Joseph Sanchell was convicted after trial by jury in Nebraska state court of one count of forcible rape and three counts of robbery. On direct appeal, the Nebraska Supreme Court first reversed the convictions, holding that one witness' in-court identification should have been excluded. State v. Sanchell, 191 Neb. 505, 216 N.W.2d 504 (1974) (Boslaugh and Spencer, J.J., dissenting). On rehearing, however, the conviction was reinstated, 192 Neb. 380, 220 N.W.2d 562 (1974) (Clinton and McCown, J.J., dissenting). Sanchell's habeas petition was denied after an evidentiary hearing in an unreported memorandum opinion by the federal district court. This appeal followed.
Petitioner urges that the identification evidence should have been suppressed because it was derived from a series of pretrial identification confrontations so suggestive as to make identification of him inevitable. The State urges us to sustain the findings of the Nebraska Supreme Court and the federal district court that even if the confrontations were unduly suggestive, the in-court identifications were reliable in any event. We hold that the identification testimony of two of the three witnesses should have been excluded, and thus the petitioner is entitled to a new trial.
The charges against Sanchell stem from events in the early morning hours of January 22, 1972, when a black male broke into five separate sleeping rooms of a women's dormitory on the University of Nebraska's Lincoln campus. The intruder raped and robbed the young woman in the first room, and later robbed two other women students. Sanchell was arrested on these charges several weeks later after one witness had identified him at his arraignment on an unrelated charge.
The facts leading up to the identification are significant. On Saturday morning, January 22, 1972, the intruder entered the dormitory between 5:30 and 5:45 A.M. and remained inside until about 6:30 A.M. The sun did not rise that day until 7:43 A.M. so no early morning light would have been present during the crimes.
The intruder entered Renee's second-floor room first, and closed the door. Renee did not awaken until he placed his hand over her mouth while she lay in bed. He ordered her not to look at him, but at trial she testified that she did stare at his face for about five seconds, then looked away, then stared at him again for another five seconds. At this time, the drapes and the door were closed and no lights were on in the room, so the only sources of illumination would have been light entering from the hall under the door or from the floodlit courtyard under the drapes. The intruder then blindfolded Renee and turned on a light. She could see his body from under the blindfold, but apparently she did not thereafter see his face again clearly. Renee was raped twice and robbed. The intruder remained in her room about 30 to 45 minutes and talked to her throughout that time.
Leaving Renee's room at 6:15 A.M., he entered Mary Ann's nearby room and robbed her, remaining in the room only a short time. Mary Ann had poor eyesight and never heard the intruder's voice nor saw him clearly, so she did not identify anyone later.
Next, the intruder entered Sharon's room. The sound of the opening door awakened Sharon, and she saw a tall black man silhouetted in her doorway, where he remained for about 10 seconds. She asked him what he was doing and he closed the door and left without answering. Sharon never identified the petitioner as the intruder.
The intruder proceeded up to the third floor, and opened the door of Ann's room. She awoke immediately and saw him silhouetted in the doorway for a second or two. He entered the dark room, closing the door behind him, then pulled her out of bed and walked behind her over to a light switch beside a mirror. When he turned on the light, Ann caught a brief glimpse of "his eyes and his face, just the outline of his face" reflected in the mirror before she closed her eyes due to his threats. He left after ordering her to give him her money.
The fifth and final room entered was Patricia's, next door to Ann's. Patricia awoke in her dark room when the intruder placed his hand over her mouth. No lights were on and her drapes were closed, although she testified at trial that the courtyards were floodlit and that her light-colored drapes did admit some light at the time. When Patricia awoke, she saw the intruder's profile for "perhaps three seconds", but she had no chance to see his full face or body at any time because he blindfolded her almost immediately. He remained in her room about five minutes, talked to her from time to time, robbed her, and then left the dormitory without being apprehended.
These events were reported soon thereafter to the University Campus Security Department. Security Officer Edmunds interviewed the witnesses about 12 hours later. The record does not show what if any description the witnesses other than Renee were able to provide at that time. Renee helped Officer Edmunds prepare a composite picture of the suspect using overlay foils. The composite was circulated to the Lincoln, Nebraska Police Department that evening, and one officer noticed that the picture and the description generally fit Joseph Sanchell, who had then recently been arrested and released on an unrelated robbery charge.1
Over the next few days, Officer Edmunds showed each of the witnesses photographs of possible suspects, with a very recent mug shot of Sanchell included in a spread of eight. The photograph display was fairly conducted and is not challenged on appeal. None of the witnesses was able to make a positive identification at this point. Patricia picked out three photos as resembling the intruder, none of which was Sanchell. Ann similarly picked out several photos which she said resembled the intruder, but did not include Sanchell's picture among them. Only Renee included the petitioner's photo along with three others that she chose as possible suspects.
