United States v. Paul Edward Bowie

515 F.2d 3 | 7th Cir. | 1975

Lead Opinion

CUMMINGS, Circuit Judge.

In April 1973, a two-count indictment was returned against Paul Bowie, Carl Higgans and Houston Watkins concerning the robbery of the University National Bank in Milwaukee, Wisconsin, on October 31, 1972. The first count alleged that these three individuals took $15,845 from the bank “by force and violence and by intimidation” in violation of 18 U.S.C. § 2113(a). The second count charged them with obtaining that sum by putting in jeopardy the lives of bank employees by using a revolver and sawed-off shotgun in violation of 18 U.S.C. § 2113(d). The trial court granted severance to all defendants. In Bowie’s trial a jury returned a verdict of guilty as to Count II after the court had instructed the jurors that “If the Government has met its burden as to Count No. II, then you don’t have to consider. Count No. I.” Thereafter the court granted the Government’s motion to dismiss Count I of the indictment. Subject to 18 U.S.C. § 4208(a)(2), Bowie received an eight-year prison term.

The evidence showed that three black males spent approximately five minutes robbing the bank of some $15,000 at 11 A.M. on October 31, 1972. They were wearing masks and were respectively described as tall, medium and short. The Government’s theory was that Bowie was the shortest of the robbers, wore a mask only over the lower portion of his face and carried a gun when he went behind the tellers’ cages to collect the money. When he reached down to pick up some of the money, his mask slipped somewhat from its original position just below his nose. The testimony does not make clear how much of his face was then visible, except that his moustache was exposed, while a small beard would not have been.

On November 2, 1972, Exhibit 25, consisting of five black and white photos, was displayed to bank employees George Dearborn and Cindy Spoerl. Both of them selected photo 25.4, a picture of Bowie, as one of the bank robbers.

On January 11, 1973, the same witnesses were shown Exhibit 13, consisting of six color photos. They selected picture 13.1A, a later photograph of the defendant, as being the photograph most closely resembling one of the bank robbers.

On September 5, 1973, bank customer Cornell Westmoreland also selected photo 13.1A from the six photographs in Exhibit 13 as most closely resembling one of the robbers.

In September 1973, at the trial of co-defendant Carl Higgans, each of the foregoing bank employees was again shown the same photographic displays as before and testified as to their selections. A physical lineup of Bowie was never conducted.

Pretrial and In-court Identification of Bowie

Bowie first asserts that all testimony regarding pretrial photographic selection and all subsequent in-court identification should have been suppressed, claiming that the photographic identification procedure was “so impermissibly suggestive as to give rise to a very substantial likelihood of irreparable misidentification.” Simmons v. United States, 390 U.S. 377, 384, 88 S.Ct. 967, 19 L.Ed.2d 1247. This issue is to be resolved only after an examination of the “totality of the circumstances” in the case. Simmons v. United States, supra at 383, 88 S.Ct. 967, 971; United States ex rel. Pierce v. Cannon, 508 F.2d 197, 200 (7th Cir. 1974).

As to the November 2, 1972, showing of five black and white photographs (Exhibit 25) to eyewitnesses Dearborn and Spoerl, Bowie complains that it was impermissibly suggestive because four of the persons portrayed were standing next to a height chart and only two of them were close in height to the shortest man involved in the robbery.¹ Actually, two of the photographs pictured subjects who were 5'8V2" tall, one was of undeterminable height (no chart appeared in that photograph), another was 5'10V2" and the last was 6'2". Only one of the photographs pictured a subject significantly taller than the defendant, and all five photographs depicted men of the same race, approximate age and basic body shape. Defendant does not assert that the investigating agents allowed multiple witnesses to view simultaneously a display which included defendant’s photo, so that the first to choose might influence the others,² nor does he contend the agents indicated to the witnesses that defendant’s photo pictured the person the agents suspected was most likely to be the bank robber. Furthermore, under the circumstances of this display, where all persons pictured were of the same race, approximate age and body type, it was not necessary to include more than five photos in the display. See United States v. Gornick, 448 F.2d 566 (7th Cir. 1971), certiorari denied, 404 U.S. 861, 92 S.Ct. 161, 30 L.Ed.2d 103; United States v. Zeiler, 447 F.2d 993 (3d Cir. 1971); and United States v. Bennett, 409 F.2d 888 (2d Cir. 1969), certiorari denied, 396 U.S. 852, 90 S.Ct. 113, 24 L.Ed.2d 256, all of which involved spreads of six photos. Consequently, the initial showings to Spoerl and Dearborn were not suggestive.

As to the January 11, 1973, showing of the six color photographs (Exhibit 13) to these same two eyewitnesses, Bowie contends that their selection of his photograph was tainted because he was the only person pictured there who was also pictured in the earlier display (Exhibit 25). Bowie claims, in effect, that the witnesses selected his photograph (13.1A) from the second display only because they had previouly seen his image in Exhibit 25. It is not sufficient to render a photographic identification invalid to show simply that defendant appeared in each of two displays. If the displays had been viewed by the .witnesses at the same time and they had been able to make an identification only after they had compared the displays and realized that only one man appeared in both, there would be good cause to suspect the reliability of the identifications. In this case the appearance of defendant’s picture in the prior display does not invalidate the identification from the second set of photographs. United States v. Cooper, 472 F.2d 64, 66 (5th Cir. 1973), certiorari denied 414 U.S. 840, 94 S.Ct. 96, 38 L.Ed.2d 77. In addition, the photographs in the second display were in color and of better quality than Exhibit 25 and Bowie’s photograph therein was more recent than 25.4. The agents might have questioned whether the better quality, more recent color photographs would strengthen or weaken the witnesses’ earlier identifications.

