The Five Reliability Factors

The Five Reliability Factors

The list divides according to what each factor reveals about the witness's memory. The first two, opportunity and attention, show how well the memory was formed. A witness who had a long, close, well-lit look and was concentrating on the culprit took in far more than one who glanced at a stranger. The last two, certainty and elapsed time, bear on how well the memory survived until the confrontation. Memory fades with delay, and confidence at the display may come from a durable memory or from the display itself. The prior description sits between the two groups, because it records what the memory held before any display could reshape it. That makes it the benchmark against which the later identification can be checked.

These factors are inputs to a judgment, and a court does not treat them as elements. The court asks whether, under the totality of the circumstances, the factors taken together outweigh the corrupting effect of the display.³ Because no single factor is a prerequisite, strength on some can offset weakness on others, and the weight each receives varies with the record. Students lose points when they treat the five factors as elements that must each be satisfied. The identification is excluded only when the suggestion, weighed against whatever reliability the factors show, creates a very substantial likelihood of misidentification.⁴

The same five considerations do a second job. When a court asks whether an in-court identification has an independent source sufficient to purge the taint of an unconstitutional pretrial identification, it looks to these factors as well.⁵ The remedy section develops that use. The sections that follow take the factors one at a time, beginning with the opportunity to view.

Opportunity to View

The first reliability factor asks how good a look the witness had at the criminal while the crime was taking place.⁷ Courts answer by examining the lighting, the amount of time the witness had to look, the quality of the witness's eyesight, and how directly the witness faced the person.⁸ Distance belongs in the same inquiry, as the cases below show, since a face two feet away yields far more detail than one across a street. The factor is tied to the moment of the offense, so it asks nothing about the later photograph or lineup. That later display is the suggestive event, and the first factor measures what the witness's memory held before the display could touch it.⁹

The reason this factor matters is that a memory can be no better than the perception it records. A long, close, well-lit, direct look gives the witness a detailed memory of the culprit's face, and a memory that rich does not depend on anything the police later show, so a suggestive procedure has little to add. A brief glimpse in poor light leaves gaps, and gaps are what suggestion fills. A witness who barely saw the culprit has little original memory to hold against the face in front of the witness, and is more likely to accept that face simply because it is on display. A strong view therefore works as an anchor, and a weak one leaves the identification adrift.

In Neil v. Biggers the victim of a rape spent up to half an hour with her assailant, which gave the Court its clearest example of a strong opportunity to view.¹⁰ She was with him under adequate artificial light in her house and later under a full moon outdoors, and she faced him directly and intimately at least twice, once in the house and once in the woods.¹¹ Duration, light, and angle of view all pointed the same way, and the Court held that the lower court's contrary conclusions on these critical facts were clearly erroneous.¹² A seven-month gap before the confrontation, which the Court called a seriously negative factor in most cases, did not overcome that record.¹³ The same case shows why courts allow a somewhat suggestive procedure to stand when the original view was this strong: the identification still rests on what the witness saw during the crime.

The view in Manson v. Brathwaite was far shorter than in Biggers and still counted heavily. The undercover officer stood within two feet of the seller for two to three minutes, the door opened twice and the seller stood at it each time, and the officer looked directly at him.¹⁴ It was near sunset, but the sun had not yet set, and natural light came into the hallway through a window and from inside the apartment.¹⁵ Half an hour in Biggers and a few minutes in Manson both favored reliability, because close distance, adequate light, and a face-to-face angle made each minute of looking yield more. That is why the factor has no minimum number of minutes. The facts also show how contestable the factor can be, because the door opened only 12 to 18 inches and the two-to-three-minute figure came from the officer's own testimony.¹⁶ Two Justices dissented in Manson and would have reversed the conviction on that record, a reminder that the strength of a view is a judgment about evidence and not a measurement.¹⁷

A federal court of appeals applied the same inputs in United States v. Stevens, where two military police officers were attacked in a bus shelter. The attack lasted at least ten minutes, the assailant stayed close and made no effort to hide his face, and he faced the officers directly several times.¹⁸ One officer testified that the light in the shelter was enough to read a book or the face of a watch, and the court described the opportunity to view as unusually lengthy and vivid.¹⁹ The three cases together give the strong pole: a long or close view, in adequate light, with the culprit's face turned toward the witness.