On February 7, Renee told Officer Jacobson of the Lincoln Police Department that she was not sure she would be able to identify her assailant's face, but that perhaps she would be able to recognize his voice.
The next day, February 8, 1972, the first of the challenged showups occurred. Either the police or the security officers called Renee and arranged for her to contact the other witnesses to ask them all to appear at the County Courthouse in Lincoln to identify someone. At the Courthouse, an officer ushered the five young women into a courtroom, telling them to watch the proceedings and not talk among themselves. Then in progress was Joseph Sanchell's arraignment on the unrelated robbery charge for which he had previously been arrested. Sanchell, who was not then represented by counsel and had not yet been arrested for or charged with the January 22 rape-robberies,2 sat at the counsel table with his back to the witnesses and did not speak. He was the only tall heavy black male in the room. After the arraignment, the witnesses were separately interviewed. Of the five, only Renee positively identified Sanchell after the arraignment. Two of the young women, Sharon and Patricia, told the officers at that time that Sanchell was not the intruder.
On the strength of Renee's identification, Sanchell was arrested and charged with the January 22 rape-robberies. Despite the uncertainty of the other witnesses, Sanchell was never placed in a lineup to see if the young women could pick him out from others fitting the general description, nor was he ever asked to speak for the witnesses to see if, given the chance to hear his and other black male voices clearly, they could identify him on that basis. Strangely enough, during his custody, the police did place Sanchell in other lineups for other unrelated crimes.
Three weeks after the arraignment, on March 1, 1972, he was again shown singly to the witnesses. On that date, two preliminary hearings were held, one for the November robbery charge and another for the January rape-robbery charges with which we are concerned. The five witnesses to the January crimes were present from the beginning. Since the November charge was to be heard first, defense counsel asked that the rape-robbery witnesses be excluded. The prosecutor resisted this motion, however, and the court denied it. As a result, the five witnesses watched while another university student, Doris Todd, positively identified Joseph Sanchell as the man who had robbed her in a dark Lincoln alley the preceding November.
Twenty minutes after the petitioner was bound over for trial on that offense, the preliminary hearing on the rape-robbery charges commenced. At this point, the court granted the defense motion to sequester the witnesses. All five testified. Renee again positively identified Sanchell as her assailant, based on his appearance. Patricia first testified only that she could not be sure, but "he fits what I saw." When the hearing was continued the next day, she said that based on her view of the petitioner's profile the day before, she believed he was the man. On cross-examination, however, she admitted she was not sure, but repeated "He fits the description of what I saw." Voice identification was never mentioned. The other three witnesses, Ann, Mary Ann and Sharon, were all unable to identify Sanchell at the preliminary hearing.
No further relevant identification proceedings were held for ten months, partly due to a written agreement negotiated after the preliminary hearing to the effect that if Sanchell submitted to and passed a polygraph examination he would not be prosecuted. Defense counsel assumed, after hearing that the petitioner had passed the test, that the charges would be dropped. In December, 1972, however, the prosecutor notified him that the State would proceed.3 Thereafter hearings were scheduled for January 24, 1973 on defense motions to contest that decision and to suppress the identification testimony.
The night before the suppression hearing, a year and a day after the crime, Patricia told the prosecutor that she would now positively identify Sanchell, despite the fact that she had not seen or heard him at any time since the preliminary hearing ten months earlier, when she had been unable to identify him.
At the suppression hearing, both Renee and Patricia testified, each positively identifying Sanchell. Both witnesses based their identifications in part on petitioner's voice. Since Sanchell had never been asked to speak for the witnesses and had never testified in their presence, Renee and Patricia were each asked when they had heard his voice so as to compare it to the voice heard at the time of the crimes a year earlier. Renee said that she had heard Sanchell talking just before the hearing that day. Patricia said that too. On cross-examination Patricia was asked how she could have promised the night before to base her positive identification on voice if she had not heard the voice until after she made the promise. Patricia then said for the first time that she had also heard Sanchell speak "to relatives or whatever" at the preliminary hearing the previous March. Since the other witnesses had not yet identified Sanchell, they did not testify. The motion to suppress was denied.
