, a bag of groceries lying nearby. Cook, appellant testified, then said “[y]ou are the one” and admonished appellant to “wait until I get back.” Appellant testified further that he continued to walk on for another half-block to a corner where he stayed until he was arrested, and denied any connection whatever with the robbery. As we have mentioned, however, the jury thought otherwise and convicted.

II

The Wade-Gilbert-Stovall trilogy articulated constitutional standards governing identification procedures by which the criminally suspect may become the criminally accused.

and
established the suspect‘s Sixth Amendment right to counsel at an identification confrontation;14
established his right to freedom from a confrontation “so unnecessarily suggestive and conducive to irreparable mistaken identification” as to amount to a deprivation of Fifth Amendment due process of law.15 Nonsatisfaction of these standards bars implicated witnesses from subsequent in-court identifications not flowing from independent sources,16 and from testimonial reference to prior illegal identifications.17

As we have indicated, appellant weaves a thesis for reversal from strands of all of these constitutional doctrines. He was without the benefit of counsel, he points out, when he was identified by Cook at the latter‘s home; that identification, he argues, occurred under circumstances so highly suggestive of his complicity in the robbery as to impinge upon due process. Therefore, so the argument runs, Cook‘s in-court identification, and Cook‘s and Officer Polk‘s testimony as to Cook‘s out-of-court identification, were improperly received at his trial, with the result that his conviction cannot stand. We first take up appellant‘s right-to-counsel contention, and then his due process claim.18

In

,19 we sustained, against a Wade-Gilbert argument, an on-the-scene identification confrontation between the victim and a singly-presented suspect transpiring quite shortly after the crime. We noted that “prompt on-the-scene identifications” were not precisely covered by the
and
holdings,20 which involved post-indictment lineups;21 we observed that in
and
“the

Court was evidently focusing primarily on the routine lineup and show-up procedure employed by the police to obtain evidence for use at trial.”22 “In these typical cases,” we continued, “where counsel has been retained and time is not a factor, [the Court] could find ‘no substantial countervailing policy considerations * * * against the requirement of the presence of counsel.‘”23 We found that in the circumstances of the case then before us “there would necessarily be a long delay in summoning appellant‘s counsel or a substitute counsel to observe the formal lineup;”24 “[s]uch delay,” we said, “may not only cause the detention of an innocent suspect; it may also diminish the reliability of any identification obtained, thus defeating a principal purpose of the counsel requirement.”25 We concluded:

Balancing all the doubts left by the mysteries of human perception and recognition, it appears that prompt confrontations in circumstances like those of this case will “if anything promote fairness, by assuring reliability * * *.”26 This probability, together with the desirability of expeditious release of innocent suspects, presents “substantial countervailing policy considerations,” which we are reluctant to assume the Supreme Court would reject. We therefore conclude, with some hesitation, that Wade does not require exclusion of [the appellant‘s] identification.26

III

On the right-to-counsel issue,

differs essentially from the instant case only with respect to the length of the time interval between the offenses and the identification confrontation.27 In
, it was a matter of about 30 minutes;28 here it was a period of an hour to an hour and a half.29 The problem at hand, then, is to determine which ambit—Wade‘s or Russell‘s—the present case falls within. The Government urges us to define the scope of the
holding in terms of a “reasonableness” test involving the circumstances

surrounding such a confrontation.30 Appellant, on the other hand, would have us confine

within the narrowest limits of the situation there presented—a confrontation “within minutes” after the observed offense.31

While a fixed time limit would reduce, for police and courts alike, uncertainties in

implementations,32 the legal test must surely be more flexible. As
instructs, the redeeming virtue in on- or near-the-scene identifications is their freshness as a factor promoting their reliability, and consequently their fairness.33 Time is, of course, of the essence, for the trustworthiness of an eyewitness identification, dependent as it is on sharpness of memory, is apt to lessen with the passage of time.34 But the pace of memory-dimming may be reflected by other circumstances—those which may either strengthen or weaken an ensuing identification35—and those circumstances vary from case to case. We can no more say that every identification within any particular fixed period is assuredly reliable than we can say that every identification beyond that period is assuredly unreliable. We reject, then, a purely artificial time approach to confrontations of the type under review, and favor instead a standard which gives other factors bearing upon the fidelity of the identification their just due.36

