HOW RELIABLE IS EYEWITNESS TESTIMONY?: A Decision By New York State's Highest Court Reveals Unsettling Truths About Juries

In The People v. Lee, the New York Court of Appeals, the highest court in New York State, recently ruled that a trial judge has discretion to admit expert testimony on the reliability of eyewitness identifications.

The ruling, which follows the trend in other states, appears at first glance to be a straightforward application of well-settled principles of evidence law. However, its underlying logic exposes unsettling truths about the limitations of juries.

The conventional wisdom, particularly among non-lawyers, is that circumstantial evidence is generally less reliable than eyewitness testimony. People sometimes say that a case is "only circumstantial" to mean that the evidence is weak. A strong case, according to this view, includes the testimony of an eyewitness.

In fact, contrary to popular opinion, circumstantial evidence is often extremely reliable. Blood of the victim that makes a DNA match with blood found on the defendant's clothing, credit card records that place the defendant at the scene of the crime, and ballistics analysis that shows a bullet removed from the victim to have been fired from the defendant's gun are all forms of circumstantial evidence. Yet, in the absence of a credible allegation of police tampering, such evidence is usually highly reliable and informative.

At the same time, numerous psychological studies have shown that human beings are not very good at identifying people they saw only once for a relatively short period of time. The studies reveal error rates of as high as fifty percent — a frightening statistic given that many convictions may be based largely or solely on such testimony.

These studies show further that the ability to identify a stranger is diminished by stress (and what crime situation is not intensely stressful?), that cross-racial identifications are especially unreliable, and that contrary to what one might think, those witnesses who claim to be "certain" of their identifications are no better at it than everyone else, just more confident.

Granted, there is one mechanism by which we attempt to separate reliable from unreliable eyewitness testimony: cross-examination. When an eyewitness for the prosecution testifies that she recognizes the defendant as the perpetrator of a crime, the defense attorney can cross-examine her by inquiring about the lighting, the duration of the encounter, the stress she experienced, and all manner of other factors that might undermine the reliability of the identification.

However, jurors typically believe an eyewitness who sticks to his or her story despite cross-examination, in no small part because most jurors are not aware of the studies tending to cast doubt on such testimony. Accordingly, over the last two decades, defense attorneys have increasingly sought to introduce expert testimony about the limitations of eyewitness identifications.

Now the New York Court of Appeals has ruled not only that a trial judge can admit such expert testimony, but that the judge must consider doing so whenever it is offered — unless the judge determines (as the trial judge did in Lee itself) that in the particular circumstances of the case, the expert testimony would not be helpful to the jury.

Put another way, under Lee's ruling, a judge cannot simply decide that expert testimony on eyewitness reliability is categorically inadmissible. The judge must decide the issue case-by-case.

Lee was based on a longstanding principle of evidence law in New York and elsewhere, under which an expert can testify on matters outside the "day-to-day experience" of the ordinary citizens called to serve on juries, as long as such testimony would "aid a lay jury in reaching a verdict." (The federal test is similar: Expert testimony can be introduced if it "will assist the trier of fact to understand the evidence or to determine a fact in issue.")

Expert testimony about eyewitness reliability certainly seems to fit the bill. Indeed, the very reason why defense attorneys want to introduce such evidence is that jurors' day-to-day experience will lead them to give too much weight to eyewitness testimony.

Expert testimony on eyewitness reliability, then, seems not just to be outside jurors' day-to-day experience, but to helpfully controvert misimpressions jurors might gather from that experience.

Nevertheless, Lee was a harder case than the unanimous Court of Appeals opinion indicates. That is because the very concept of trial by jury is based on the assumption that the day-to-day experience of jurors equips them to weigh testimony and reach accurate conclusions. Yet Lee acknowledges that jurors are ill-equipped to perform this most basic function.

Moreover, once we start asking whether jurors can do their job, we may discover that expert testimony is needed to correct other misimpressions as well.

