Remedy and Remaining Safeguards

Remedy and Remaining Safeguards

The weighing in the previous section ends in a ruling, and the ruling decides what the jury will hear. When the court finds that an unnecessarily suggestive procedure created a very substantial risk of misidentification, the identification is excluded, and the exclusion reaches the courtroom identification as well as the earlier display.¹ When the court finds that reliability outweighs the suggestion, the identification is admitted, and the defense still has several ways to test the witness in front of the jury, none of which requires proof that the police did anything wrong.² The division reflects a division of labor. The due process screen addresses the narrow danger of suggestion created by the police, and the ordinary adversarial process addresses the broader danger that an honest witness is simply mistaken. The first subsection below covers the hearing and the scope of suppression, and the second covers the safeguards that remain after the screen has done its work.

Suppression of Both Identifications

The defense raises the due process claim by a motion in limine, which is a request made before the witness testifies that asks the judge to forbid any testimony about the identification, and the trial court holds a hearing on it.³ At the hearing the court takes proof on each step of the inquiry: whether the procedure was suggestive, whether it was unnecessarily suggestive and arranged by law enforcement, and, if both are true, how the five reliability factors come out.⁴ The admissibility question should be decided at a hearing held outside the jury's presence, although the Constitution does not require the jury to be excluded. Hearing it first serves a practical function. The testimony that shows how the witness came to choose the defendant is the same testimony the court may rule inadmissible, and a jury that has already sat through it cannot be made to forget it. The burden of proof at the hearing rests on the defense, because the defendant must prove an alleged due process violation. That allocation differs from the right-to-counsel claim, where the government must prove that counsel was present, that counsel was waived, or that an independent source supports the in-court identification.

Once the court has weighed the five factors, the hearing produces one of two results. If the court concludes that the unnecessarily suggestive procedure created a very substantial risk of misidentification, due process is violated and neither the out-of-court identification nor an in-court identification is admissible.⁵ If the court concludes that the reliability of the identification outweighs the unnecessary suggestiveness, the identification is admissible through testimony about either the out-of-court identification or an in-court identification.⁶ The remedy reaches the in-court identification because of what the display does to the witness's memory. A witness who was steered to the defendant at the display carries that choice to the stand, and the pointing finger at trial then looks to the jury like fresh recognition of the culprit when it may be recognition of the person the witness was shown earlier. Excluding only the first identification would let the prosecution recover the same tainted result through the second.

The admitting result is illustrated by Manson v. Brathwaite, where an undercover officer who had bought heroin from the defendant picked him out of a single photograph two days later and also identified him at trial, and the Court rejected the due process claim.⁷ The weighing in that case came out in favor of reliability, so the identification evidence remained available to the jury, which is the outcome the totality approach was adopted to permit. The Court preferred that approach to a rule of automatic exclusion because a rule that bars reliable evidence can let the guilty go free, and reversal of a conviction on that ground is a drastic sanction.⁸

The one recognized limit on excluding the in-court identification is the independent source. A witness may identify the defendant in court despite an unconstitutional pretrial identification if the in-court identification rests on a basis in the witness's own observation of the crime that does not depend on the tainted display. The court tests that basis with the same five considerations used to measure reliability: the opportunity to observe the criminal at the time of the crime, the witness's degree of attention, the accuracy of the prior description, the level of certainty at the confrontation, and the time between the crime and the confrontation. The overlap explains why the hearing's findings carry forward. A court that has just found the witness's memory too weak to offset the display has little left to point to as an independent source, while a witness who watched the crime closely and described it accurately before any display has a source in the crime itself.

