(concurring).
The issue principally litigated on this appeal is whether the rule set forth in People v Riley (70 NY2d 523), striking down as unconstitutional station house showup identifications except in exigent circumstances, applies where a witness to the crime has identified someone as the responsible person on the street shortly after the crime under circumstances making that identification constitutionally admissible, and is thereafter asked to view the arrested person again at the station house in an effort to make sure that the witness had not erred in the original identification.
However, a study of the hearing minutes leaves me with a disturbing sense of uncertainty as to whether the witness had actually seen the features of those whom he had preliminarily identified. In fact, the Assistant District Attorney at the hearing expressed the view that the witness had not seen the features of 1 of the 2 persons whom he identified without, however, specifying which one the assistant believed that the witness had not actually seen. Accordingly, I am in agreement with the court that the facts set forth in the hearing minutes do not provide a basis for deciding this case on the basis of the issue primarily argued, or for distinguishing it from Riley (supra). Because of the importance of the issue, and because the court’s memorandum opinion, although ultimately resting on the uncertainty surrounding the street identification, may be construed as setting forth a more inclusive principle, I think it appropriate to discuss the issue that was presented by the parties on this appeal.
On analysis, it is a variation of the question addressed by *504the Court of Appeals in People v Morales (37 NY2d 262) with regard to confirmatory showup identifications by undercover police officers. The Morales analysis opened with the following sentence (supra, at 271): "The vice to be avoided by the salutary safeguards required to avoid suggested or mistaken identifications generally apply to initial identifications.” The opinion went on at length to detail the differences between identifications by civilian witnesses, usually the victims of crimes occurring quickly under exciting circumstances, and identifications by trained professional police officers who, as in the case of an undercover officer, observed the crime under circumstances in which it was part of the officer’s professional duty to observe closely the features of the person with whom he was involved.
I agree that this latter part of the analysis was decisive in People v Morales (supra). But this does not in any way diminish the importance to the issue argued here of the absolutely accurate statement in that analysis by the Court of Appeals that the vice to be avoided by the safeguards (i.e., lineup identifications) "generally apply to initial identifications." (Supra, at 271.)
Notwithstanding the unsatisfactory nature of the hearing testimony as to whether the identifying witness had actually observed the features of those whom he identified on the street, the facts of this case present a helpful introduction to the issue. At issue here are station house viewings by the victim of a robbery of two persons arrested on the basis of his "identification” of them on the street while he was accompanying the police who had responded to his complaint. The complaining witness reaffirmed his identification of 1 of the 2 arrested persons, the defendant. As to the other arrested person, the witness stated he was no longer sure that he was the second participant in the robbery.
Almost certainly as a result of this station house viewing, the charges against the second arrested person were dismissed. It is consistent with the limited information available that the station house viewing, and the expression by the witness of his doubts as to the accuracy of his street identification, may well have precluded the possibility that the second arrested person would have been tried, and conceivably even convicted, on the basis of the street identification made under circumstances that may well have been persuasive to a jury of that person’s guilt.
Nor is there any reason to suppose that the clearly positive *505result of the station house viewings here is somehow unique to the case before us. Human experience suggests that from time to time a crime victim who has identified someone on the street in the immediate aftermath of a crime may, after a period of time and on a second viewing, be less certain of the accuracy of his original identification. Where the confirmatory identification procedure used here is followed, and has such a result, it may, as in this case, result in the dismissal of charges against a person arrested on the basis of a constitutionally admissible street identification, and in other cases it may result in the immediate release of such persons from custody. It is not easy to believe that a procedure motivated by a desire to protect an arrested person against the possibility of a misidentification, and which from time to time will have results protective of such persons, is violative of rights guaranteed by the Constitution. Nor do I believe that the opinion of the Court of Appeals in People v Riley (supra) is reasonably interpreted as supporting such an anomalous result.
In his appellate argument that a station house viewing of an already identified person violates that person’s constitutional rights under the rule set forth in Riley (supra) unless a lineup is utilized, the defendant fails to consider three characteristics of that procedure which the Court of Appeals had no occasion to address in Riley, and which fundamentally transform the nature of the relevant constitutional inquiry.
First, the concern underlying Riley (supra) that the suggestiveness of a station house showup viewing may unacceptably increase the possibility of a misidentification has little, if any, relevance to a situation in which the witness has already identified the arrested person under constitutionally acceptable circumstances, and is thereafter asked to look again at someone whom he has already identified, and whom he knows he has already identified, to determine whether he continues to be sure of the correctness of that identification.
Second, the reasons for preferring lineup identification procedures are virtually nonexistent where a member of the lineup is someone whom the viewing witness has already identified. Whatever may be the correctness of the original identification, it surely will be rare that the viewing witness will not recognize in the lineup the person whom he had identified only a short period before. Where the witness recognizes a person in the lineup as someone he previously identified, the presence of others in the lineup is not likely to add any additional trustworthiness to the reliability of the second *506identification. Where a lineup includes someone identified shortly before by the viewing witness, it is realistically as suggestive as a showup, if suggestiveness is relevant to the intended purpose of the second viewing, which I do not believe to be the case. In fact, the principal practical effect of using a lineup for a second station house viewing of an already identified person will be to enable a trial prosecutor to argue deceptively that the jury should give additional weight to the station house identification because the defendant was selected out of a lineup.
Finally, it is unrealistic to suppose that busy police officers will think it necessary to allocate police time and resources, and to further impose on the time and convenience of a crime victim, to arrange a lineup with respect to someone already identified under constitutionally permissible circumstances in order to confirm that the earlier identification was accurate. In effect, the rule urged by defendant discourages the use of a confirmatory procedure with a real potential to exculpate a person unjustly accused of a crime on the basis of a prior misidentification. It discourages the use of such a procedure in the form in which it is most likely to be used by police officers, and does so without any realistic expectation that in the usual situation the police will conduct any confirmatory viewing at all if they are required to use lineups in such circumstances. Coupled with the reality that lineups have little, if any, advantage over showups in the situation with which we are concerned, the effective result of such a rule would be to discourage the use of a practical procedure whose principal beneficiaries will be persons arrested on the basis of a street identification as to which the complaining witness may be less certain on a second viewing. I do not believe that the Constitution, or the opinion in Riley (supra), requires so dubious a result.
It does not follow from the conclusion that the Constitution would not be violated by such a procedure that testimony with regard to such a station house confirmatory viewing, should necessarily be admitted as evidence. In the kind of situation with which we are concerned, I think that a Trial Judge might reasonably exercise his discretion to exclude testimony with regard to the station house identification, not because the defendant’s constitutional rights were violated, but because the probative value of the confirmatory identification of an already identified person is insufficient to justify a repetition of what is in essence a single identification.
I have considered whether it would be appropriate for an *507appellate court to set forth as a general principle, on the basis of this balancing test, that testimony with regard to a station house confirmatory viewing should be excluded. I am inclined to think that such a general rule would be unwise because factual variations in individual cases, as well as unexpected trial developments, make the issue more appropriately one for the exercise of discretion by a trial court. But the statement of such a general principle based on familiar balancing considerations appropriate to admissibility would seem to me far preferable to an untenable constitutional rule that would discourage police use of a confirmatory procedure whose principal beneficiaries are likely to be erroneously identified defendants.