A weak opportunity to view does not by itself require exclusion, because the court weighs it with the other four factors against the corrupting effect of the suggestive display.²⁰ A frequent error is treating the length of the view as the whole factor. Time is one input among several, and a short look at close range in good light can outweigh a long look from a distance in the dark. Where the view was weak, the prosecution needs strength from the other factors to answer the suggestion, and the next sections take up those factors in turn, beginning with the witness's degree of attention.

Degree of Attention

The second reliability factor asks how closely the witness was paying attention to the culprit while the crime was happening.²² It is easy to blur this factor with the opportunity to view, but the two measure different things. Opportunity describes what the witness was in a position to see, which depends on time, light, distance, and angle. Attention describes how much of that available view the witness actually took in, and a person can stand in plain sight of a face while registering almost nothing about it because the mind was on something else. Courts therefore sort witnesses by a practical question: was this person trained to observe, a casual passing observer, or, as often happens, a victim who is likely to be especially alert?²³

The reason the factor bears on reliability is that people form detailed memories of what they attend to and thin ones of what they barely notice. A witness who concentrated on the culprit's face carries a rich original memory, and a memory that rich leaves the later display little room to reshape it. A witness who gave the stranger a passing glance carries very little, and a photograph or lineup shown afterward can supply the missing details and be accepted as if the witness had remembered them. Attention therefore works as a partner to the first factor, because a long view is worth more when the witness spent it looking at the right thing.

At one end of the range stands the trained observer on duty. In Manson v. Brathwaite the undercover officer who bought heroin was treated as no casual or passing observer, since he was a trained police officer on duty in specialized and dangerous work.²⁴ The Court reasoned that a specially trained, assigned, and experienced officer could be expected to pay scrupulous attention to detail, because he knew he would later have to find and arrest the seller and knew his claimed observations would face close examination at trial.²⁵ Two separate links carry that conclusion. Training gives the officer the skill to notice, and the assignment gives him a reason to memorize. The Court also acknowledged the counterpoint that an officer in that work is expected to produce results, and it still found that the training and the expectation of later scrutiny outweighed that pressure.²⁶

The crime itself can supply attention that training does not. In United States v. Stevens, a federal court of appeals found that two military police officers attacked in a bus shelter were trained in observation, and it added that neither was a casual or passing observer because a terrifying assault had riveted their attention to the assailant.²⁷ Both officers testified that they watched closely throughout, and one said that during the attack she was already thinking ahead to identifying him at the police station, which is the same later-accountability reasoning that supported the officer in Manson.²⁸ The court cited Manson for the point that specially trained officers can be expected to attend to detail.²⁹ Stevens differs from Manson in where the attention came from: the Manson officer was carrying out a deliberate operation, while the Stevens officers were drawn into the attack against their will.

Ordinary victims occupy the middle ground, and their position rests on the same mechanism as the Stevens officers. In Neil v. Biggers the Court described the rape victim as no casual observer, because of the nature of the crime she suffered.³⁰ A person under direct threat has every reason to study the person making the threat, so victims are likely to be especially alert, though the word is likely and not always.³¹ Research on eyewitness memory observes that events happen quickly, often under great stress, and that memories are faulty, which is why a court treats a victim's alertness as a tendency to be weighed and not as a certainty.³²

The trained-observer pole carries a limit that students often miss. Police work does not make an identification reliable by itself, and the exam error is treating an officer's badge as proof of adequate attention. Attention is one input among five, and the court weighs it against the corrupting effect of the display together with the view, the description, the certainty, and the delay.³³ The officer in Manson prevailed because the training coincided with a two-foot view in natural light, a prompt description, firm certainty, and a two-day gap, and not because of the uniform alone. The dissent by Justice Marshall and Justice Brennan targeted this very structure: it regarded a single photograph as inherently suggestive and concluded that a factor-based reliability test does not adequately guard against suggestion.³⁴ High attention answers the question of how well the memory was formed, and the court must still ask whether the display overwrote it. The next factor, the accuracy of the witness's prior description, provides the record of what the memory held before any display.