Bowie assails both the January 11th showing of Exhibit 13 to Dearborn and Spoerl and its September 3rd showing to Westmoreland, stating that these displays did not fulfill any investigative need because Bowie had already been arrested and indicted when they took place. A prosecutor need not forego a presentation of a better quality and more recent display to witnesses who have already identified a defendant, nor a presentation of such a display to other witnesses, who have not yet identified anyone, simply because an indictment has been returned. Cf. United States v. Ash, 413 U.S. 300, 93 S.Ct. 2568, 37 L.Ed.2d 619. Defendant has shown no improper motivation on the part of the prosecution in offering the color display to the witnesses after the indictment. The initial showing of the color display to Westmoreland was in all other respects, as well, proper.

Bowie’s contention that the pretrial displays were unfair because they were not coupled with a lineup must be rejected. The failure to conduct a lineup does not invalidate such pretrial photographic identification. United States v. Evans, 484 F.2d 1178, 1186 (2d Cir. 1973).

Under the rationale of Simmons, upon which defendant relies so heavily, the initial identification by each witness is the crucial point. The risk described by Justice Harlan in Simmons is that an initial photographic misidenti-fication, which was the product of improper suggestions, might be perpetuated in any subsequent identification, including that made in court. This is so because a witness might tend to identify the defendant based on his memory of the photograph selected from the first, suggestive display, rather than from actual memory of the event.³ If, however, the initial display is not improper and the witnesses make subsequent selections in some part from memory of the first photograph that they selected, the risk of misidentification is reduced, for the original selection was not a product of impermissible suggestion. See Simmons v. United States, supra, 390 U.S. at 384, 88 S.Ct. 967; United States v. Evans, 484 F.2d 1178, 1184 (2d Cir. 1973); United States v. Hines, 455 F.2d 1317, 1330-1331, 147 U.S.App.D.C. 249 (1972), certiorari denied, 406 U.S. 969. Because there was nothing improperly suggestive in the initial showings to the three witnesses in this case, defendant’s attacks on their later photographic and in-court identifications are not persuasive. Since we find all the photographic identifications to be without the taint of impermissible suggestion, any in-court identification was not the product of-a pretrial procedure “so impermissibly suggestive as to give rise to a very substantial likelihood of irreparable misidentification.” See Simmons v. United States, supra, 390 U.S. at 384, 88 S.Ct. at 971; United States v. Evans, supra, 484 F.2d at 1184. We recently rejected somewhat similar attacks on photographic identification in the appeal of one of Bowie’s co-defendants, Carl Higgans (United States v. Higgans, 507 F.2d 808 (7th Cir. 1974)), and are not persuaded to depart from that holding here.

It is also noteworthy that defendant’s counsel fully explored the identification procedures before the jury while cross-examining the witnesses who had identified defendant. With each witness he explored all of the circumstances surrounding the prior photographic identification and specifically inquired whether the witnesses were able to identify defendant in court from memory of the actual robbery or from pictures which they had selected prior to trial. We realize that a verdict of guilty after wide-ranging cross-examination of identifying witnesses does not automatically dispose of a defendant’s claim of impermissible suggestion in pretrial photographic displays. Where, however, defendant’s counsel has on cross-examination explored and attacked each witness’ basis for identifying the defendant, irreparable prejudice from pretrial photographic identification is not likely. In light of the “totality of the circumstances” involved in this case, the district court correctly admitted both the earlier photographic and the in-court identifications.

Bowie also submits that he had a constitutional right to assistance of counsel at the post-indictment displays on January 11, 1973, to Dearborn and Spoerl and September 5, 1974, to West-moreland. This point is without merit. United States v. Ash, 413 U.S. 300, 93 S.Ct. 2568, 37 L.Ed.2d 619; United States v. Robinson, 406 F.2d 64, 67 (7th Cir. 1969), certiorari denied, 395 U.S. 926, 89 S.Ct. 1783, 25 L.Ed.2d 243.

Refreshing Witness Dearborn’s Testimony

Next Bowie contends that the trial court abused its discretion in allowing George Dearborn to refresh his recollection as to which robber went behind the tellers’ cages. He initially testified that the man who went behind the tellers’ cages was the middle-sized robber. The next day he testified that he was uneasy as to whether that was a correct statement. On redirect examination, Dearborn was permitted to review Government Exhibit 28, a summary of an FBI interview with him on the day of the bank robbery over a year before the trial. Thereafter Dearborn testified that the smallest of the three men went behind the tellers’ cages and scooped the money out of the tellers’ counters. This was a permissible trial tactic, and the district court did not err in allowing Dearborn to view Exhibit 28. Kirschbaum v. United States, 407 F.2d 562, 566 (8th Cir. 1969); McCormick on Evidence, § 9, p. 19 (1954); 3A Wigmore on Evidence, § 904, p. 675 (Chadbourn rev. 1970).