We have recognized three distinct interests toward which prompt on- or near-the-scene identifications may wholesomely contribute. The first, and dominant, is that identifications occurring shortly after the observations made when the crime was committed possess a greater likelihood of accuracy because of their freshness, while a postponed identification could be detrimental to that end. This consideration contributed significantly to our holdings in

and other similar cases,37 as it has to a number of pre-Wade confrontations in which a degree of suggestivity was questioned on due process grounds.38 Time may warp the memory,39 and there is no way of telling at what point or to what extent this disintegration takes place in particular humans.40 Delay incidental to the arrangement of a formal lineup in the presence of counsel thus becomes important in assessing whether an earlier identification procedure, though without counsel, is constitutionally justified.41

Once the time factor, in the surrounding circumstances, enhances the trustworthiness of a near-the-scene identification, two other interests also come to the fore. These are the combined needs for fast and effective police action in apprehending criminals, and the avoidance of unnecessary inconvenience and embarrassment to innocent citizens incidental thereto.42 By such confrontations, a suspect not identified as the perpetrator of the crime is exonerated forthwith without being hauled to a police facility for a lineup, and possibly without suffering an arrest. Simultaneously, the police can quickly ascertain the necessity for any further investigation and search for the real culprit. Such confrontations, then, serve also to protect the innocent, and to pro-

mote efficiency in the deployment of limited police resources.43

We do not, however, in summarizing the interplay of the two last-mentioned interests, intimate that apart from the freshness of an identification they are constitutional justifications for confrontations which omit the procedure prescribed by Wade. The foundation of our

decision was an early confrontation aiding fairness by improving the reliability of the identification; the “expeditious release of innocent suspects” then became an additional reason for dispensing with the necessity for counsel upon an identification quite independently deemed reliable because it was fresh.44 We find the elements of the
foundation present in the case at bar, and thereon we rest our decision.45 It is in the process of doing so that other interests to which we have given recognition are also subserved.

IV

The one circumstance differing here from

, we repeat, is the time lapse—about 30 minutes, as against an hour to an hour and a half—between the identifying witness’ initial confrontation with his assailants and the confrontation at which appellant was identified as one of them. In all other material respects the cases are alike; only the longer time interval here could upset the result which
held to be proper. In our view, that single difference is not enough to produce a different outcome.

The truth of the matter is that we have not limited

to confrontations occurring within moments of the criminal episode. Rather, we have imparted to the time factor a modest degree of elasticity, thus accommodating the flux of ever-changing surrounding circumstances. In both Wade- and Stovall-type situations, we have recognized that confrontations delayed somewhat more than momentarily may still meet constitutional requirements although some amount of delay was inevitable.46 In some instances, of both types, where the duration of the delay approximated

that here, we have nonetheless found the confrontation justified.47 As we look to the events transpiring in the instant case, we find the justification satisfying.

The assault and robbery were committed in broad daylight. The episode, by Cook‘s estimate, occupied a space of about two minutes. Cook was able to observe his assailants while the robbery was in progress and upon his release at its conclusion. Cook‘s encounter with the one later identified as appellant was not the first; just minutes before, on his way to the grocery store, Cook had seen him standing on a street corner.48 Cook went home right away and reported the affair to the police. The report—made no later than about 12:00 noon—does not seem to have been unduly delayed, despite the uncertainty in the record as to exactly when it and the reported incidents respectively occurred.49 Cook, very properly, took time to supply descriptions of his assailants in substantial detail.50 The descriptions were disseminated by police radio, as they had to be; this must have been done promptly, since Officer Polk heard the broadcast at 12:20 p. m. Then, recalling his observation earlier that day of a man in the offense area fitting one of the descriptions—appellant, as it later turned out51—the officer dutifully embarked upon a search for him. Appellant and a man fitting the second description were sighted, and the officer had to radio for help and await its arrival. When, at about 1:45, the men were taken into custody, the officer faced a choice be-

tween taking them to Cook‘s house for a viewing or transporting them to a police facility for a lineup. The officer elected the first alternative, and followed through on it immediately.

We recount these unfolding events because for us they reveal a societally indispensable endeavor to capture two criminals at large which from beginning to end occupied but a relatively brief time span. They reveal, too, efforts in that direction—by police and victim alike—pressed with as much diligence as can reasonably be expected, followed quickly by an effort to ascertain the accuracy of the results.52 The victim‘s prompt report to the police set in motion an equally prompt search for his assailants, an activity continuing aggressively and without pause until two suspects were apprehended. With the chances of intercepting two robbers on foot dwindling rapidly as time passed, the need to know whether the suspects were the culprits was urgent. Delay for a formal lineup with counsel, under the circumstances here, could easily have meant that unless the search was already successful, its resumption was doomed to almost certain failure.