Sanchell's trial began February 5, 1973. Renee identified Sanchell by sight and voice. At trial she said for the first time that she, like Patricia, had heard Sanchell talk to "a friend" at the preliminary hearing and had recognized his voice then. Patricia also identified Sanchell on the basis of his profile and his voice. This time, however, she said that when she heard Sanchell at the preliminary hearing he had been talking to his attorney while she testified (rather than "to relatives" as she had testified at the suppression hearing in January). Mary Ann and Sharon, the next two witnesses, were unable to identify Sanchell at trial.
Ann, the fifth witness, identified Sanchell for the first time at trial. She based her identification entirely on his voice, which she said she had heard just before the suppression hearing. That concluded the identification testimony, the only evidence linking Sanchell to these crimes.
Due Process Considerations
The fundamental issue before us is whether admission of the identification testimony was proper under Fourteenth Amendment due process. We turn to a brief analysis of the law.
As a general rule, the identification testimony of one who has actually observed a crime is admissible and it is for the trier of fact to determine whether the testimony is worthy of credence. A different rule applies, however, where improper police suggestions may have influenced the witnesses to identify an innocent suspect. Limits on the admissibility of identification testimony which may have been influenced by governmental suggestion are mandated by the Sixth Amendment right to counsel clause and the due process clauses of the Fifth and Fourteenth Amendments. In United States v. Wade, 388 U.S. 218, 18 L. Ed. 2d 1149, 87 S. Ct. 1926 (1967), the Supreme Court examined the problem of pretrial lineups and one-man showups at length, noting that such confrontations are "peculiarly riddled with innumerable dangers . . . which might . . . derogate from a fair trial." 388 U.S. at 228. The Court further observed:
A major factor contributing to the high incidence of miscarriage of justice from mistaken identification has been the degree of suggestion inherent in the manner in which the prosecution presents the suspect to witnesses for pretrial identification. A commentator has observed that "the 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. Suggestion can be created intentionally or unintentionally in many subtle ways.
388 U.S. at 228-29.
Where improper suggestion by police or prosecution does lead to mistaken identification, the mistake is too often irreparable, for,
it is a matter of common experience that, once a witness has picked out the accused at the line-up, he is not likely to go back on his word later on, so that in practice the issue of identity may (in the absence of other relevant evidence) for all practical purposes be determined there and then, before the trial. 388 U.S. at 229.
The Court in Wade held that once an accused had been formally charged, identification confrontations arranged by the government were a critical stage of the prosecution at which the accused was entitled to assistance of counsel. Violation of that rule would require exclusion of all testimony about the pretrial identification itself. In-court identifications by the witness exposed to the improper confrontation would be admitted only if the government could show
388 U.S. at 240.
See also Gilbert v. California, 388 U.S. 263, 18 L. Ed. 2d 1178, 87 S. Ct. 1951 (1967).
In Stovall v. Denno, 388 U.S. 293, 18 L. Ed. 2d 1199, 87 S. Ct. 1967 (1967), announced the same day as Wade and Gilbert, the Court indicated that the Fourteenth Amendment's due process clause offered an additional ground for exclusion of identification testimony, a basis independent of any right to counsel claim. Due process would require suppression where the defendant could show, in "the totality of the circumstances," that the identification was procured by pretrial procedures " so unnecessarily suggestive and conducive to irreparable mistaken identification " as to deny due process. 388 U.S. at 301-02 (emphasis added).
Subsequently, the Court restated the due process test to be whether the 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, 19 L. Ed. 2d 1247, 88 S. Ct. 967 (1968). In Foster v. California, 394 U.S. 440, 22 L. Ed. 2d 402, 89 S. Ct. 1127 (1969), the Court reversed a robbery conviction, finding that a succession of lineups and showups in that case had
made it all but inevitable that [the witness] would identify petitioner whether or not he was in fact "the man." In effect, the police repeatedly said to the witness, " This is the man." . . . This procedure so undermined the reliability of the eyewitness identification as to violate due process. 394 U.S. at 443.