The usual slip is forgetting the in-court identification. A student who writes that the showup is suppressed and stops has left the witness free to point at the defendant in front of the jury, so the answer needs both identifications and, for the courtroom one, the independent source question tested by the five factors. A second slip is to assume that exclusion follows from suggestiveness alone. Suppression is required only if the unnecessary suggestiveness creates a very substantial risk of misidentification, so a suggestive display that the factors outweigh leaves both identifications admissible.⁹

Safeguards Beyond Suppression

Admission of an identification after the weighing does not close the defense's case against it. The due process screen answers one danger, a misidentification produced by suggestion that law enforcement arranged, and it excludes evidence only when the resulting risk is very substantial.¹¹ Eyewitnesses go wrong for many reasons that have nothing to do with the police, such as a brief look, a high-stress event, or the passage of months, so an identification can pass the due process test and still be shaky. The totality approach accepts that result because it keeps identifications that have features of reliability before the jury, and other devices exist to test them.¹² Those devices operate whether or not law enforcement did anything improper, and they include counsel at post-indictment lineups, vigorous cross-examination, protective rules of evidence such as Rule 403, and jury instructions on the fallibility of eyewitness identification and on proof beyond a reasonable doubt.¹³

Each device answers a different weakness in eyewitness evidence, as follows:

• Counsel at a post-indictment lineup can watch the display as it happens and object to suggestive features while they can still be corrected.
• Cross-examination lets the defense bring out before the jury the same facts the judge weighed at the hearing, including the length of the view, the lighting, the witness's attention, the first description, and any early uncertainty.
• Evidence rules such as Rule 403 give the judge a general power to exclude evidence whose risks outweigh its value, which supplements the due process screen.
• Jury instructions tell the jurors that eyewitness identifications can be mistaken and that guilt must be proved beyond a reasonable doubt, which keeps a confident witness from standing in for proof.

Expert testimony adds what cross-examination cannot supply. A cross-examiner can show that the witness had a brief look, but a lawyer's question cannot teach jurors that stress, a long delay, or a difference in race between witness and culprit can reduce the accuracy of memory, and ordinary experience may not have taught them. Perry recognized the admissibility of expert eyewitness testimony on problems such as cross-racial identification, identification after long delays or under stress, and psychological phenomena, and the defense may use it to prove suggestiveness or unreliability.¹⁴

These tools mostly go to weight and do not replace the in limine motion. The motion asks whether police-created suggestion made the identification unreliable enough to keep from the jury, while the safeguards help the jury decide how much an admitted identification is worth, including when the police arranged nothing at all.¹⁵ The common mistake is to treat the in limine motion as the only way to attack an eyewitness, so that an identification that survives the weighing is regarded as settled. It is still open to cross-examination, expert testimony, evidentiary objection, and instruction, and an exam answer about an admitted identification should say so.

Suppression of Both Identifications

A due process challenge to an identification is resolved before the witness testifies. The defense files a motion in limine, which is a request made ahead of the testimony asking the judge to bar any mention of the identification, and the trial court holds a hearing at which it takes proof on each step of the inquiry: whether the display was suggestive, whether it was unnecessarily suggestive and arranged by law enforcement, and, if both answers are yes, how the five reliability factors come out.¹⁷ The timing has a practical reason. The testimony showing how the witness came to choose the defendant is the same testimony the court may rule inadmissible, and jurors who have already heard it cannot be told to forget it, so the admissibility question should be decided outside the jury's presence, although the Constitution does not require the jury to be excluded. The defendant must prove the alleged due process violation, while the government carries the burden of showing an independent source for an in-court identification.

The hearing ends in one of two rulings, and both concern evidence rather than the charge. If the court concludes that the unnecessarily suggestive procedure created a very substantial risk of misidentification, due process has been violated and neither the out-of-court identification nor an in-court identification is admissible.¹⁸ If the court concludes instead that the reliability of the identification outweighs the suggestiveness, the identification comes in through testimony about either the earlier display or an in-court identification.¹⁹ The remedy is therefore exclusion of identification testimony, and the prosecution may go forward on whatever other evidence it has.