We realize, of course, that the interest in reliable-because-fresh identifications was not assisted in this case quite as well as it was by the 30-minutes-later identification confrontation in

. Yet we cannot believe that so small a time difference—a half hour to an hour—could seriously depreciate Cook‘s at-home identification in terms of freshness, the value of which
extolled.53 And certainly the public interest in efficient police operations with a minimum of incommodity to the citizenry54 weighs just as heavily here as it has in other cases.55

In our view, the confrontation we now scrutinize served the end that any ensuing identification might be much fresher, any innocent arrestee might be released much sooner, and any further police work might resume much faster than would have been possible if the delay incidental to arrangement of an orthodox lineup had been suffered. In this case, then, no less than in

, we are presented with “‘substantial countervailing policy considerations’ which we are reluctant to assume the Supreme Court would reject.”56 In the circumstances here, as in
, we are led to the conclusion that Cook‘s near-the-scene identification of appellant should stand.

We realize that “[a]t some point the nexus of time and place between offense and identification must become too attenuated to outweigh the admitted dangers of presenting suspects singly to witnesses.”57 We recognize the importance of confining identification presentations—even those manageable freshly after criminal episodes—“to situations in which possible doubts as to identification need[ ] to be resolved promptly.”58 We agree—and reaffirm—that such presentations must be restricted to situations where “the interest in speedy identification * * * justifies the failure to arrange a formal lineup * * * [and thus] the failure to provide the suspect with counsel.”59 Our holding is limited strictly to such situations and we caution, as we have before, that “absent such need the conventional line-up is the appropriate procedure.”60 In the case at bar, we find ample justification for the questioned confrontation in the combination of the necessity appearing, the relatively short time interval, and the potentially reliable identification. We also note that, after this case arose, the Metropolitan Police Department put into operation a regulation restricting on- and near-the-scene identification confrontations to suspects arrested within 60 minutes after the alleged offense and in close proximity to the scene.61 We see in this regulation a careful and commendable administrative effort to balance the freshness of such a confrontation against its inherent suggestiveness, and to balance both factors against the

need to pick up the trail while fresh if the suspect is not the offender.62 We see no need for interposing at this time any more rigid time standard by judicial declaration. Our conclusion is fortified by the consideration that the absence of counsel, though lawful, is one of the circumstances that may be taken into account by a court, along with the ability of the witness to observe the offender in the first instance, in determining the fairness of the confrontation. The court‘s review of the situation must consider whether there was a sufficient likelihood that the witness was making an identification that was his own, rather than one thrust upon him, and that in turn requires a consideration of the likelihood that the witness was capable of making an accurate identification, or whether the identification was inherently unreliable.63

V

There remains only appellant‘s contention that the District Judge erred in ruling that the confrontation at Cook‘s house did not impinge on due process. We think, however, that the ruling must be sustained. While we have no doubt that some suggestiveness unavoidably inheres in every showup64 the factors already discussed65 countervail to insulate the confrontation against the ignominy of a constitutional viola-

tion.66 And by a process of balancing the competing considerations, we have consistently held that “absent special elements of unfairness,”67 a showup close in time and place to the crime does not encroach on due process.68

We find nothing in the circumstances to remove the case at bar from the purview of our previous holdings on this score. Both Cook and Officer Polk testified that no suggestion was made to Cook that appellant was believed to be one of the two robbers, but only that Cook was asked to view two men because they seemed to fit the descriptions Cook had furnished the police. It is doubtful, too, that the suggestivity inherent in the situation went beyond that present in other showups since Cook demonstrated a power of discrimination by identifying only one of the two men.69 Moreover, the District Judge heard the witnesses, weighed the testimony, and by a fully correct application of the governing legal principles concluded that the test of a due process violation was not met. Giving the judge‘s conclusion the deference70 due on a record providing ample support, we find no error in his due process ruling.

The judgment of conviction appealed from is accordingly

Affirmed.

BAZELON, Chief Judge (dissenting):

Judge Robinson‘s careful opinion for the court seems to me so substantially correct that it is with great reluctance that I dissent.