The sequence of events which led the Court to that result included an initial suggestive lineup and then, when the witness was unable to identify the defendant in the lineup, the police arranged a one-to-one confrontation between the defendant and the witness. Even then, the witness was unsure. A week later, the defendant was put into another lineup. He was the only person in this lineup who had also been in the first lineup. At the second lineup, the witness said he was convinced the defendant was the man.4
In the most recent Supreme Court case involving a due process challenge to identification testimony, however, the Court emphasized that testimony is not to be suppressed if the court finds that
Neil v. Biggers, 409 U.S. 188, 199-200, 34 L. Ed. 2d 401, 93 S. Ct. 375 (1972) (emphasis added).
This court in United States v. Dailey, 524 F.2d 911 (8th Cir. 1975), a due process case, indicated that once the defendant has proved that an unnecessarily suggestive confrontation has been used, testimony as to that confrontation must be excluded and the government then assumes the burden of proof by clear and convincing evidence that the witness' in-court identification was not tainted by the suggestive confrontation.5 Dailey, however, turned also on overwhelming proof of probable misidentification. 524 F.2d at 915.
We adhere to the principle enunciated in Dailey. However, because of the uncertainty as to the proper rule governing burden of proof in post- Stovall cases, see note 5, supra, we would note that the result here would be the same even if the petitioner had the burden of proof throughout. We find that the petitioner has shown that the showups were so suggestive as to create a substantial likelihood of irreparable misidentification by witnesses Patricia and Ann.
The One-on-One Confrontations.
We have previously warned that "showing only a single suspect to the witness is 'the most suggestive and, therefore the most objectionable method of pretrial identification.'" United States v. Dailey, 524 F.2d 911, 914 (8th Cir. 1975); United States v. Cook, 464 F.2d 251, 253 (8th Cir.), cert. denied, 409 U.S. 1011, 34 L. Ed. 2d 305, 93 S. Ct. 457 (1972). Accord, Foster v. California, 394 U.S. 440, 443, 22 L. Ed. 2d 402, 89 S. Ct. 1127 (1969); United States ex rel. Kirby v. Sturges, 510 F.2d 397, 403-08 (7th Cir.), cert. denied, 421 U.S. 1016, 44 L. Ed. 2d 685, 95 S. Ct. 2424 (1975).
The arraignment and preliminary hearing on the unrelated November, 1971 robbery charge were used as showups for this case. The suggestiveness inherent in any showup was surely aggravated here by choosing a forum that would tell the witnesses that Sanchell had been charged with another recent robbery and had been positively identified by the victim of that crime, another female student. See United States v. Luck, 447 F.2d 1333 (6th Cir. 1971). The government had no legitimate interest in displaying the petitioner to the witnesses in so suggestive a manner, for these witnesses had no necessary part to play in those hearings.
The Unnecessary Suggestiveness.
Second, the police did not need to resort to the one-man showups used. There was no emergency as in Stovall. After the photographic display, the police could have asked the persons whose pictures had been selected to participate in a fair lineup and a fair voice identification procedure.6 Voice identification was suggested by Renee but never followed up. Ironically, as earlier mentioned, Sanchell was placed in other lineups after his arrest but no lineup was arranged for the five young women to view.
The Resultant Irreparable Misidentification.
In addition, we find under Biggers, that the totality of circumstances created a "substantial likelihood of irreparable misidentification." We view the repeated presentations of Sanchell as a single suspect, particularly to Patricia and Ann, who had failed to identify the suspect initially, as the very abuse condemned by the Supreme Court in Foster v. California, 394 U.S. 440, 22 L. Ed. 2d 402, 89 S. Ct. 1127 (1969), for the reason that it makes identification virtually inevitable.
Patricia saw the intruder only for three seconds or less in a dark room, so she was presumably unable to give a detailed description before the showup. When she saw the petitioner's photograph only a day or two after the crime, she did not think it even resembled the intruder, and after the first showup, she told the interviewing officer that Sanchell was "probably not" the right man. Only after a second suggestive showup and after she learned that Renee had identified Sanchell, did Patricia tentatively identify Sanchell, and even then she was extremely uncertain. It was not until a year and a day after the crime that she decided for certain that Sanchell was the man, a decision made after last seeing or hearing him ten months earlier.
Ann may have had a better opportunity to see the intruder's face at the time of the offense than either Renee or Patricia, for she saw it at close range when the light was turned on. Yet she was unable to identify the defendant by appearance either by photograph or in person at any time, even at trial. Instead, she stated at trial that she had heard the defendant talking to someone before the suppression hearing began, a year and two days after the crime, and that she had suddenly recognized the voice. Even then, however, she did not so state to the prosecutor, but instead waited until trial, after she learned that both Renee and Patricia had identified the defendant. It is perhaps worthy of note that at the time of trial, Patricia and Ann were roommates, and they admitted discussing this case from time to time. Under the totality of the circumstances we find that the identification testimony of Patricia and Ann was so influenced by the various suggestive proceedings that it could not have any reliable independent basis in memory of the crime itself.