Cross-examination of Witness Serazio

Bowie also asserts that he was denied effective cross-examination of government witness Charlotte Serazio. She testified that her prior photographic and in-court identification of co-defendant Higgans was based on his eyes. Since Higgans was not present in court due to the successful severance motion, Bowie claims that he was denied a meaningful opportunity to cross-examine Serazio. However, Higgans’ photograph (Government Exhibit 10) was available for the jury to compare with her description of Higgans, and Bowie could have subpoenaed that co-defendant to appear if he wished to challenge the accuracy of her description. Therefore, Serazio’s testimony was properly admitted.

Receipt of Stipulations 2, S and 5 into Evidence

Bowie argues that stipulations 2, 3 and 5 were irrelevant and should not have been received in evidence.

Stipulation 3 states that on the night of October 31, 1972, Lamont Butts, Michael Johnson and Curtis Jones registered as Milwaukeeans at the Sherman House Motel in Chicago and that the fingerprints of “Michael Johnson” were those of co-defendant Higgans.

Stipulation 2 states that on the following day, these three persons flew from Chicago to Los Angeles, paying individual fares of $378 each.

Stipulation 5 states that in November 1972 Lohman Brooks, a resident of a suburb of Los Angeles, California, was visited in El Segundo, California, by co-defendant Higgans, another individual called “Mr. Bowie,” and a third person. The stipulation also shows that the three individuals bought a 1971 Cadillac in Brooks’ name with a $1500 cash down payment, and that co-defendant Higgans counted out that payment.

These stipulations were received in evidence following testimony of Robert Walrath, a Milwaukee roommate of Bowie’s brother Donald. Walrath testified that appellant Paul Bowie came into their apartment to talk to Donald Bowie between 2:30 and 4:30 a. m. on October 31, 1972. On November 4th, the following Saturday, Walrath received a long-distance collect telephone call from Paul Bowie. In that conversation Walrath was asked to tell Donald Bowie to call his brother Paul at a certain telephone number with a 312 area code ⁴ and to ask for Lamont Butts. From Walrath’s testimony, the jury could infer that Lamont Butts was Paul Bowie. The stipulations were admissible as evidence of flight and use of an alias and could be considered to support an inference of guilt. United States v. Pate, 342 F.2d 646 (7th Cir. 1965), reversed on other grounds, 386 U.S. 1, 87 S.Ct. 785, 17 L.Ed.2d 690. The stipulations were also properly received as evidence of defendant’s wealth, showing that he was one of the persons paying $378 for an airline ticket to California and also one of the three purchasing a 1971 Cadillac under another’s name. No showing of Paul Bowie’s prior impecuniousness was required. United States v. Higgans, 507 F.2d 808 (7th Cir. 1974); United States v. Crisp, 435 F.2d 354, 360 (7th Cir. 1970), certiorari denied, 402 U.S. 947, 91 S.Ct. 1640, 29 L.Ed.2d 116.

Aiding and Abetting Instructions

Finally, Bowie asserts that the court’s instructions 8 and 9 regard-ing multiple defendants were prejudicial to him. These instructions provided as follows:

“In the case where two or more persons are charged with the commission of a crime, the guilt of any defendant may be established without proof that the accused personally did every act constituting the offense charged.”

* H! * ‡ ¡H &

“You of course may not find any defendant guilty unless you find beyond reasonable doubt that every element of the offense as defined in these instructions was committed by some person or persons, and that the defendant participated in its commission.”

Bowie claims that by the use of the phrase “any defendant” in these instructions, the jury was permitted to determine the guilt of co-defendant Higgans and could use that determination to Bowie’s prejudice. Since defendant did not make such an objection to these instructions, any error in this respect has been waived. United States v. Kurfess, 426 F.2d 1017, 1020 (7th Cir. 1970), certiorari denied, 400 U.S. 830, 91 S.Ct. 60, 27 L.Ed.2d 60; Rule 30, F.R.Crim.P. Furthermore, other portions of the charge told the jury that it must address itself to the guilt or innocence only of Bowie and that the jury could not consider him an aider and abetter unless it found that he “was a participant and not merely a knowing spectator.” Thus taking the instructions as a whole, the jury could not have used Higgans’ guilt as proof against Bowie and, therefore, the assailed instructions do not constitute plain error under Rule 52(b), F.R.Crim.P.

The judgment of conviction is affirmed.

1 . Bowie was one of those pictured before tall. height chart, which showed him to be 5'872"

2 . See United States v. Wade, 388 U.S. 218, 234, 87 S.Ct. 1926, 18 L.Ed.2d 1149; Note, Pretrial Identification Procedures — Wade to Gilbert to Stovall: Lower Courts Bobble the Ball, 55 Minn.L.Rev. 779, 795 (1971).

3 . Because we have found that the displays were not impermissibly suggestive, we need not consider the factors set out in Neil v. Biggers, 409 U.S. 188, 93 S.Ct. 375, 34 L.Ed.2d 401, and discussed in United States ex rel. Pierce v. Cannon, 508 F.2d 197, 204 (7th Cir. 1974), which aid a court in determining whether a suggestive pretrial identification taints subsequent in-court identification.

4 . The witness misspoke; the Los Angeles area code is 213.

---

Dissenting Opinion

WILLIAM J. CAMPBELL, Senior District Judge

(dissenting).