Accuracy of Prior Description

The third reliability factor looks at the description the witness gave of the criminal before any suggestive display, and asks how accurate and how complete it was.³⁶ The first two factors concern how the memory was formed, since opportunity to view and degree of attention describe what the witness saw and how closely. The description is the earliest record of what that memory actually held, made before the police showed the witness anyone, and it can be set beside the person the witness later picked. Courts place descriptions along a spectrum that runs from poor, through not detailed but reasonably accurate, to complete and accurate.³⁷ A related question is whether the accused looked different at the time of the crime than at the time of the identification, because a change in appearance can explain a gap between the description and the person picked.³⁸

The description matters because it supplies a yardstick that the suggestive display could not have shaped. A witness who reports particular features to the police before seeing any photograph or suspect has fixed what the memory contained. If the person the witness later picks has those features, the pick probably came from memory, since a display shown afterward cannot have supplied details the witness had already reported. If the description was thin, there is little to compare, and the witness's sense of recognition may be a product of the face on display. If the description conflicts with the person picked, the display more likely drove the choice. A detailed description is therefore evidence of where the identification came from, which is why a thorough one makes a court more willing to accept an identification that followed a somewhat suggestive procedure.³⁹

In Neil v. Biggers the victim of a rape gave the police what the federal district court characterized as “only a very general description”: the assailant was fat and flabby, with smooth skin, bushy hair, and a youthful voice.⁴⁰ The Supreme Court read the hearing record more fully. She had also said that her assailant was between 16 and 18 years old, between five feet ten inches and six feet tall, 180 to 200 pounds, and dark brown in complexion, and a police officer testifying from his notes substantially corroborated her.⁴¹ Taking in everything she reported (approximate age, height, weight, complexion, skin texture, build, and voice), the Court called the description “more than ordinarily thorough.”⁴² The description also did work during the seven months of searching that followed. She viewed suspects at her home and at the station in lineups and showups, was shown between 30 and 40 photographs, told the police that one photograph showed features similar to her assailant's, and identified none of them.⁴³ That history suggests the description was operating as a standard against which she tested each candidate.

The strong pole in Manson v. Brathwaite rests on promptness more than on volume. The undercover officer described the seller to a fellow officer within minutes of the sale, giving the seller's race, height, build, the color and style of his hair, a high cheekbone, and the clothing he wore.⁴⁴ The fellow officer, who suspected he knew the seller's identity, reacted positively at once, and two days later the undercover officer picked out the lone photograph that the fellow officer produced.⁴⁵ The Court added that no one claimed the defendant lacked the physical characteristics described.⁴⁶ Accuracy, in other words, is tested against the actual features of the accused. The two cases show different ways a description can support reliability. In Biggers the strength lay in detail, offered by a witness who had spent up to half an hour with her attacker. In Manson the strength lay in timing and in the quick match to a person another officer already had in mind.

Because the factor is a judgment about context, the same description can look strong to one reader and thin to another. The federal district court in Biggers had treated the description as general, and the Supreme Court disagreed after considering the whole record.⁴⁷ In Manson, two Justices would have reversed the conviction, reasoning that a single photograph is inherently suggestive and that the reliability test does not adequately protect against suggestion.⁴⁸ A list of race, height, build, and hair fits many people, so a court looks at how specific the description was, how soon it was given, and whether it fit the accused, and does not simply count features.

The same factor also exposes descriptions that change. In Kyles v. Whitley the Court explained how damaging an eyewitness's shifting account can be, citing Biggers and Manson for the point that reliability depends in part on the accuracy of the prior description.⁴⁹ One witness had told the police that the assailant was about five feet four or five, 140 to 150 pounds, and of medium build, yet testified at trial that Kyles, who was six feet tall and thin, was the shooter.⁵⁰ Another said at the scene that he had not seen the shooting or the assailant outside the car, and at trial described the struggle and the gun in detail.⁵¹ Kyles arose from the prosecution's failure to disclose those earlier statements, so it is not a suppression case, but it shows what the factor measures: whether what the witness says now fits what the witness said first.⁵²

A common error is treating a description given after the display as a prior description. A statement made after the witness has seen the photograph may echo the display and does not show what the memory held beforehand, so only descriptions that precede the suggestive procedure carry the weight this factor assigns. Even a properly timed description that is poor or inconsistent does not by itself require exclusion, because the court weighs it with the other four factors against the corrupting effect of the suggestive display.⁵³ The next factor, the witness's level of certainty at the confrontation, is closely tied to this one, because certainty seems in large part to be a function of how extensive the description was.⁵⁴

