Kenneth Solomon v. Harold J. Smith, Superintendent, Attica Correctional Facility, Respondent , 645 F.2d 1179 ( 1981 )

KEARSE, Circuit Judge:

The State appeals from a judgment of the United States District Court for the Southern District of New York, Vincent L. Broderick, Judge, granting the petition of appellee Kenneth Solomon for a writ of habeas corpus on the ground that the identifications of Solomon at his state trial were unreliable and resulted from impermissible police procedures, in violation of Solomon’s rights under the Fifth, Sixth and Fourteenth Amendments to the Constitution. This case, which arises from Solomon’s conviction of robbery, rape, and sodomy, solely on the basis of eyewitness testimony, presents an unusual confluence of circumstances. The events commenced as a robbery of a doctor’s office and ended with the three robbers, one of whom wore a hood and was addressed by his companions as “Kenny,” raping and sodomizing the receptionist, Nancy Padovani. Padovani, the principal eyewitness, identified Solomon at trial with positive certainty. Her initial selection of a picture of Solomon, however, had been contemporaneously characterized by the police as a “negative” but “possible” identification. Between that initial selection of the Solomon picture and her positive identification at a Wade hearing¹ a week before trial, Padovani was shown the Solomon picture several times; she viewed Solomon at an extended, uncounseled show-up during his arraignment, and she viewed him in a lineup in which the only person even close to his height was 65 pounds heavier than he. The most striking feature of this case, however, was Padovani’s selection from an unsuggestive lineup that did not include Solomon, of another man as her hooded assailant. It transpired that this man’s name too was Kenny, Kenneth An-scombe, that Kenneth Anscombe fit Pado-vani’s initial description of the hooded assailant better than did Solomon, and that Kenneth Anscombe has been convicted of participating in other armed robberies nearby, with one of the non-hooded assailants identified by Padovani.

The combination of circumstances persuaded the district court that the identification of Solomon at trial was impermissible under prevailing constitutional standards, and it ordered the State either to release Solomon or to retry him within 60 days. We affirm.

A. The Crimes

On October 7, 1974, three black males, one wearing a hood and carrying a gun, entered the office of Dr. Jean-Louis Cas-seus and announced a “stick-up.” The receptionist, Nancy Padovani, led them to Dr. Casseus’s office where they robbed the doctor'at gunpoint of $50 and his watch. The three then returned to the reception area, took Padovani’s watch and some cash from her pocketbook and ordered her to lock the doctor and a patient in a closet. The intruders then raped and sodomized Padova-ni. The hooded individual stood behind Pa-dovani during some of this time and was the last to assault her. As he did, the other two assailants sought to leave and, apparently becoming nervous, shouted, “Come on Kenny, quick. Let’s go.” After leaving •the doctor's office, the three fled in a taxi. The incident lasted approximately ten to fifteen minutes. Afterwards, Padovani released the doctor and patient from the closet and called the police. She was taken to a hospital for treatment and tests and it was there that she first spoke with Detective Lawrence Doherty. Padovani described the gunman to Detective Doherty as male, black, age 21, 5'7" tall, weighing 145 pounds, and wearing a blue parka with a hood. Although Padovani was later to testify at the Wade hearing that she thought this initial description included a prominent facial scar, there is no mention of a scar in the contemporaneous report of Detective Doherty, whose practice was to ask about and make note of any uncommon identifying marks. Doherty did not recall mention of a scar.

B. Identification by Padovani

On October 8, the day after the incident, Padovani went to the police station where she was shown two trays of photographs. One tray was labeled “Negro, Male, Sex” and the other “Negro, Male, Robbery under 5'8"." Each tray contained approximately 100 pictures. She selected five or six pictures from these photos and then from that smaller set selected that of petitioner Kenneth Solomon. Doherty made a copy of the picture of Solomon, on which Padovani drew a hood around the face with a blue marker. Although Padovani testified at the Wade hearing nine months later that as soon as she drew the hood she was “positively sure” that Solomon was the hooded man who had raped her, on October 8 Detective Doherty recorded Padovani’s identification in his photo log book by placing a check mark in the column headed “Negative,” rather than in that headed “Positive,” and merely wrote “possible” next to his entry in the Negative column. This negative/possible entry in the police log book is the only contemporaneous evidence as to the initial level of Padovani’s certainty that Solomon was her hooded assailant.