The Voice Identification.
The Nebraska Supreme Court originally found that the showups had impermissibly tainted Patricia's visual identification. The court observed:
We believe that the evidence . . . conclusively shows that the identification by Patricia was the product of the tainted showup on February 8, 1972, and the suggestive circumstances which preceded and followed it. It was not based upon her observations at the time of the crime. It must therefore be excluded. 216 N.W.2d at 514 (emphasis added).7
On rehearing two justices reversed their decision and the Court voted 4-2 to affirm Sanchell's conviction. In doing so the Court found that Patricia's in-court identification was not based on her visual observations at the showups but instead upon the subsequent independent voice recognition.8 Since the witnesses had not heard the petitioner speak at the showups the Court reasoned that those proceedings did not affect the ultimate voice identification. It further stated that since the witnesses other than Renee had not identified Sanchell immediately after the showups, these suggestive procedures did not enter into their ultimate identification. We must respectfully disagree. This analysis fails to give proper consideration to the legal factors governing the nexus between the earlier suggestive procedures, the manner in which the voice identification was made, and the ultimate identification itself.
The primary evil of the one-to-one confrontation is not that a witness might make a visual identification of a suspect. On the contrary, a fair lineup is conducted for that very purpose. The basic wrong of the showup is that authorities have singled out one individual and have at least implicitly pointed an accusatory finger at the suspect in the presence of possibly uncertain witnesses. It is this pervasive influence set in motion by the police which often becomes difficult to counteract. As the Supreme Court said in Wade, "the dangers for the suspect are particularly grave when the witness' opportunity for observation was insubstantial, and thus his susceptibility to suggestion the greatest." 388 U.S. at 229.
The District of Columbia Court of Appeals has said:
These dangers may result from such diverse influences as the witness's [sic] desire to cooperate with the police, from his knowledge that he is expected to identify some one he sees, . . . from uncertain recollections of a stranger's face distorted by a mental focus on particular features, from a generalized feeling of anger or vengeance, from suggestions subtly planted by the conduct or demeanor of a nearby policeman or other witness, from a calling of the defendant's name or an overheard description of his offense, and of course from a fortuitous line-up grouping which makes the defendant conspicuous or unique.
Mason v. United States, 134 U.S. App. D.C. 280, 414 F.2d 1176, 1180 (1969).
To obviate the substantial likelihood of irreparable misidentification after these forces have been loosed by earlier suggestive procedures, it is imperative to explore the circumstances under which the witness finally decided the accused was "the man." At a minimum the record should show, in such a case, that the subsequent identification was secured by fair procedures insulating it from the prior suggestion.9 It must show that the factors mentioned in Biggers, including "the level of certainty demonstrated by the witness at the [time of the] confrontation," indicate an independent and reliable identification. 409 U.S. at 199.
Without this minimal assurance, it becomes simply a matter of judicial rhetoric to say that the earlier denial of due process is no longer influential in the witness' ultimate identification. Due process, under these circumstances, and the test of substantial likelihood of irreparable misidentification are more than mere subjective tools of the judge viewing the facts. Such tests necessarily must lend themselves to objective evaluation and the assurance necessary to overcome the pervasive dangers of misidentification.
The fact that the witness ultimately bases identification on voice rather than appearance does not of itself indicate an independent source, especially where the voice identification is made when the witness knows that the speaker is the person already charged with the crime and already identified by other witnesses, as was the case here. The record reveals remarkably little information as to just what were the circumstances under which the witnesses heard the petitioner's voice so that they might compare it with the voice they heard at the time of the crimes. What we do know, however, is that when the witnesses heard Sanchell speak, his was the only voice of a black male that they heard. We do not know, for example, whether he spoke in a whisper nor how many words he said. Further, when they heard him speak they could also see him, and they were well aware by that time that he had been formally charged with the crimes. When Patricia made her voice identification she knew that Renee had already made a visual identification of Sanchell. When Ann made her voice identification, she knew that both Renee and Patricia had already identified Sanchell. Renee and Patricia's voice identifications were based in part, and Ann's entirely, on words heard a year and a day after the crime. Patricia told different versions as to when she heard Sanchell's voice. These circumstances make it clear that the voice identification procedures were not independent of the tainted visual showups.