As an initial matter, let me add a slightly different perspective to the court‘s discussion of the different interests which underlay our decision in

. The court says:

Once the time factor * * * enhances the trustworthiness of a near-the-scene identification, two other interests also come to the fore. These are the combined needs for fast and effective police action in apprehending criminals, and the avoidance of unnecessary inconvenience and embarrassment to innocent citizens incidental thereto. * * *

We do not, however, in summarizing the interplay of the two last-mentioned interests, intimate that apart from the freshness of an identification they are constitutional justifications for confrontations which omit the procedure prescribed by Wade. The foundation of our Russell decision was an early confrontation aiding fairness by improving the reliability of the identification. The “expeditious release of innocent suspects” then became an additional reason for dispensing with the necessity for counsel upon an identification quite independently deemed reliable because it was fresh.1

If the needs for “fast and effective police action in apprehending criminals” and the “expeditious release of innocent suspects” are viewed as additional reasons for the

rule, it is vital to recognize, as I am sure the court does, that these reasons are strictly dependent upon the reliability of the identification. Police action, in the long run, is not made more efficient if the police are likely to be misled into thinking that they have apprehended the criminal when in fact they have not; similarly, innocent suspects are not protected if

there is a strong likelihood that they will be misidentified as criminals. In short, the additional interests allegedly served by the

rule are served only to the extent that we are confident that prompt near-the-scene identifications are reliable.

I set forth this reminder of the importance of the reliability of the identification because it seems to me that the court may have lost sight of that importance in Part IV of its opinion, where it turns to applying

to this case. Part IV summarizes the circumstances of the identification at issue here and then states:

We recount these unfolding events because for us they reveal a societally indispensable endeavor to capture two criminals at large—which from beginning to end occupied but a relatively brief time span. They reveal, too, efforts in that direction—by police and victim alike—pressed with as much diligence as can reasonably be expected, followed quickly by an effort to ascertain the accuracy of the results. * * * With the chances of intercepting two robbers on foot dwindling rapidly as time passed, the need to know whether the suspects were the culprits was urgent. * * *3

I am as delighted as any citizen to learn that the police have been diligent, and I am sympathetic to the cry of urgent need. What troubles me is that diligence and urgency may both continue for many hours after a crime, so I cannot say that these are the critical considerations in applying or extending the

rule. The court goes on to say:

We realize, of course, that the interest in reliable-because-fresh identifications was not assisted in this case quite as well as it was by the 30-minutes-later identification confrontation in Russell. Yet we cannot believe that so small a time difference—a half hour to an hour—could seriously depreciate Cook‘s at-home identification in terms of freshness, the value of which Russell extolled.4

Here, then, is the crux of the difference between us: the court thinks that an identification made an hour to an hour and a half after the crime is roughly as reliable as one made 30 minutes after, while I cannot be so confident. We deal here in a misty area of intuitive psychological judgments. If the Government had brought to our attention psychological studies showing that mental images of the sort at issue do not fade for several hours, I might feel compelled to push the

exception farther. But since they have not, I feel bound to be exceedingly careful about extending an exception which threatens to undermine the principles of fair identification procedures set down by the Supreme Court in Wade.5

Caution about extending

requires careful attention to factors which contribute to the reliability of the identification. In
6 I concurred in affirming a

conviction involving an on-the-scene identification 45 minutes after the offense, but there all that was being identified were clothes and appellant‘s general appearance. The court‘s opinion in this case suggests other factors that will be relevant: “the quality of the witness’ opportunity to observe when the offense was committed, and the caliber of the description of the perpetrator he furnished the police.”7 I am not convinced that the caliber of the description is a useful guide to reliability since, as we said in

:

[T]he conscious attempt to separate the ensemble impression into particular verbalized features, in order to preserve some recollection, may well distort the original accurate image so that it is the verbalized characteristics which are remembered and not the face or the man.8

But what particularly troubles me about Part IV of the court‘s opinion is the inadequacy of its focus on the factors it admits are relevant to credibility. Granted that Cook, the victim, described the clothes and general appearance of his assailants to the police, his identification was not, as in Cunningham, so limited. More important, I cannot believe that the court is particularly impressed by Cook‘s opportunity to observe. Of the two robbers, appellant is said to be the one who grabbed Cook from behind and held him on the ground while the other went through Cook‘s pockets. Cook testified that he was able to observe appellant‘s face “during and after” the robbery. But he admitted that the entire incident took no longer than one or two minutes, and from his description of the events, it could well have taken considerably less than that. We are left with the very real possibility that he got no more than a fleeting look at his assailant.9

I conclude that this identification was not an unusually reliable one. Given the length of time between the crime and the identification, therefore, I cannot agree that this case falls within the

rule as it has been applied. Since, in addition, the court has not persuaded me that we should expand the
rule, I respectfully dissent.