Level of Certainty

The fourth reliability factor asks how certain the witness was at the confrontation with the suspect, whether that confrontation was a lineup, a showup, or a photograph display.⁵⁶ Certainty can run from nonexistent to virtual certainty, and it seems in large part to be a function of how extensive the witness's description was, which is why the previous factor leads directly into this one.⁵⁷ A witness who gave the police many details beforehand has something concrete to hold each candidate against, and so can say with some confidence whether this person fits. The factor counts because the entire inquiry asks whether the choice came from memory or from the display, and a firm recognition expressed at the moment of the confrontation is evidence that memory was doing the work. Certainty is also the factor most easily produced by the procedure itself, since it is reported by the same witness whose memory the display may have reshaped. Two checks therefore decide how much a statement of certainty is worth: when the witness expressed it, and whether it fits what the witness had done before.

The second check, consistency with the witness's history, is illustrated by Neil v. Biggers. The rape victim told the police she had “no doubt” that the man at the showup was her assailant.⁵⁸ The Court gave that statement weight because of the record behind it: over the previous seven months she had viewed suspects in lineups and showups and had been shown between 30 and 40 photographs, and she had identified none of them.⁵⁹ The Court concluded that her record for reliability was a good one, since she had previously resisted whatever suggestiveness a showup carries.⁶⁰ That history is what turns her confidence into evidence. A witness who has repeatedly declined to choose when shown candidates has demonstrated that she does not say yes to whichever person the police place in front of her, so a decisive answer at the one confrontation where she did choose cannot be explained away as a reaction to being shown a suspect. The same record offset the seven-month gap, which the next factor treats as a negative in most cases, because a witness who had resisted every earlier display was less likely to have been swayed by this one.⁶¹ History can also cut the other way, since what matters is whether the witness previously viewed other physical or photographic displays and either picked no one or picked someone else.⁶²

Foster v. California shows the opposite pattern. The only eyewitness to a Western Union robbery viewed a lineup in which the defendant was far taller than the other two men and wore a jacket like the robber's, and afterward the witness only thought the defendant was the man but was not sure.⁶³ A one-to-one meeting in an office left the witness still unsure.⁶⁴ A week or ten days later the police arranged a second lineup in which the defendant was the only person who had appeared in the first, and this time the witness was “convinced.”⁶⁵ Foster was decided before the five factors were catalogued, so it is not a factor-by-factor holding, but its sequence is the plainest picture of contaminated certainty. What changed between doubt and conviction was the number of times the police had put the same man in front of the witness, and the Court said the police had in effect told the witness again and again that this was the man.⁶⁶ The two cases differ in what the history shows. In Biggers the record shows a witness who withheld a choice through many displays and then chose with confidence, while in Foster the record shows confidence that rose only as the same suspect was displayed again, which is the suggestion itself at work and cannot support reliability.

Certainty also comes in degrees, and two more cases show what firm certainty looks like when no contamination story explains it. In Manson v. Brathwaite the undercover officer was asked whether the photograph showed the man who sold him heroin, and he answered that there was no question whatsoever and repeated that assurance.⁶⁷ The Court found little pressure on the officer to go along with the display, because the detective had left the photograph at the officer's office, the officer viewed it alone two days after the purchase, and he could do so with care and reflection.⁶⁸ In United States v. Stevens two military police officers picked the defendant's photograph from a wanted board about an hour after an attack, and the court of appeals treated both as highly certain.⁶⁹ Officer McCormack's identification was unqualified, and Officer Smith said the first thing that hit her was that it was the man, because of his eyes, though she admitted that the assailant's face looked slightly heavier in real life than in the photograph.⁷⁰ Her small caveat about weight did not turn recognition into hesitation, which shows that the factor measures how firm the recognition was, with absolute certainty at the top of a range.

The error that costs the most points on this factor is treating certainty displayed at trial as certainty at the confrontation. By the time of trial the witness may have seen the defendant at a lineup, at a hearing, and across the courtroom, and each look can raise confidence without adding anything to what the witness remembers of the crime. A dissenting opinion by Justice Sotomayor collects research showing that jurors give the greatest weight to an eyewitness's confidence even though confidence is a poor gauge of accuracy.⁷¹ A court applying the factor therefore looks to what the witness said when first confronted with the suspect, and it discounts certainty that surfaced only after repeated exposure. Certainty remains one input among five, weighed with the view, the attention, the description, and the delay against the corrupting effect of the display, so strong certainty does not by itself carry the identification and weak certainty does not by itself defeat it.⁷² The fifth factor, the time that passed between the crime and the confrontation, supplies the last input to that weighing.