I respectfully dissent. Despite the use of inherently suggestive pretrial photographic identification procedures, not a single witness was able to state with any degree of certainty that the defendant’s photograph was that of one of the bank robbers. Nor could a single witness positively identify the defendant in the courtroom. After thoroughly reviewing a trial record conspicuously bereft of any truly corroborative evidence, I am left with the unsettling feeling that almost any 5'7" to 5'9", slightly built, male negro who could not disprove his presence in Milwaukee, Wisconsin on October 31, 1972, could just as surely have been convicted of robbing the University National Bank as was Bowie. The criteria prescribed in Simmons v. United States, 390 U.S. 377, 88 S.Ct. 967, 19 L.Ed.2d 1247 (1968) for determining whether an in-court identification may be made by a witness who has previously viewed photographs of the defendant, and the requirements of Neil v. Biggers, 409 U.S. 188, 93 S.Ct. 375, 34 L.Ed.2d 401 (1972) for determining the admissibility of extrajudicial identification evidence, were designed to prevent this eventuality and, in my view, require that the defendant’s conviction be reversed.

Simmons concerned the propriety of allowing witnesses to identify the defendant in the courtroom after they had viewed his photograph in the course of an investigation. (Evidence of the ex-trajudical photographic identification was not introduced.) Cautioning that “each case must be considered on its own facts”, the Simmons Court held that:

“[Cjonvictions based on eyewitness identification at trial following a pretrial identification by photograph will be set aside on that ground only if the photographic identification procedure was so impermissively suggestive as to give rise to a very substantial likelihood of irreparable misidentification.” 390 U.S. at 384, 88 S.Ct. at 971.

The word “irreparable” is essential to the holding in Simmons, for even if the pretrial identification procedures are found to be suggestive, it does not necessarily follow that the in-court identification must be prohibited. It may be that other facts provide an independent basis for the in-court identification.

But where evidence of the extrajudicial identification is also presented, the standard for suppression is somewhat different. In Neil v. Biggers, supra, the Court noted that:

“[T]he primary evil to be avoided is ‘a very substantial likelihood of irreparable misidentification.’ Simmons v. United States. . . . While the phrase was coined as a standard for determining whether an in-court identification would be admissible in the wake of a suggestive out-of-court identification, with the deletion of ‘irreparable’ it serves equally well as a standard for the admissibility of testimony concerning the out-of-court identification itself.” 409 U.S. at 198, 93 S.Ct. at 381.

Since evidence of pretrial photographic identification ¹ was introduced in the instant case, we must look to Neil to determine whether this evidence should have been excluded. This determination requires first that the court discern whether the identification procedure was suggestive, and if so, “whether under the ‘totality of the circumstances’ the identification was reliable even though the confrontation procedure was suggestive.” 409 U.S. at 188, 93 S.Ct. at 382. If, upon consideration of the factors set forth in Neil, the pre-trial identifications are found to be unreliable, the conviction must be reversed, regardless of whether the courtroom identifications of the defendant were proper under Simmons.²

SUGGESTIVENESS

The photographic identification procedure employed in this case constituted a suggested confrontation. The first spread of photographs shown to witnesses was a selection of five black and white pictures. It was stipulated that, at the time witnesses Westmoreland, Dearborn, Knabe and Spoerl were shown the photospreads (Exhibits 13 and 25), they were told by FBI Agents Thorne and Walsh that the pictures pertained to the shorter individual (in relation to the other robbers) who allegedly moved behind the tellers’ cages during the robbery. Rather than being asked if any of the pictures depicted the aforementioned robber, the witnesses were requested only to select the picture which “most closely resembles” the individual who moved behind the counter (Tr. pp. 75, 81, 113-116, 239, 327-28, inter alia).

To lend further suggestiveness to the procedure, all but one of the photographs depicted an individual standing next to a height chart. Although the witnesses agreed on little else, they did concur that the shorter robber stood between 5'7" and 5'9" in height and had a 'slight build. The five black and white-photos comprising Exhibit 25 depict one individual who stands at a height of 6'2"; another who stands at a height of 5'10V2" and has a heavy, muscular build; and a third whose picture is so out of focus that the “mug shot” lettering is barely readable. The individual depicted in the latter photograph is not standing beside a height chart. Thus, the witnesses were actually shown only two pictures (25.1 and 25.4) which depicted individuals of the described height and build. Having been asked to select the picture which “most closely resembles” the shorter robber who moved behind the counter, the witnesses might well have thought that they did not even have the option of rejecting both of the remaining photographs. (See, for example, the testimony of Sandra Knabe at Tr. pp. 323 and 327).

Of the six color photographs comprising Exhibit 13, two are of individuals whose height exceeds 5T0". While the remaining four comprise a more representative sampling of individuals meeting the description of the bank robber, the witnesses were again asked only to select the picture which “most closely resembles” the shorter robber. No picture included in Exhibit 25, other than Bowie’s, was repeated among the pictures in Exhibit 13, and the color photographs were shown to the witnesses after Bowie had been taken into custody, at a time when a physical line-up could easily have been conducted instead of the more inconclusive method of identification by photograph.³

The identification procedure was rendered suggestive by repeating only the defendant’s picture in the subsequent photo-spread (Exhibit 13), and by the fact that the witnesses were advised as to which of the three robbers the pictures were intended to represent. Further, use of the height-charts and, in one instance, a discernibly out-of-focus picture, represented highly suggestive techniques which, through the process of elimination, served to highlight the defendant’s photograph. To then require only that the witnesses select the photograph which most closely resembled the robber increased the possibility of mis-identification. I find no difficulty in concluding that the procedure employed was suggestive,⁴ and therefore turn to the factors set forth in Neil for determining whether “the identification was reliable even though the confrontation procedure was suggestive.”