Notes

1
D.C.Code § 22-2901 (Supp. I 1968). Pp. 1033-1034 supra, footnotes omitted.
2
D.C.Code § 22-504 (1967). It is perhaps worth noting, as I have before, that “the argument from reducing unnecessary detention of innocent suspects is very weak. First, I believe that an innocent man, were he as aware of the dangers of misidentification as lawyers and judges are, would generally prefer to have the necessary time taken for a fairly conducted line-up. The logic of protecting the innocent forces one to the position that the supposedly innocent man should have some voice in the decision to return to the complaining witness. ‘Protecting the innocent’ by giving them no choice at all has a very hollow ring to me.

“Second, it seems to me that innocent men are adequately protected by the requirement that the police have probable cause before they arrest a suspect. * * * ”

 (Bazelon, C. J., dissenting). Police effectiveness, therefore, is the critical “countervailing policy consideration” underlying the Russell rule.
3
See note 6, infra. P. 1036 supra, footnotes omitted.
5
See
. See Note, Right to Counsel at Scene-of-the-Crime Identifications, 117 P.Pa.L.Rev. 916, 921 (1969).
6
The record leaves uncertain the time at which the robbery occurred. At the hearing, Cook accepted counsel‘s approximation of 11:30 a. m.; at trial, appellant accepted the same approximation as the time at which he allegedly saw Cook getting up from the ground, as hereinafter related. Cook, at trial, however, testified that “it could have been 11:30 and it could have been a few minutes to 12:00. * * *”

This uncertainty, of course, prevents exact computation of the time lag between the offenses and the identification confrontation in question. The times at which other significant events took place, however, are fixed by the record with fair precision. As stated hereafter in text, Cook‘s description of his assailants was on the police radio network at least by 12:20 p. m., appellant was apprehended at about 12:45 p. m., and the confrontation was around 1:00 p. m.