Time Elapsed

The fifth reliability factor asks how much time passed between the crime and the confrontation with the suspect.⁷⁴ A short interval favors reliability, and a long one is usually a negative, though it is a variable that courts tend to weigh less heavily than the other four factors.⁷⁵ The factor's wording fixes the two endpoints. The clock starts when the witness saw the crime and stops at the confrontation, which is the moment a suggestive display could first act on whatever the witness still remembered. The reason the interval matters is easiest to see by splitting the concern into two parts.

The first part is decay. Memory of a stranger's face fades, and the available scientific evidence suggests that most of the loss occurs in the first forty-eight hours after a crime.⁷⁶ The second part is contamination. Eyewitness recollections are highly susceptible to distortion by post-event information and social cues, so every added week gives conversation, news coverage, and earlier looks at the suspect more opportunity to fill in details that memory no longer supplies.⁷⁷ An identification made within hours or days leaves little room for either process, which makes it more likely that the choice reflects the crime and less likely that it reflects the display. A gap of weeks or months raises both concerns at once, and that is why a long interval is usually a negative.

Manson v. Brathwaite supplies the favorable pole. The undercover officer bought heroin on May 5, described the seller to a fellow officer within minutes, and viewed a single photograph two days later.⁷⁸ The Court contrasted that gap with the passage of weeks or months and counted the two days among the indicators of the officer's ability to identify accurately.⁷⁹ A shorter interval appears in United States v. Stevens, where two military police officers recognized the defendant on a wanted board about an hour after an attack, and the court of appeals said that so near an identification added considerable weight to their emphatic certainty and that the assailant's image could reasonably be assumed to remain fresh.⁸⁰ The two cases show how this factor works with the others. In Manson the favorable interval was reinforced by a description given within minutes, and in Stevens it made the officers' certainty more believable, because certainty voiced while memory is fresh is harder to attribute to repeated exposure. Freshness still answers only the reliability question: two Justices in Manson would have reversed anyway, reasoning that a single photograph is inherently suggestive and that the reliability test does not adequately guard against suggestion.⁸¹

The unfavorable pole is Neil v. Biggers, where seven months separated the rape from the showup.⁸² The Court called that lapse “a seriously negative factor in most cases.”⁸³ It did not stop there, because the undisputed testimony was that the victim had made no identification at any earlier showup, lineup, or photographic display, so her record for reliability was a good one and she had previously resisted whatever suggestiveness a showup carries.⁸⁴ That reasoning follows from the contamination concern described above. A long gap is troubling because later displays and information can overwrite memory, and a witness who passed through many displays without choosing anyone has shown that the memory did not bend toward whichever candidate appeared. Weighing all the factors, including a view lasting up to half an hour, a description more thorough than usual, and “no doubt” at the showup, the Court found no substantial likelihood of misidentification.⁸⁵ Manson and Biggers therefore teach different things about the same factor: in Manson the interval helped, and in Biggers it hurt and was outweighed.

Courts tend to give this factor less weight than the other four.⁸⁶ The first four factors examine this witness's own encounter with this culprit, while elapsed time is a general indicator of how much memory may have eroded. A witness who saw the culprit clearly, attended closely, and described the person precisely has shown that the memory started out strong enough to tolerate some delay. The reverse does not hold, since a short interval cannot supply a good view that the witness never had, and a recognition an hour after a glimpse in poor light remains weak evidence. That asymmetry previews the weighing in the next section: a factor can be negative and still be outweighed, and the court asks how all five compare with the corrupting effect of the suggestive display.⁸⁷

A frequent error is measuring the interval to the trial date. The factor, as stated, compares the crime with the confrontation, so a court reads the clock at the moment the display could have acted on memory, and the delay before trial is not what it counts.⁸⁸ A related mistake is to treat a long interval as an automatic bar. The same decision that called seven months seriously negative admitted the identification, so a long lapse earns a negative mark and a place in the weighing and does not by itself decide admissibility.⁸⁹