The remedy reaches the courtroom because of what an unreliable display does to memory. A witness who was steered to the defendant at the display carries that choice to the stand, and the pointing finger at trial then looks to the jury like fresh recognition of the culprit when it may be recognition of the person the witness was shown earlier. Excluding only the first identification would let the prosecution recover the same tainted result through the second. The recovery is not always blocked, and the limit is the independent source rule: a witness may identify the defendant in court despite an unconstitutional pretrial identification if the courtroom identification rests on the witness's own observation of the crime rather than on the display. Courts decide that by weighing the witness's opportunity to view the criminal at the time of the crime, the witness's degree of attention, the accuracy of the prior description, the level of certainty demonstrated at the confrontation, and the time elapsed between the crime and the confrontation.²⁰ Those are the same five considerations that measured reliability at the hearing, so the findings carry forward. A court that found the witness's memory too thin to offset the display has little left to call an independent source, while a witness who watched closely and described the culprit accurately before any display has a source in the crime itself.²¹

The right-to-counsel cases supplied the model for this inquiry and also mark where due process differs. Under United States v. Wade, an in-court identification that followed an uncounseled lineup had to be excluded unless the prosecution established by clear and convincing evidence that it rested on observations of the suspect other than the lineup.²² The factors for that determination were somewhat broader than the five used today: the prior opportunity to observe the crime, any discrepancy between the pre-lineup description and the defendant's actual appearance, any earlier identification of another person or from a photograph, any earlier failure to identify the defendant, and the time that passed.²³ Gilbert v. California went further when the prosecution itself elicited testimony about the tainted lineup identification, and barred that testimony outright, with no chance to show an independent source, so that police would respect the right to counsel.²⁴ Due process analysis has no counterpart to that automatic bar, because a suggestive display excludes nothing until the court has weighed the reliability of the identification against it.²⁵ The independent source question also resembles the attenuation inquiry in Brown v. Illinois, which asks whether a later confession has been purged of an earlier illegality by looking at elapsed time, intervening events, and the flagrancy of the misconduct, but the identification version substitutes factors that measure the witness's memory of the crime.²⁶

The cases show both rulings. In Foster v. California the only eyewitness to the robbery of a Western Union office viewed a three-man lineup in which Foster, close to six feet tall, stood beside two men about five feet five or six inches and alone wore a jacket like the robber's, and the witness could not positively identify him.²⁷ The police then arranged a one-to-one confrontation in an office, and the witness was still unsure; a week or ten days later a second lineup of five men included Foster as the only repeat participant, and this time the witness was convinced.²⁸ At trial the witness testified about the lineup identifications and repeated the identification in the courtroom, and the Court held that the procedures made the resulting identifications virtually inevitable and denied due process, reversing the conviction and leaving the harmless-error question for remand.²⁹ The lineups predated the counsel requirement of Wade, so due process was the only ground available, which shows the claim stands apart from the right to counsel.³⁰ A courtroom identification offered after that sequence would restate a choice the police had manufactured, and that is why the remedy has to reach the in-court identification unless the prosecution can point to a source in the witness's own view of the robbery.

Manson v. Brathwaite shows the admitting result. An undercover officer who bought heroin from the defendant watched the seller for several minutes, described him to another officer soon after the purchase, and two days later picked the defendant out of a single photograph; the officer also identified him at trial, and the Court rejected the due process claim.³¹ Justice Marshall, dissenting, regarded a single photograph as inherently suggestive and would have required a proper lineup.³² The majority let both identifications go to the jury because the close view, the prompt description, and the short delay left a memory that the display had not created.³³ Both cases involved displays that pointed at one man, and they came out differently because of the memory behind the choice. Foster's witness stayed uncertain through a lineup and a face-to-face confrontation and became sure only after the police put the suspect before him a second time, so the certainty came from the display. Manson's witness had studied the seller closely and described him before seeing any photograph, so the choice could be traced to the crime.

The usual slip is suppressing only the lineup. An answer that excludes the pretrial display and stops leaves the witness free to point at the defendant before the jury, so the answer must also address the in-court identification and ask whether it has an independent source under the five factors. A second slip imports the automatic exclusion of the counsel cases into the due process claim. Suggestiveness alone does not require suppression, so a display that the factors outweigh leaves both identifications admissible, and the defense is left to the trial safeguards covered next.³⁴