Patricia and Ann's inability to identify the petitioner by his appearance and their lack of opportunity to hear other voices for purposes of comparison brings this case close to Palmer v. Peyton, 359 F.2d 199 (4th Cir. 1966), which the Supreme Court cited with approval in Stovall. In Palmer, a conviction was reversed on due process grounds where the victim had identified the defendant solely on the basis of his voice without opportunity to compare it with other voices. There the Fourth Circuit said:
Where the witness bases the identification on only part of the suspect's total personality, such as . . . voice alone, prior suggestions will have most fertile ground in which to grow to conviction. This is especially so when the identifier is presented with no alternative choices; there is then a strong predisposition to overcome doubts and to fasten guilt upon the lone suspect. 359 F.2d at 201 (emphasis added).
Although Renee was also exposed to the suggestive showups, we agree with the district court that there existed sufficient basis for her in-court identification of Sanchell. Renee's opportunity to observe the intruder at the time of the crime was much greater than that of the others and her ability to offer identifying evidence to the police prior to the showups lends credence to her subsequent identification. Prior to the showups, she had chosen Sanchell's picture along with three others and she gave a description sufficient to allow the police to make a composite picture of the suspect. She identified Sanchell the first time she saw him in person. Although the basis of her identification was challenged on cross-examination, under the circumstances we find that her visual identification of the petitioner was admissible, and its credibility was a question for the jury.10
The order of the district court is reversed and remanded with directions that the State of Nebraska grant petitioner a new trial within 60 days of the issuance of the court's mandate. In the event no retrial is held, the federal district court shall grant a writ of habeas corpus discharging the petitioner.
Reversed and remanded.
Reversed and remanded.
1. Sanchell had been arrested for robbery of Doris Todd, a university student, in November, 1971. These charges were later dropped.
2. Sanchell contends that since he had been formally charged with that unrelated robbery but was without counsel, it follows that using the February 8 arraignment as an identification confrontation violated his Sixth Amendment right to counsel with reference to the rape-robbery charges. This argument misapprehends the meaning of Kirby v. Illinois, 406 U.S. 682, 32 L. Ed. 2d 411, 92 S. Ct. 1877 (1972). We agree with the district court that that case controls, and that under Kirby only those identification confrontations which occur after the defendant has been formally charged with the offense for which the identification testimony is sought require presence of counsel.
3. Both the Nebraska Supreme Court and the federal district court found that the prosecutor did make a written offer to dismiss all the charges if defendant Sanchell would submit to and "pass" a polygraph test. 216 N.W.2d at 507. The agreement further provided that regardless of the results, the test would not be mentioned at trial in any way. The defendant did submit to the test, and the Lincoln police officer who administered it told defense counsel that Sanchell had "passed" since the polygraph indicated his denials of the rape and robbery were true. Months later, however, the police decided on the basis of other undisclosed information that the defendant had lied on the examination. The prosecutor then refused to drop the charges, and the trial court overruled the defendant's motion to dismiss on these facts without explanation. At trial, the prosecutor indirectly alluded to the polygraph examiner's "interview" with the petitioner, but did not disclose that the test indicated Sanchell was telling the truth. Both the Nebraska Supreme Court and the federal district court held that Sanchell had in fact passed the test within the meaning of the agreement. The Nebraska Supreme Court held, however, that such an agreement was unenforceable unless approved in advance by the trial court. 216 N.W.2d at 508. We in no way condone the conduct of the prosecutor in this case, but we agree with the district court that neither the refusal to drop the charges nor the reference at trial to the interview amounts to a denial of due process. The validity of the agreement is peculiarly a matter of state law, at least where, as here, there is no showing that the accused was prejudiced in preparing his defense. Cf. People v. Reagan, 395 Mich. 306, 235 N.W.2d 581 (1975).
4. In Foster, the Court apparently found the procedures so suggestive that it did not expressly consider other factors which might indicate whether or not the identification could nonetheless be reliable. The witness' two prior failures to identify the defendant were no doubt convincing evidence that the eventual identification was the product of police suggestion rather than memory of observations at the crime itself. See also, Palmer v. Peyton, 359 F.2d 199 (4th Cir. 1966).