Consider lying. In both civil and criminal trials, appeals courts strongly defer to a jury's decision to believe one witness rather than another. This practice is based on the assumption that jurors who see and listen to the witnesses firsthand are in a better position to assess their credibility than a reviewing judge looking at a cold transcript.

Yet here too, as with eyewitness testimony, the empirical evidence suggests otherwise. Studies show that jurors tend to equate signs of nervousness with untruthfulness. Many practiced liars, however, are cool as a cucumber, while first-time witnesses are often nervous about testifying, quite apart from whether they are telling the truth or not.

Moreover, there are cultural biases. Americans often want a witness to "look them in the eye" as a sign of honesty, while in other cultures it is a sign of respect to keep one's gaze focused downward.

The logic of the Lee decision suggests that litigants should be permitted to offer expert testimony about the signs of honesty and dissembling. Under Lee's analysis, this too appears to be something outside of the day-to-day experience of lay jurors.

Indeed, carried to its conclusion, this logic may suggest that a litigant should even be permitted to call an expert to testify that a particular witness exhibited signs of lying visible to the expert. In fact, this was permitted in a celebrated case, the perjury prosecution of accused communist agent Alger Hiss in 1949-50.

Hiss's first perjury trial ended in a hung jury. At his retrial, Hiss's lawyer called Dr. Carl Binger, who testified that Hiss's chief accuser, Whittaker Chambers, was a pathological liar.

Binger based this conclusion on his assessment of Chambers's testimony in the two cases as well as some of Chambers's writing (including his translation of Bambi from German to English!) By all accounts, the tactic backfired when Binger wilted under withering cross-examination by prosecutor Thomas Murphy.

Since the Hiss case, defendants have rarely sought to introduce expert testimony on the veracity of witnesses — individually or as a group. Prosecutors, however, routinely seek to introduce the expert opinion of an examining doctor that a child victim of sexual abuse is telling the truth.

Moreover, the fact that Binger, a supposed expert in the evaluation of credibility, could be so easily discredited, should make us wonder about the ability of untrained jurors to make such determinations.

Accordingly, the Hiss case — like the Lee decision — could be invoked in support of the admissibility of expert testimony cautioning jurors not to put too much stock in their ability to judge credibility.

The New York Court of Appeals was right to hold that expert testimony on eyewitness credibility should not be per se inadmissible. The solution reached by the Court of Appeals is to leave the matter to the discretion of the trial judge. That answer is technically correct as a matter of evidence law: trial judges have broad discretion over what expert testimony to admit.

But by punting the question to the trial courts of New York, the Court of Appeals did not address the truly difficult issue: If jurors commonly place too much weight on eyewitness identifications (and make other cognitive errors), why shouldn't litigants be permitted to offer such expert testimony in every single case in which the other side offers eyewitnesses?

The best reason to exclude such evidence in any given case will be efficiency. If eyewitness testimony is a relatively small part of the government's case, it may waste everyone's time to hear extensive expert testimony on the reliability of eyewitnesses.

On the other hand, if the government thinks an eyewitness is sufficiently important to call to the stand, it would also seem sufficiently important to inform the jury about eyewitness reliability.

The extent to which such expert testimony is allowed will now depend largely on the attitude of the particular judge before whom a case is tried. Some judges will routinely allow it; others will usually forbid it; and virtually all such decisions will be affirmed on appeal. That inconsistency itself is somewhat troubling.

Ultimately, however, this uneasy state of affairs may be the best we can do. The New York Court of Appeals was understandably reluctant to mandate a side-trial on eyewitness reliability in virtually every case tried in the New York courts. Nor, given the stark facts about eyewitness unreliability, could the Court of Appeals approve a blanket rule of inadmissibility. Such a ruling would have led to injustice in many cases that are based largely on eyewitness testimony.

Trial by jury is a safeguard against oppression. As the Lee case shows, however, it comes with a cost. The ordinary citizens we trust to protect us against potentially overzealous government are, by virtue of their very ordinariness, often inexpert at the tasks they must perform.