The Neil decision identifies five factors to be considered in “evaluating the likelihood of misidentification”:

(1) The opportunity of the witness to view the criminal at the time of the crime.

(2) The witness’ degree of attention.

(3) The witness’ prior description of the criminal.

(4) The level of certainty demonstrated by the witness at the confrontation.

(5) The length of time between the crime and the confrontation.

A review of the record discloses that, on the basis of the first four factors, the pre-trial photographic identification of the defendant was unreliable.

(1) Opportunity to view the criminal at the time of the crime.

Each of the witnesses testified that the bank robbers wore masks, and I believe the record clearly establishes that the full face of the robber in question was at no time visible. Only two of the witnesses, Westmoreland and Spoerl, testified that they saw the mask slip at all. Westmoreland testified that the mask, which extended over the robber’s nose shielding his lower face, “slipped a little” (Tr. p. 27). The incident took no longer than a few seconds. Although he testified that it was at that time that he saw the robber’s face, further questioning established that he had not seen the robber’s full face; that the mask slipped only as far as the robber’s lip (Tr. p. 70); and that while it was thus possible for him to determine that the robber wore a mustache, he was not able to discern whether he also wore a beard (Tr. p. 80).

Westmoreland’s ability to observe the criminal was further retarded by two additional factors. According to this witness, the robber wore “a wide — a big hat” (Tr. p. 26) which came down as far as his midforehead: “and the way he had his hat on [I] couldn’t tell how his hair was — ”. Secondly, Westmoreland acknowledged that although he wears glasses almost every day, he had not been wearing glasses on the day of the robbery (Tr. pp. 82-83).

Witness George Dearborn had even less of an opportunity to view the criminal. Asked if he could describe the man who went behind the tellers’ cages, he responded: “No, this is the one man I would have the most difficulty describing. Other than his height and approximate weight, I don’t remember anything significant about him” (Tr. p. 105). In explaining the characteristics which caused him to select defendant’s photograph as that which most closely resembled the robber, Dearborn testified: “The nose and the eyes, because really that’s all I could see. The head was covered by a cap and the lower portion of the face was covered by a scarf” (Tr. p. 175).

Witness Cindy Spoerl saw the robber’s mask (which she described as a scarf) slip to a point “just below his nose” (Tr. p. 261). She did not notice whether he had a mustache (Tr. p. 233). She testified that she observed the robber who went behind the tellers’ cages for only a “few seconds”, after which she moved to a corner in the hallway where she remained for the duration of the robbery. Thereafter, she had no further occasion to see the robber in question (Tr. p. 234).

Sandra Knabe testified that she and other bank employees were directed by the tall robber to move from the tellers’ cages to an area near the vault, and that only after she had done so did she observe the shorter robber move behind the tellers’ counter. She testified that approximately twenty people were herded into a 10' by 4' area near the vault (Tr. p. 310), and that her ability to observe was obscured because she was standing “toward the wall” (Tr. p. 303) while other employees, some of whom were taller than she, were standing in front of her (Tr. p. 310). She further testified that she did not spend a long time observing the shorter robber (Tr. p. 304) because “I was looking at other people and I looked at the (taller) man that was holding the gun” (Tr. p. 310).

A review of the record clearly establishes that, while the robbery consumed five or six minutes, no witness observed the robber behind the tellers’ cages for any appreciable length of time. All observed that he was masked, and no one spoke with him. As will be discussed infra, the limited extent to which the witnesses were able to observe the robber is reflected in the remarkably vague descriptions given by the witnesses after the incident and at trial. On the basis of the first factor set forth in Neil, the pre-trial photographic identifications were not reliable.

(2) The witness’ degree of attention.

The record indicates that Knabe was understandably more preoccupied by the taller robber who was pointing a gun at her. During most of the robbery, Spoerl did not see the shorter robber at all, since she did not have an opportunity to observe him after she moved to the hallway corner at the taller robber’s direction (Tr. p. 234). Dearborn testified that the shorter robber remained behind the tellers’ cages for about a minute-and-a-half, and that he did not watch the robber during that entire period: “No, I wasn’t watching during the entire time; I was looking around the lobby” (Tr. p. 164). Only Westmoreland’s testimony reflects any particular degree of attention paid to the shorter robber, and this fact is rendered almost inconsequential by the fact that he was not wearing his glasses, and by the difficulty which he encountered in his attempts at in-court identification.

(3) The accuracy of the witness’ prior description of the criminal.

The descriptions given by the witnesses of the robber behind the tellers’ counter were vague, to say the least: 5'7" to 5'9" in height; male; negro; slightly built. It is a description which might fit any number of Milwaukee’s citizens. But what is perhaps even more significant in terms of the reliability of subsequent photographic identification is the witnesses’ inability to either agree or express an opinion regarding various other descriptive characteristics which were observable during the robbery.