By the time Solomon was to be tried in the summer of 1975, however, Padovani was quite certain that Solomon was the hooded assailant. In the interim, several events occurred that seem likely to have contributed to her objectively perceptible increase in certainty. First,' near midnight on October 17 she was summoned to the precinct house for a lineup by Doherty, who told her Solomon had been arrested. She did not actually see Solomon that night, since by the time she arrived it had been decided not to have a lineup after all; she did, however, see Solomon’s picture tacked to a bulletin board in the police station.

On the following day, October 18, Solomon was arraigned. Padovani testified that Doherty had in his hand the same picture of Solomon that she had seen on the police bulletin board the night before. It is unclear how long Padovani and Solomon were in the courtroom together on the occasion of the arraignment, but Padovani met Doherty in court on the morning of October 18, and Solomon was not actually arraigned until nearly 5 p. m. During at least part of the afternoon Solomon was in a detention area of the courtroom with several other persons. Doherty made no effort to ask Padovani at that time if she could pick out Solomon. He waited until after the arraignment had actually taken place — a proceeding that lasted some 20-30 minutes, with only the prosecutor, Solomon, Doherty and Padovani standing before the judge— and then asked, “Did you get a good look at him?” Padovani said she had. Doherty asked, “Is that him?” She said it was. At no time during the arraignment proceeding was Solomon represented by counsel.

After the arraignment there was a grand jury proceeding at which Padovani testified. Before she testified Doherty showed her Solomon’s picture again. After Pado-vani testified before the grand jury, Doherty at least once showed her and other witnesses pictures of persons who might have been the nonhooded assailants. (One such person, Clayton Smalls, was identified by Padovani at a lineup in November.) On each occasion Doherty also showed Padova-ni the picture of Solomon again.

Finally, on January 16, 1975, Padovani was shown a lineup of five men that included Solomon, whom she identified. Having initially described the hooded assailant as 5'7" tall and weighing 145 pounds, her selection of Solomon from this particular array was perhaps not difficult. Solomon stands 5'6" tall and weighs 130 pounds; the only other person in the lineup who was within two inches of this height weighed in at 195 pounds:

Height Weight

Suspect # 1 5'8" 195 lbs

Suspect #2 6' 170 lbs

Suspect #3 6' 165 lbs

Kenneth Solomon 5'6'' 130 lbs

Suspect # 5 5'10" 146 lbs

In addition, although none of the intruders had been described as having beards or mustaches, and Solomon had no such facial hair, some of the other participants in Solomon’s lineup had beards or mustaches.

Before picking Solomon out of this suggestive lineup, Padovani had attended two other lineups on that day, neither of which included Solomon. At the second of these lineups there was an array that was composed entirely of persons close in height and weight to the description initially given by Padovani:

Height Weight

1. Larry Sutton 5'8V4" 135 lbs

2. James Aliord 5'7" 169 lbs

3. Thomas White 5'7" 150 lbs

4. Kenneth Anscombe 5'6" 147 lbs

5. Vernon Tull 5'7" 150 lbs

As soon as the shade went up on this lineup, Padovani instantly picked out Anscombe as her hooded assailant: “I see him already, four. That’s the one I said.”

After Padovani left the viewing room, she retracted her identification of An-scombe. Her testimony at the Wade hearing on this identification was as follows:

Q. But in the second line-up, what you are saying, when you picked out that man, was that this was the number one man in the office [i. e., the gunman], right?

******

A. Yes, but I only seen him for one second, not even — I just got scared because I seen the facial features and I, you know, didn’t really look. All I saw was the face and I walked away, run out of there because I got scared and I told them outside, “I made a big mistake because he was much too heavy, you know, Solomon was a little skinnier.”

Kenneth Anscombe was indeed fifteen pounds heavier than Solomon; but he weighed just two pounds more than the hooded assailant originally described by Pa-dovani, and he lacked the facial scar that was omitted from Padovani’s first description. And Kenneth Anscombe was later convicted of participating in other robberies, in the vicinity of Dr. Casseus’s office, with Clayton Smalls, whom Padovani had already identified as one of her nonhooded assailants.

II

As the Supreme Court has stated, [t]he vagaries of eyewitness identification are well-known; the annals of criminal law are rife with instances of mistaken identification.... A commentator has observed that “[t]he influence of-improper suggestion upon identifying witnesses probably accounts for more miscarriages of justice than any other single factor— perhaps it is responsible for more such errors than all other factors combined.” Wall, Eye-Witness Identification in Criminal Cases 26.