Safeguards Beyond Suppression

Admissibility under the due process test settles only whether the jury will hear an identification, and the test reaches a narrow band of cases: those in which law enforcement arranged an unnecessarily suggestive procedure whose suggestion the five reliability factors fail to offset. The narrowness follows from the rule's purpose. In Perry v. New Hampshire the Court held that due process does not require a trial judge to assess the reliability of a suggestive identification the police did not arrange, because the reliability inquiry exists to deter improper lineups, showups, and photo arrays, and there is nothing to deter when no officer arranged anything.³⁶ The witness in Perry, when asked for a more specific description of the man she had seen, pointed through her kitchen window at the defendant standing in the parking lot beside a police officer. The trial court denied suppression because the police had not manufactured that confrontation.³⁷ A defendant in that position gets no preliminary reliability ruling, so the tools described next carry the whole burden, and they are available whether or not law enforcement did anything improper.³⁸

The Court has identified the devices that fill the gap: counsel at post-indictment lineups, vigorous cross-examination, protective rules of evidence such as Federal Rule of Evidence 403, and jury instructions on both the fallibility of eyewitness identification and the requirement that guilt be proved beyond a reasonable doubt.³⁹ Each answers a different weakness in eyewitness evidence:

• Counsel at a post-indictment lineup watches the display as it unfolds and can object to suggestive features before the witness has chosen anyone.
• Cross-examination puts before the jury the facts the judge weighed at the hearing, such as the length of the view, the lighting, and the witness's first description.
• Rule 403 lets the judge exclude evidence whose probative value is substantially outweighed by its prejudicial impact or its potential to mislead the jury, and that standard does not ask who arranged the identification.⁴⁰
• Jury instructions tell jurors that identifications can be mistaken and that guilt must be proved beyond a reasonable doubt, so a confident witness cannot stand in for proof.

Cross-examination does its best work on the witness's first description, because that statement records the memory before later events can reshape it. Kyles v. Whitley shows the technique, although the case itself concerned statements the prosecution failed to disclose. One eyewitness had told the police that the gunman was about five feet four or five inches tall and of medium build, and the Court reasoned that cross-examination on that statement would have left him struggling to explain how he later identified a thin six-footer. A second witness's account had grown from not having seen the shooting to a detailed description of it, which would have invited a withering cross-examination.⁴¹ The Court added that the way an eyewitness's description evolves over time can be fatal to the reliability of the identification.⁴² Cross-examination of this kind informs the jury without deciding the question for it. The Kyles dissent thought a jury might well credit an unhesitating identification by a witness who saw the killer from under ten feet away over a rough estimate of height and weight.⁴³ That disagreement concerns weight, which belongs to the jury. In Neil v. Biggers the Court noted that the testimony had been addressed to the jury, which apparently found the identification reliable.⁴⁴ And Manson v. Brathwaite refused to adopt automatic exclusion because a per se rule keeps reliable and relevant evidence from the jury.⁴⁵

Cross-examination has a limit, and expert testimony and instructions address it. A lawyer's questions can show that the view was brief, but they cannot teach jurors that stress, a long delay, or a difference in race between witness and culprit can degrade memory. The research collected in Justice Sotomayor's dissent in Perry found that jurors routinely overestimate the accuracy of eyewitness identifications and give the greatest weight to the witness's confidence, even though confidence is a poor gauge of accuracy.⁴⁶ Suggestion makes this worse because it can inflate a witness's confidence, which complicates the jury's credibility judgment and blunts the very cross-examination that would expose the weakness.⁴⁷ Perry recognized the admissibility of expert eyewitness testimony on problems such as cross-racial identification, identification after long delays or under stress, and psychological phenomena, and the defense may use that testimony to prove suggestiveness or unreliability.⁴⁸ The fallibility instruction then gives the jury the same warning in the judge's voice at the close of the case.

The common mistake is to treat the due process motion as the only attack on an eyewitness, so that an identification that survives it is regarded as settled. It remains open to every device above, and suggestion that does not come from the police leaves the defense with only those devices. Because the day-to-day decision whether and when to object rests with defense counsel, the challenge must be raised at trial, and a claim left unraised risks being barred from later collateral review absent cause and prejudice.⁴⁹ Everything outside the screen goes to weight, while the screen itself keeps its two steps: the court first asks whether law enforcement arranged an unnecessarily suggestive procedure, and only if so weighs the five reliability factors against the corrupting effect of the suggestion to decide whether there is a very substantial likelihood of misidentification.⁵⁰