Westmoreland and Knabe testified that the robber wore a wide-brimmed hat (Tr. pp. 26 and 311); Dearborn stated that he wore a stocking cap (Tr. p. 176); Spoerl could not recall whether or not he wore any kind of head covering (Tr. p. 260). Westmoreland was at first almost positive, and later less certain, that the robber did not wear glasses (Tr. p. 84); no one else could recall whether he did or not. None of the witnesses could remember whether he wore gloves. Spoerl could describe neither the kind nor the color of the robber’s clothes (Tr. p. 260) except that they were “shabby” (Tr. p. 269); Dearborn could not describe the kind of clothes, but believed they were dark in color (Tr. p. 184); West-moreland testified that the robber wore a three-quarter length tan cashmere coat (Tr. p. 26). Only Westmoreland and Spoerl recalled that the robber’s mask slipped from his nose. Westmoreland stated that he wore a mustache — Spoerl could not recall seeing one.

Two witnesses described the robber’s mask as a scarf; one described it as a knitted ski mask; the fourth remembered it as a handkerchief. Dearborn could not remember “anything significant about him” other than his “height and approximate weight” (Tr. p. 105).

The government’s case depended entirely upon the witnesses’ ability to identify Bowie as one of the robbers. Yet, because the robbers wore masks and hats, the witnesses were unable to provide specific information regarding most of the robber’s facial characteristics. This fact, combined with the witnesses’ inability to recall or agree regarding such clearly observable identification factors as the foregoing, demonstrates that the pre-trial photographic identifications were not reliable.

(4) The level of certainty demonstrated by the witness at the confrontation.

As previously discussed, due to a number of circumstances surrounding the robbery, including the fact that the robber was masked, the witnesses did not have a particularly good opportunity to view the criminal at the time of the crime. Their attention was distracted, and they were unable to describe discernible identification factors (e. g., clothing) in an accurate, detailed or consistent manner. Under these circumstances, even a certain identification of the defendant at the time of confrontation would be suspect and only marginally reliable. But where, as here, the level of certainty at confrontation is, at best, equivocal, I find no basis for concluding that the identification is anything but utterly unreliable.

Cindy Spoerl testified that when the FBI agents showed her Exhibit 25, she selected pictures 25.1 and 25.4 as those which “most closely resembled” the bank robber. She made the same determination in her selection of picture 13.1(a). With respect to pictures 25.1 and 25.4, she was not “able to say that the persons depicted in those photographs were the people that participated in the robbery” (Tr. p. 265).

Cornell Westmoreland viewed a spread of five black and white photographs shortly after the robbery. It is unclear from the record whether these pictures were the same as those shown to Spoerl and Dearborn on November 2, 1972 (Exhibit 25). Westmoreland was unable to identify any of those depicted in this photo-spread as one of the bank robbers (Tr. pp. 33-34).

On September 5, 1973, almost a year after the robbery, Westmoreland was shown the six photographs which comprise Exhibit 13. He was asked to select the picture which most closely resembled the bank robber behind the counter, and selected 13.1(a) [Bowie’s picture] as “resembling” that individual. His selection was based upon “the size and the face features and the complexion” (Tr. p. 47), and was to “a certain degree positive, I really could not set a degree” (Tr. p. 47). ■Thus,'Westmoreland was, to “a certain degree positive” that the person depicted in photograph 13.1(a) “resembled” the robber.

At trial, Dearborn was shown Exhibits 25 and 13, which he had viewed on November 2, 1972 and January 11, 1973, respectively. He could not recall whether he selected 25.1 or 25.4 (25.4 is Bowie’s picture; 25.1 is the only other photograph in Exhibit 25 that closely fits the witnesses’ description of the robber). Regarding Exhibit 13, Dearborn testified that he had selected picture 13.2(a). Bowie’s picture is 13.1(a). The government was permitted at a later point in the trial to show, through FBI Agent Thorne’s testimony, that Dearborn initially selected both 13.1(a) and 13.2(a) as “individuals who resembled” one of the bank robbers. According to Thorne, Mr. Dearborn later amended his selection to include only picture 13.1(a).

Thus, not only were Dearborn’s initial identifications uncertain ones, but the witness was unable to testify with any certainty or accuracy regarding which photographs were actually selected.

Finally, Sandra Knabe was shown Exhibit 25 (the black and white pictures) shortly after the robbery. She testified that she selected picture 25.4 (Bowie) “according to size and color”. Asked if she had an impression as to the “strength of the resemblance”, she responded: “there — just—I don’t know how to express it. Not that I thought that that was him, but that that would fit most closely” (Tr. p. 323).

She further testified that “I told them (the agents) that I couldn’t identify him, but the one that would fit my recollection the best, as far as those pictures went, would be that one” (i. e. 25.4) (Tr. p. 327).

The witnesses were asked to select the photo which “most closely resembles” the robber. They did so on the basis of height, color, complexion and in some instances body size — hardly what might be considered selective criteria. Even then, the best that could be said by any witness was that picture 25.4 and/or 13.1(a) “resembled” the robber in question. In short, the witnesses failed to manifest a “level of certainty at the confrontation” sufficient to render the extrajudicial photographic identifications reliable.