United States v. Wade, 388 U.S. 218, 228-.29, 87 S.Ct. 1926, 1932-1933, 18 L.Ed.2d 1149 (1967). In recognition of these realities, we have always given close scrutiny to the possibility of constitutional violations in criminal cases resting largely or solely on eye-witness testimony:

Centuries of experience in the administration of criminal justice have shown that convictions based solely on testimony that identifies a defendant previously unknown to the witness is highly suspect. Of all the various kinds of evidence it is the least reliable, especially where unsupported by corroborating evidence.

Jackson v. Fogg, 589 F.2d 108, 112 (2d Cir. 1978). On the basis of the facts detailed above, we agree with Judge Broderick’s conclusion² that the admission of Padova-ni’s identification testimony at trial violated Solomon’s rights under the Due Process Clause of the Fourteenth Amendment³ and his right to counsel under the Sixth and Fourteenth Amendments.

A. Due Process

A defendant’s right to due process of law includes the right not to be the object of suggestive police identification procedures that create a “substantial likelihood of irreparable misidentification.” Neil v. Biggers, 409 U.S. 188, 198, 93 S.Ct. 375, 381, 34 L.Ed.2d 401 (1972); Simmons v. United States, 390 U.S. 377, 384, 88 S.Ct. 967, 971, 19 L.Ed.2d 1247 (1968); Jackson v. Fogg, supra, 589 F.2d at 111; see Manson v. Brathwaite, 432 U.S. 98, 97 S.Ct. 2243, 53 L.Ed.2d 140 (1977). This protection must encompass not only the right to avoid improper police methods that suggest the initial identification, but as well the right to avoid having suggestive methods transform a selection that was only tentative into one that is positively certain. The jury, in order to convict, must find the evidence probative of guilt beyond a reasonable doubt. It will surely be more likely to convict on the basis of testimony that is certain rather • than tentative. Thus, while a witness is always entitled to become surer of an identification, due process precludes the generation of that increased certainty through a suggestive lineup or a prolonged one-on-one viewing not preceded by a proper lineup. See Simmons v. United States, supra, 390 U.S. at 383, 88 S.Ct. at 970. Cf. United States v. Wade, supra, 388 U.S. at 240-41, 87 S.Ct. at 1939.

In the present case, the start of the procedures was apparently irreproachable; Padovani was shown a large array of pictures, from which she selected one. But how certain was she of her selection? The only contemporaneous evidence of her level of certainty was Detective Doherty’s recording in his log book. There was a column headed “Positive” and one headed “Negative.” He checked “Negative” and wrote “possible” in that column. Hence, we should start with the premise that Padova-ni’s initial identification was not certain. What occurred between this initial uncertainty and the eventual certitude that was conveyed to the jury was a stream of displays of Solomon in contexts that could only fix his image in Padovani’s mind. She saw his picture — uncomplicated by any accompanying array — at least four times. Cf. Simmons v. United States, supra (the danger of an incorrect identification is “increased if the police display to the witness only the picture of a single individual who generally resembles the person he saw”). Her first opportunity for a post-arrest viewing was a one-person “showup” at his arraignment, when she was in his presence for 20-30 minutes.⁴ Cf. Moore v. Illinois, 434 U.S. 220, 229, 98 S.Ct. 458, 465, 54 L.Ed.2d 424 (1977) (“It is difficult to imagine a more suggestive manner in which to present a suspect to a witness for their critical first confrontation” than at a preliminary hearing to determine whether to bind the accused over to the grand jury.). She was allowed to see him in a lineup only after she had seen his isolated picture many , times and had been standing one person away from him for 20-30 minutes at arraignment; and in the only lineup in which he appeared, all of the other lineup participants were significantly larger than he. The combination of these repeated showings of Solomon was impermissibly suggestive. Cf. Foster v. California, 394 U.S. 440, 89 S.Ct. 1127, 22 L.Ed.2d 402 (1969).