5. In Dailey, this court relied on a right to counsel case, United States v. Valez, 467 F.2d 600 (8th Cir. 1972), without discussion of the conflict of authority as to the burden of proof in due process cases. There is a conflict of authority on whether the burden of proof is the same in due process challenges to identification testimony as it is in Wade right to counsel cases. Neil v. Biggers, 409 U.S. 188, 34 L. Ed. 2d 401, 93 S. Ct. 375 (1972), has been interpreted as placing a greater burden of proof on the defendant in due process cases, so that he must show not only that the procedures were defective because unduly suggestive, but also that the resulting identification evidence was unreliable under all the circumstances. See Pulaski, Neil v. Biggers : The Supreme Court Dismantles the Wade Trilogy's Due Process Protection, 26 Stan. L. Rev. 1097 (1974). In right to counsel cases, on the other hand, once the defendant shows that the procedures were defective, the burden shifts to the government to show that the evidence is nevertheless reliable. United States v. Wade, 388 U.S. 218, 240, 18 L. Ed. 2d 1149, 87 S. Ct. 1926 (1967). Biggers concerned a pre-Stovall v. Denno showup, and Mr. Justice Powell left open the question of whether a different, more stringent rule might apply to post-Stovall due process cases. Neil v. Biggers, supra 409 U.S. at 198-99. At least three circuits have adopted a different rule for post-Stovall due process cases, one which shifts the burden of proof to the government after the defendant shows that the identification witnesses were exposed to unnecessarily suggestive procedures. See Brathwaite v. Manson, 527 F.2d 363, 18 Cr.L.Rep. 2290 (2nd Cir. 1975); United States v. Sanders, 156 U.S. App. D.C. 210, 479 F.2d 1193, 1198 (1973); Kimbrough v. Cox, 444 F.2d 8 (4th Cir. 1971). The Seventh Circuit, on the other hand, has refused to adopt a different rule for post-Stovall cases in a well-reasoned opinion by Judge Stevens (now Mr. Justice Stevens). United States ex rel. Kirby v. Sturges, 510 F.2d 397 (7th Cir.), cert. denied, 421 U.S. 1016, 44 L. Ed. 2d 685, 95 S. Ct. 2424 (1975).
6. Since there is no evidence that the police ever requested Sanchell to participate voluntarily in a lineup before they resorted to the February 8 showup at the arraignment, we need not determine whether the police could have required petitioner to participate without his consent by arrest or otherwise at that time. We note, however, that Sanchell was at least technically in police custody on the unrelated robbery charge. Wade, Gilbert and Kirby presume that the police have the power to require persons in custody (and perhaps others as well) to appear in lineups regardless of whether there is probable cause to arrest the person on the charge for which the lineup is being held. This is necessarily so since a lineup must contain not only the police "suspect" if they have one, but also several "fillers." The problem of how to display a suspect to witnesses for identification before the police have probable cause to arrest him has been considered in United States v. King, 461 F.2d 53, 55 (8th Cir. 1972); United States v. Callahan, 439 F.2d 852, 865-66 (2nd Cir.), cert. denied, 404 U.S. 826, 30 L. Ed. 2d 54, 92 S. Ct. 56 (1971); United States v. Greene, 139 U.S. App. D.C. 9, 429 F.2d 193, 195-97 (1970); Bates v. United States, 132 U.S. App. D.C. 36, 405 F.2d 1104, 1106 (1968).
7. Two justices dissented on the ground that the credibility of the identification evidence was a question of fact for the jury.
8. The Court added that this was corroborated by her identification at trial of both profile and hand characteristics. These characteristics were, of course, observed at the time of the impermissible showups, earlier disapproved by the Court.
9. For example, Sanchell's voice along with the voices of other black males could have been portrayed to the witnesses sequestered in a room where the speakers were not visible. We suggest this simply as one manner in which a fair voice identification could have been made.
10. We are less sure of the propriety of permitting Renee to identify Sanchell's voice at trial. Certainly, she heard the intruder's voice enough at the time of the offense to make a subsequent voice identification possible. However, the circumstances as to when and how she heard the petitioner's voice in order to compare it with her assailant's are at best unclear. She did not mention voice identification until more than a year after the offense and the record is barren as to what she may have heard the petitioner say on the two occasions she said she heard him. These factors, we think should be explored before voice identification is permitted on retrial. This is more than a question of admissibility of evidence in a state trial. Identification evidence in a criminal trial which is unreliably obtained or which lacks proper foundation goes directly to the question of due process.