On the basis of the foregoing analysis of the record, I would hold that the confrontation procedures employed in this case were sufficiently suggestive to warrant an inquiry into whether the photographic identifications of the defendant were reliable. And upon consideration of the factors set forth in Neil v. Big-gers, I would conclude that the photographic identifications were not reliable, and evidence thereof should have been suppressed.

In-Court Identifications

On the basis of Simmons v. United States, supra, I do not believe the witnesses should have been allowed to “identify” the defendant in open court as one of the bank robbers. Having concluded that the pre-trial identifications carried a very substantial likelihood of misidentification, the only question that remains is whether such misidentification is “irreparable”. That determination depends upon whether there existed a basis, independent of the photographic identification, for the witnesses’ conclusions that Bowie resembled one of the robbers.

The record in this case, as it relates to the factors discussed previously regarding the reliability of photographic identifications, refutes any suggestion that an independent basis existed for the in-court identifications. In addition, the witnesses viewed defendant’s photograph on several occasions after the robbery, including prior to and during the trial of Carl Higgans, thus reinforcing in their minds the conception that the person depicted in the photograph was the robber who went behind the teller’s counter. Still, the witnesses could testify only that Bowie “resembled” the robber.

The testimony of George Dearborn is illustrative of the problem. He testified that he had the pictures which he selected “fairly clearly in mind”, and that if those pictures were placed in a grouping of other pictures, he would be able to identify those which he had previously selected.

“Q. Now, Mr. Dearborn, I’d like to ask you to think in your own mind when you conjure up a picture of one or more of the robbers on that day, do you think of a person you saw in the bank or are you thinking of the photographs you have selected?

A. Very hard to separate the two” (Tr. p. 118).

Asked at a later point if he could “clearly separate in your own mind the impressions you formed of the identification of the robbers on the day of the robbery and the images of the photographs you have in your mind from seeing them several times,” Dearborn responded: “that’s very difficult” (Tr. p. 180).

The absence of an independent basis for in-court identification is even more clearly reflected in the testimony of Cornell Westmoreland. During direct examination, the following colloquy occurred:

“Q. Mr. Westmoreland, can you identify anyone in the courtroom today as being the five-foot-nine gentleman that you have described thus far?

A. No, sir.

Q. Is there anyone in the courtroom who looks familiar to you in that regard?

A. No, sir.” (Tr. p. 28).

Westmoreland was then shown Exhibit 13, and testified that on September 5, 1973, he had selected picture 13.1(a) as “resembling” the robber on the basis of “the size and the face features and the complexion” (Tr. p. 47). During redirect examination, the witness, having viewed defendant’s photograph once again, was allowed to step down from the witness stand and observe the defendant at even closer range. Having done so, he testified that the defendant had certain features which were similar to the bank robber’s features: forehead, height and build. The record reflects the following colloquy between the prosecutor and the witness:

“Q. The gentleman in Exhibit 13.1 you were shown on September 5, 1973, his facial features, mous-tache and height are also about the same as the person who went behind the teller’s cage: Right?

A. Right. (Tr. p. 89).

Q. Having seen Mr. Bowie when you went off the witness stand and looked at him after he took- off his glasses,⁵ can you identify him as being one of the bank robbers?

A. Positive: no, sir.

Q. Can you indicate whether he resembles one of the bank robbers?

A. Yes, sir (Tr. p. 91).

Q. In fact, does he resemble the person in Exhibit 13.1?

A. Yes, sir.

Q. Looks very much like the same guy: Right?

A. Yes, sir. (Tr. p. 92).

Having “singled out” the defendant to a witness who had been unable to select anyone in the courtroom who resembled the robber, the government was allowed to lead the witness into acknowledging that Bowie’s features resembled those of the bank robber because after all, he bore such a striking resemblance to the person depicted in the photograph. The suggestiveness of this in-court “show-up” is even more extreme than the pre-trial showup procedure which the Supreme Court condemned in Foster v. California, 394 U.S. 440, 89 S.Ct. 1127, 22 L.Ed.2d 402 (1968). The fact that Bowie’s confrontation occurred in the courtroom, while Foster’s took place in a police station, is a distinction without a difference. Foregoing the temptation to further digress upon the propriety of the government’s redirect examination, suffice it to say that Westmoreland’s in-eourt identification was obviously based upon photograph 13.1(a) and not upon an independent recollection of the robber.

The majority notes that the defendant’s counsel “fully explored the identification procedures before the jury while cross-examining the witnesses Where . . defendant’s counsel has on cross-examination explored and attacked each witness’ basis for identifying the defendant, irreparable prejudice from pretrial photographic identification is not likely”.

I must respectfully disagree. Where suggestive pre-trial identification procedures have been employed and the surrounding circumstances of the crime render that identification unreliable, there exists a very substantial likelihood of irreparable misidentification in court, unless the record shows that the in-court identification is based upon independent data, e. g., the witness’ observations at the time of the crime. I do not believe that Simmons and Neil can be interpreted in a manner consistent with one another without reaching this conclusion. Vigorous cross-examination is pertinent only in terms of the witness’ credibility, i. e., whether the jury believes the identification was erroneous. But the rules prescribed by Simmons and Neii are designed to prevent the presentation of such testimony in the first instance. Simmons recognized that the Supreme Court’s decision in United States v. Wade, 388 U.S. 218, 87 S.Ct. 1926, 18 L.Ed.2d 1149 (1967) and Gilbert v. California, 388 U.S. 263, 87 S.Ct. 1951, 18 L.Ed.2d 1178 (1967) “departed from the rule that the manner of an extra-judicial identification affects only the weight, not the admissibility, of identification testimony at trial.” 390 U.S. at 382, 88 S.Ct. at 970. Thus, hindsight observations regarding the extent and vigorousness of cross-examination are inapposite.