Given these impermissible procedures, the teaching of Neil v. Biggers, supra, and Manson v. Brathwaite, supra, is that the admissibility of Padovani’s identification of Solomon at trial depends on its reliability, and as Brathwaite makes clear, 432 U.S. at 114, 97 S.Ct. at 2253, the constitutional assessment of reliability requires a balancing of the factors outlined in Biggers, against the degree of suggestiveness in the impermissible procedures. Biggers and Brathwaite refer to five factors to be in-eluded in an evaluation of the likelihood of mistaken identification. These factors are (1) the witness’s opportunity to observe the criminal at the time of the crime, (2) the degree of the witness’s attention at that time, (3) the accuracy of the witness’s initial description of the criminal, (4) the certainty with which the witness first identified the suspect, and (5) the time lapse between the crime and the identification. Normally, if the first four of these are rated good, especially if the time interval is short, the identification of the suspect will be considered reliable and the likelihood of misidentification insubstantial. Normally, however, the analysis is concerned with only one suspect and there are no serious alternate candidates in the picture. This case presents the unusual situation in which the witness not only identified the defendant, she also selected a highly plausible alternate candidate from a lineup. In these circumstances, the third and fourth factors of the Biggers-Brathwaite analysis — initial description and certainty — take on a heightened importance that alters the significance of the first two factors. If the witness’s initial description more accurately describes the alternate candidate than the defendant and the selection of the defendant was not positive, thus pointing toward a misidentification rather than toward reliability, then a finding that the witness had a good opportunity to observe and a high degree of attentiveness will further support a conclusion of mis-identification rather than suggesting reliability. In the present case the first four Biggers-Brathwaite factors point toward a mistaken identification, and the fifth factor is wholly inconclusive.

First, Padovani’s opportunity to view the criminal at the time of the crime: hers was not a fleeting glimpse; the event lasted 10-15 minutes, and while the hooded assailant was not in her line of vision the entire time, there was more than enough time to see his size and shape and so much of his face as was visible. The hood prevented a comprehensive viewing of the face; it obscured his ears, his hairline, and the shape of his head. Indeed Padovani was forced to draw a hood on Solomon’s picture to increase her certainty that this was the right face; and Dr. Casseus remained uncertain through the Wade hearing that the unhood-ed Solomon he could see was the hooded assailant he had seen.⁵ Second, the degree of Padovani’s attention: attention to the event frequently must be distinguished from attention to the assailant. The event was chaotic, harassing, and frightening; Dr. Casseus, for example, said at the Wade hearing that he was “watching the gun most of the time.” Given the nature of the assault on Padovani, there can be little doubt that she was sufficiently attentive to the assailants to form a strong mental image of their sizes and shapes and so much of their faces as she saw. Third, the accuracy of Padovani’s initial description of the assailant: her description was only fractionally accurate as to Solomon; it was a remarkably accurate description of Anscombe. Pa-dovani said the hooded assailant was 5'7" tall and weighed 145 pounds. Anscombe is 5'6" tall and weighs 147 pounds. Solomon too is 5'6" tall, but he is 15 pounds lighter. Padovani described the hooded assailant as 21 years old. Anscombe was 18; Solomon was 16. Solomon also has just above his left eye a prominent scar that would not have been concealed by a hood. Padovani did not, according to Doherty’s recollection and his notes, mention a scar. Anscombe had no scar. Fourth, the level of certainty demonstrated by Padovani in selecting Solomon’s picture: the level of certainty must be considered low. Without in any way impugning Padovani’s veracity, we discount her recollection nine months later- of this initial identification as “positively sure,” since it had been contemporaneously recorded by the police as negative/possible, and in the interim she had been subjected to at least six suggestive police procedures, some of which she did not even recall.⁶ In addition, one’s faith in Padovani’s certainty must be shaken by her sudden identification of Anscombe in a nonsuggestive lineup. Fifth, the length of time between the crime and the initial identification: this factor viewed in isolation would not point toward a misidentification since the selection of Solomon’s picture was made the day after the event. But the immediacy of the selection cannot be evaluated independently of several material considerations, including the facts that (a) the record does not reveal whether Anscombe’s'picture was in the arrays originally shown to Padovani, (b) the person whose face was selected did not, in two significant respects (weight and scar), match the description Padovani had given, compare Chavis v. Henderson, 638 F.2d 534 (2d Cir. 1980), and (c) the identification was not positive.

Thus there is little in the event or in the initial phases of the investigation to suggest that Padovani’s certainty was reliable. And during the period between the initial tentative identification and the ultimate certainty there were no fair procedures (such as unsuggestive lineup) that could provide any indicia of reliability.