.
7
Prior to entering the store, Cook had observed several men, one of whom Cook later said was appellant, standing on a street corner. P. 1032, n. 35 supra.
8
At the time, appellant made no attempt to flee. The other suspect walked away as the officer approached, and was picked up a short distance away. Appellant tenders no issue as to the legality of his apprehension.
.
9
Cook informed Officer Polk that the other man had been on the street corner when he entered the store but was not involved in the robbery. Cook testified that on his way to the grocery, just before the crime, he noticed appellant standing on the corner nearby, in a small group. While this evidence may be appropriate for the jury to consider, I do not find that it materially strengthens the reliability of Cook‘s identification for the purposes of Russell. In the first place, a glance at a stranger on the street is not so likely to make a vivid impression on one‘s mind as we have assumed that a criminal assault is. In the second place, we have no independent evidence tending to confirm that the man Cook saw on the street corner was the man who assaulted him.
10
The record discloses only that, upon appellant‘s apprehension, Officer Polk “advised him of his rights.” We find much too little in this unelucidated revelation to anchor a waiver of right to counsel at the subsequent identification confrontation at Cook‘s house, the imminence of which, from aught that appears, appellant was completely unaware of. See
. Compare
;
.
11
Supra notes 4-5.
12
, ruled that Wade and Gilbert apply prospectively, and thus only to confrontations occurring after June 12, 1967.
13
This ruling predated
, but in both of its aspects the judge relied upon
, in which attacks grounded on due process and Fed.R.Crim.P. 5(a) failed to invalidate an on-the-scene confrontation. The judge disposed of the Stovall contention on a finding that the confrontation was “not so unnecessarily suggestive as to create the likelihood of mistaken identity” but, similarly to Wise, produced “a fresh identification, based upon a prior description given by the witness himself.” And with the confrontation occurring “immediately after [appellant‘s] apprehension and shortly after the offense,” the judge felt that Wise was persuasive to the conclusion that Wade-Gilbert principles were not infringed.
14
;
. The ruling was based prominently upon the dangers accompanying eyewitness identifications and the unfairness resulting from the inability of the accused to reconstruct at trial the circumstances surrounding the identifications.
.
17
;
. There is an exception in consequence of our holding in Clemons that the Gilbert per se exclusionary rule does not operate retroactively in the due process area where the impugned confrontation took place before the Supreme Court‘s decision in Stovall,
.
18
We consider the latter infra Part V.
19
Supra note 13.
21
Since both
and
involved post-indictment lineups after the accused had secured counsel, some courts have held that pre-indictment lineups are not protected by the Wade-Gilbert rule. E. g.,
;
. Other courts have rejected this view, e. g.,
;
, as have we. See
, discussed in text at notes 19-26;
(identification at accused‘s preliminary hearing);
(same).
27
In Russell, the confrontation occurred about 5:00 o‘clock in the morning, and we observed that “there would necessarily be a long delay in summoning appellant‘s counsel or his substitute counsel to observe a formal lineup.”
. But the early hour at which the confrontation occurred, though of importance, was not critical, for we have since held that a near-the-scene confrontation taking place during normal business hours calls for an application of Russell.
;
, at 8.
28
Our holding in Russell “approve[d] only those on-the-scene identifications which occur within minutes of the witnessed crime.”
. The “within minutes” confrontation Russell condoned is one within a half-hour, since the crime there occurred at about 4:30 a. m. and Russell was not arrested until about 5:00 a. m.
. See also note 47, infra, and accompanying text.
29
See note 6, supra.
30
The Government also contends that Cook‘s in-court identification rests adequately upon an independent source, and that alternatively we might so hold. Since we conclude that appellant‘s pretrial identification involved no constitutional violation, it is unnecessary to reach that question, and it could hardly be resolved on this appeal in any event. The issue as to independent source was not raised in the District Court and in consequence no ruling was made as to it; the District Judge did not have the benefit of our suggestion that such a ruling be made whether requested or not,
, because Clemons was decided after the instant case was tried. Moreover, resolution of the issue over independent source might well have required judgments on credibility, which should be made at the trial level. Cook said that he observed appellant standing on a street corner prior to the robbery, appellant said that Cook misidentified him immediately after the attack, and the testimony of both parties had relevance to any determination on independent source.
31
But see note 28, supra.
32
It has been suggested that Russell-type exceptions to Wade-Gilbert be limited to identifications made within 30 minutes of the crime and within 15 blocks of the crime scene, or 30 blocks if the suspect left in a car. Note, Right to Counsel at Scene-of-the-Crime Identifications, 117 U.Pa.L.Rev. 916, 924-925 (1969). And see text infra at note 60.
34
See text infra at notes 37-41.
35
Prominent among these are the quality of the witness’ opportunity to observe when the offense was committed, and the caliber of the description of the perpetrator he furnished the police.
36
As we have stated, the choice between Wade and Russell must accommodate all pertinent considerations—including very importantly the time lapse from the offense. See text supra at notes 34-35. Where, however, the result on balance is uncertain, Wade is presumptively to be favored as the primary expositor of constitutional values. “In the absence of legislative regulations adequate to avoid the hazards to a fair trial which inhere in lineups as presently conducted, the desirability of deterring a constitutionally objectionable practice must prevail over the undesirability of excluding relevant evidence.”
.
39
“[R]ecognition of a person or face would seem to be as much the product of a subjective mental image as of articulable, consciously remembered characteristics. A man may see clearly in his ‘mind‘s eye’ a face or a figure which he is hard put to describe adequately in words. Though the image of an ‘unforgettable face’ may occasionally linger without any translation into words, photographic recall is most often ephemeral. Vivid in the flash of direct observation, it fades rapidly with time. And the conscious attempt to separate the ensemble impression into particular verbalized features, in order to preserve some recollection, may well distort the original accurate image so that it is the verbalized characteristics which are remembered and not the face or the man.”
.
40
Where the identification occurs moments after the crime, as when the suspect is apprehended after a brief chase, so little time is involved that the identification has been likened to the hearsay exception for spontaneous declarations. See
. Where the confrontation was not so immediate, an important inquiry is whether it was so postponed that the observed identification stimuli have lost their acuity.
41
As to each of these considerations, speed is essential. The Supreme Court has recognized that “prejudicial delay” could well occur, and premised a suggestion as to possible use of substitute counsel on that consideration.
.
43
Our dissenting colleague argues that these considerations “are strictly dependent upon the reliability of the identification,” and “are served only to the extent that we are confident that prompt near-the-scene identifications are reliable.” Infra pp. 1038-1039 (emphasis in original). Coupling the premise that the identification in this case “was not an unusually reliable one” with the circumstance that the time lapse here was somewhat longer than in Russell, he further argues that the situation at bar is beyond the purview of the Russell ruling. Infra, p. 1040. The proposition that reliability-in-fact so largely shapes the judicial ruling as to on- or near-the-scene identifications without counsel is the pervasive theme of the dissent.