CONCLUSION

On the basis of Neil and Simmons, I would reverse defendant’s conviction. Most of the “exclusionary rules” which have been developed over the past two decades are rooted in the need to “deter future unlawful police conduct and thereby effectuate” the constitutional rights guaranteed to all citizens under the Fourth, Fifth and Sixth Amendments. United States v. Calandra, 414 U.S. 338, 347, 94 S.Ct. 613, 38 L.Ed.2d 561 (1974). In their application, these rules necessarily benefit the criminal and the innocent alike.

But the exclusionary rules developed through the Neil and Simmons decisions have a somewhat different motivation. They demonstrate a recognition that suggestive identification procedures can easily result in misidentification; that eye witness identification testimony is perhaps the strongest and most convincing evidence in the minds of jurors; and that an innocent person who has been erroneously identified therefore stands a strong chance of being erroneously convicted. The criteria set forth in Neil and Simmons are thus designed to exclude false evidence, thereby reducing the chance that an innocent person will be convicted.

I offer these observations because, on the basis of this record, I am struck by the clear possibility that the photographic and in-court identifications in this case may have produced the very result which Simmons and Neil were designed to prevent, the conviction of an innocent man.

The government’s case against Bowie was a house of cards consisting of pretrial photographic selections of the person “most closely resembling” the robber, and in-court identifications of the defendant as a person who resembled the robber on the basis of such generally applicable characteristics as race, height, weight and complexion. No fingerprints were introduced into evidence, and no photographs of the crime in progress were available because the bank’s internal camera was not operating on the day of the robbery. Although Dearborn acknowledged that marked money (referred to as “bait money”) was stolen in the course of the robbery (Tr. p. 167), none was introduced into evidence to link Bowie with the crime.

The evidence which was introduced by the government, other than the identification testimony, was highly conjectural and certainly insufficient to withstand a motion for a directed verdict of acquittal. Since the government’s case against Bowie thus rested entirely upon the identification testimony of the witnesses, and since, in my opinion, the identification procedures employed in this case created a “very substantial likelihood of misidentification”, I find the conclusion inescapable that Bowie may well have been convicted of a crime he did not commit. At any rate, I believe that Neil and Simmons require that the conviction be reversed.

1 . The words “identification" and “identify” are used here somewhat loosely, primarily as a matter of convenience. As will be more fully discussed infra, no witness was able to positively identify the defendant’s photograph as that of one of the robbers, and no witness was able to positively identify the defendant in court.

2 . In such cases, the only purpose for determining whether there existed an independent basis for the courtroom identification is to discern whether, at any future trial, a courtroom identification may be permitted.

3 . “Because of the inherent limitations of photography, which presents its subject in two dimensions rather than the three dimensions of reality, and which presents a ‘frozen’ image, a photographic identification, even when properly obtained, is clearly inferior to a properly obtained corporeal identification. Consequently, witnesses should be asked to examine photographs only when a proper corporeal identification is impossible ... or difficult. In any other case, the use of photographs is improper, for it constitutes the unnecessary employment of an identification procedure clearly inferior in reliability to one which is available.” P. Wall, Eye-Witness Identification in Criminal Cases, 70 (1965). See also, Simmons v. United States, supra, at 386, n. 6, 88 S.Ct. 967.

4 . This court’s recent decision in United States ex rel. Pierce v. Cannon, 508 F.2d 197 (1974) strongly criticized certain aspects of a pre-trial confrontation (in that case, a line-up) as being suggestive, noting that “[i]f we were formulating rules to be applied by federal courts in federal cases the foregoing considerations might very well compel us to adopt, in the exercise of our supervisory power for the courts of this circuit, the exclusionary rule petitioner suggests.” At p. 202. The petitioner had argued that because the pre-indictment line-up in that case was unnecessarily suggestive, identification testimony stemming therefrom should be per se excluded.

I would not favor a rule which excludes pretrial identification evidence solely on the ground that the pre-trial identification procedures were suggestive, unless they were so purposely and blatantly suggestive as to constitute an effort to “manufacture” evidence. In most cases, the Neil approach should be followed: the court should determine whether the identification was reliable, even though the procedure was suggestive. Since reliability of identification should be the court’s principal concern, however, any identification procedure which falls short of a completely objective confrontation should be viewed as sufficiently suggestive to trigger an inquiry into whether the identification was nevertheless reliable.

In Pierce, for example, it appears that the pre-trial confrontation was considerably Jess suggestive than the photographic identification procedure employed in the instant case, but was nevertheless sufficient to warrant an inquiry, on the basis of the factors set forth in Neil, into whether the pre-trial identification was reliable.

5 . The “identification” was not necessarily enhanced by removal of the defendant’s glasses, since Westmoreland testified that he was “really not sure” whether or not the robber wore glasses. (Tr. p. 84). What is certain, as the prosecutor observed, is that the individual depicted in picture 13.1(a) is not wearing glasses (Tr. p. 84).