Finally, Padovani’s identification of An-scombe strongly suggests that her initial selection of Solomon’s picture was a mis-identification. Her description of her instant selection of Anscombe from a lineup of five men of nearly the same heights and weights (“I just got scared .... All I saw was the face and I walked away, run out of there because I got scared ....”) gives every evidence of genuine recognition. When to this is added the fact that An-scombe’s name is Kenneth, that he matches Padovani’s initial description nearly exactly, and that he had been associated in other robberies in the same neighborhood with one of the non-hooded assailants already identified by Padovani, we are compelled to agree with the district court that there was “a very substantial likelihood of irreparable misidentification,” and that due process compelled the granting of the writ.

B. Right to Counsel

It is undisputed that Solomon was not represented by counsel at his arraignment. Under New York law arraignment marked the commencement of the criminal action against Solomon. N.Y. Crim.Proc.Law §§ 1.20(1), (8), (17); 100.05; 170.10(1); 180.10(1) (McKinney 1971, 1980 Supp.); People v. Blake, 35 N.Y.2d 331, 339^40, 361 N.Y.S.2d 881, 320 N.E.2d 625 (1974). Thus, it was a crucial stage of the proceedings, and Solomon was entitled to have counsel to aid him. Moore v. Illinois, supra; Kirby v. Illinois, 406 U.S. 682, 92 S.Ct. 1877, 32 L.Ed.2d 411 (1972); Hamilton v. Alabama, 368 U.S. 52, 82 S.Ct. 157, 7 L.Ed.2d 114 (1961). In theory, at least, counsel might have been able either to forestall the prejudicial showup of Solomon at arraignment, or to limit the duration of the witness’s observation, or, preferably, to precipitate a proper lineup prior to the arraignment.⁷

When the accused has been subjected to a lineup or showup in the absence of counsel at a crucial stage of the proceedings, an identification at trial will not be permitted unless the witness’s ability to identify the defendant has an origin “independent” of the uncounseled identification proceeding.⁸ Kirby v. Illinois, supra; United States v. Wade, supra, 388 U.S. at 239-41, 87 S.Ct. 1926, 1938, 18 L.Ed.2d 1149. The tests of “independent origin” set forth in Wade appear to be functionally identical to the reliability tests articulated in Neil v. Biggers, supra.⁹ For the reasons detailed above in the discussion of the due process issue, we find in the record no independent source of Padovani’s certainty that was not itself the product of suggestive police procedures.

The judgment of the district court is affirmed.

1 . A Wade hearing is held to determine whether an ■ in-court identification has an independent basis and is therefore admissible. United States v. Wade, 388 U.S. 218, 87 S.Ct. 1926, 18 L.Ed.2d 1149 (1967).

2 . The state court judge at the Wade hearing refused to suppress the identifications by Padovani and Dr. Casseus, concluding that each was based on observations made during the crime on October 7, rather than on subsequent events or procedures. He did not, however, either use the analytical framework of Neil v. Biggers, 409 U.S. 188, 93 S.Ct. 375, 34 L.Ed.2d 401 (1972), see infra, or make detailed factual findings with respect to suggestiveness or reliability; he merely concluded that the police procedures were not unfair or unconstitutional. In these circumstances, the district court “quite properly made its own assessment of the undisputed facts” in order to apply the appropriate governing principles. Jackson v. Fogg, 589 F.2d 108, 111 (2d Cir. 1978); United States ex rel. Williams v. La Valiee, 487 F.2d 1006 (2d Cir. 1973), cert. denied, 416 U.S. 916, 94 S.Ct. 1622, 40 L.Ed.2d 118 (1974); Gonzalez v. Hammock, 477 F.Supp. 730, 732 n.6 (S.D.N.Y.1979). We agree - with Judge Broderick’s assessment that the state court’s determination is not fairly supported by the record as a whole, and thus is not entitled to a presumption of correctness under 28 U.S.C. § 2254(d). Taylor v. Lombard, 606 F.2d 371 (2d Cir. 1979), cert. denied, 445 U.S. 946, 100 S.Ct. 1346, 63 L.Ed.2d 781 (1980). See Sumner v. Mata, — U.S.-, 101 S.Ct. 764, 66 L.Ed.2d 722 (1981).