As we read Russell, the decision hinged upon the probability of enhanced reliability of identification at an early confrontation—in other words, the essential fairness of the identification procedure—and not upon a judge‘s attempt to assess its actual reliability, which is the traditional function of the jury. Indeed, it could hardly be said that the identification in Russell was “an unusually reliable one,” for it was based upon the identifying witness’ observation of the offender as he passed along the sidewalk on the opposite side of the street at 4:30 in the morning. As the Russell opinion explicates, the court‘s “conclusion does not rest on a determination that [the witness‘] identification was in fact especially reliable. It rests instead on a general rule that it is not improper for the police immediately to return a freshly apprehended suspect to the scene of the crime for identification by one who has seen the culprit minutes before.”

(emphasis supplied). On such a rule also have rested our other decisions treating in other contexts the problem of on- or near-the-scene identifications. See cases cited supra notes 37-38.

45
Since we deem Russell controlling, we decided this case by application of the doctrine expounded in that case, and intimate no view on the issues we would face were the situation otherwise.
46
As we have pointed out, supra note 28, the delay in Russell itself was about 30 minutes. See also
, involving a Stovall-type claim, where the interval between offense and confrontation was about the same.
47
The record in
, affirming a conviction without opinion, discloses the passing of 45 minutes to an hour from the offense to a showup attacked on right-to-counsel grounds. Trial Transcript at 11-12. See also
(identification confrontation, without counsel, 90 minutes after offense). In two cases involving confrontations assailed on due process grounds, the interval was about two hours.
;
, Trial Transcript at 12. And see
, sustaining an “on-the-scene ‘identification’ within an hour of [a] robbery primarily * * * to establish similarity of clothing and general appearance as an additional circumstance in the chain leading to appellant‘s apprehension with a portion of the stolen money in his possession.”
48
See note 7, supra.
49
See note 6, supra.
50
The description which appellant matched reasonably well specified a man about 30 years old, five feet eleven inches tall, 200 pounds in weight, wearing a blue waist-length jacket and blue pants.
51
The dissent, quoting language from Russell, questions whether “the caliber of the description is a useful guide to reliability” of an ensuing identification. Infra p. 1040. That it is, seems clear, we think, both from Wade and from our own numerous decisions, which assign it importance as a factor indicating an independent source for a later in-trial identification.
;
;
;
;
;
;
;
. Surely a reasonably accurate pre-arrest description makes the same contribution to the potential reliability of an identification forthcoming much closer to the observations made at the time of the offense. And the passage from Russell upon which the dissent relies in this connection was a part of the court‘s rationale for—not against—the prompter though lawyerless identification confrontation in that case.
52
Obviously the situation here—an identification confrontation assisting a determination as to whether a freshly-apprehended suspect is the offender—contrasts sharply with a confrontation designed to build up evidence for use at the trial of one already charged formally with the crime, as was the objective in
and
.
55
Surely no basis for distinguishing Russell can be found in the mere fact that the confrontation took place, not on the sidewalk where Cook was robbed, but at his house, to which he had repaired immediately thereafter to report the crime to the police.
, expands the Russell doctrine to other than strictly on-the-scene confrontations where a valid reason appears for an identification effort at a location near in time and place to the point at which the offense was committed, and our Stovall-type decisions have established a similar principle. See
;
;
.
57
. There we held that that point was surpassed in a due process context by the lapse of nearly four hours from the commission of the offense.
61
“If a suspect is arrested within 60 minutes of an alleged offense and within an area reasonably proximate to the scene of the crime, he shall be returned to the scene of the offense, or the eyewitnesses shall be transported to the scene of the arrest, for identification of the suspect.” Memorandum Order No. 16, Series 1970, ¶ I-1 (May 15, 1970).
62
See text supra at notes 37-43.
64
“The presentation of only one suspect, in the custody of the police, raises problems of suggestibility that bring us to the threshold of an issue of fairness.”
. See also
;
.
65
See text supra at notes 37-43.
66
See cases cited supra note 38.
68
See cases cited supra note 38.
69
Compare
;
.