3 . Judge Broderick also properly found that the identification by Dr. Casseus violated due process. Dr. Casseus twice viewed a group of ten mug shots in his attempt to identify the intruders. At least one and possibly both of the arrays shown Dr. Casseus included the picture of Solomon on which Padovani had drawn the hood. This emphasized Solomon and was im-permissibly suggestive. See Simmons v. United States, 390 U.S. 377, 383, 88 S.Ct. 967, 970, 19 L.Ed.2d 1247 (1968). Further, at the Wade hearing, Dr. Casseus’s recollection and identification of Solomon were anything but impressive. He could give only the most general of descriptions of the intruders: they were black; they were young. He did not remember what descriptions he had given the police. He did not recall noticing any scars. His best effort at identifying Solomon was to testify that Solomon “reminds me of the fellow,” and that “[i]t looks like him.” He attributed his lack of certainty to the fact that the assailant had been wearing a hood (see note 5 infra). Unaccountably, at the trial one week later, Dr. Casseus was positive Solomon had been the hooded intruder, and Dr. Casseus now remembered that the hooded assailant had had a scar. Analysis of the facts in accordance with Neil v. Biggers, supra, see text infra, suggests that Dr. Casseus’s identification of Solomon at trial was not reliable.

4 . A one-person showup does not inevitably taint a subsequent courtroom identification, Neil v. Biggers, supra, 409 U.S. at 198, 93 S.Ct. at 381, but the procedure here is cause for concern because the initial photographic identification was not positive, no lineup preceded the arraignment, and the subsequent lineup was flawed.

5 . At the hearing, Dr. Casseus testified in part as follows:

Q Now Doctor, as you testified, you indicated that you looked at Mr. Solomon, and you said he looked something like the person?

A Yes.

Q And in effect, are you telling us that after the passage of eight months, you are unable to say with certainty that this is the person?

A If I’m able to—

Q I’m sorry.

A I didn’t hear you. I don’t understand.

Q Is what you are saying, when you say he looks like the person—

A Yes?

Q —an indication that—

A Well, he doesn’t have a hood now. I mean I cannot be exactly the way I saw him.

Q So you can’t be sure that this is the person?

A It looks like.

Q And that’s the best you could say? A Yes, because he had a hood at that time. He’s not exactly the same I saw him.

6 . Padovani did not recall being shown Solomon’s picture on any occasion after the arraignment. Doherty, however, testified that he showed her the picture on at least two occasions thereafter.

7 . We do not know whether the benefit of having counsel at the arraignment in this case would have been more than theoretical.

If an accused’s counsel is present at the pretrial identification, he can serve both his client’s and the prosecution’s interests by objecting to suggestive features of a procedure before they influence a witness’ identification.

Moore v. Illinois, supra, 434 U.S. at 225, 98 S.Ct. at 463. Ordinarily we would not assume that such requests would be vain. Here, however, the extremely suggestive lineup in which Solomon was placed was held despite his counsel’s objection and despite the indisputable availability of five other men each of whom was far closer to Solomon’s description than any of the men placed in Solomon’s lineup. The practical helplessness of counsel to forestall unfair identification procedures, however, does riot diminish the accused’s right to counsel nor minimize the breach of that right.

8 . There was no trial testimony as to Padovani’s identification of Solomon at the uncounseled arraignment showup. Evidence of such an identification is per se inadmissible. Gilbert v. California, 388 U.S. 263, 87 S.Ct. 1951, 18 L.Ed.2d 1128 (1967); see United States v. Wade, supra.

9 . Compare Wade, in which the Court listed factors to be considered as follows:

the prior opportunity to observe the alleged criminal act, the existence of any discrepancy between any pre-lineup description and the defendant’s actual description, any identification prior to lineup of another person, the identification by picture of the defendant pri- or to the lineup, failure to identify the defendant on a prior occasion, and the lapse of time between the alleged act and the lineup identification. It is also relevant to consider those facts which, despite the absence of counsel, are disclosed concerning the conduct of the lineup.

388 U.S. at 241, 87 S.Ct. at 1939 (footnote omitted), with Biggers:

[Tjhe factors to be considered in evaluating the likelihood of misidentification include the opportunity of the witness to view the criminal at the time of the crime, the witness’ degree of attention, the accuracy of the witness’ prior description of the criminal, the level of certainty demonstrated by the witness at the confrontation, and the length of time between the crime and the confrontation.

409 U.S. at 199-200, 93 S.Ct. at 382.