Tuesday, March 16, 2010

Spottswood on Hearings

Mark Spottswood (Northwestern) has posted on SSRN his article Hearings. Here's the abstract:

This article explores a constantly recurring procedural question: When is fact-finding improved by a live hearing, and when would it be better to rely on a written record? Unfortunately, when judges, lawyers, and rulemakers consider this issue, they are led astray by the widely shared - but false - assumption that a judge can best determine issues of credibility by viewing the demeanor of witnesses while they are testifying. In fact, a large body of scientific evidence indicates that judges are more likely to be deceived by lying or mistaken witnesses when observing their testimony in person than if the judges were to review a paper transcript of the testimony. Witness presence, in other words, may often harm, rather than improve, the accuracy of credibility assessments. The fact that legal actors value hearings for mistaken reasons does not mean that hearings have no value, but it does raise the concern that hearings will be employed when they are unneeded or even harmful, especially given the lack of available guidance on this question. In this article, I attempt to remedy this problem by providing a sound set of guiding principles concerning both the utility and the harms of live hearings.

Hearings will often, but not always, do more harm than good. In addition to the fact that demeanor cues generally impair, rather than aid, credibility judgments, there are a number of cognitive biases that may arise from having one’s first impressions of a witness be visual and auditory impressions. These include a persistent human tendency to trust or distrust witnesses based on their physical attractiveness, their social status, their race, or other features that may make them similar to, or different than, the fact-finder. On the flip side, hearings may help a judge make sense of confusing evidence. In addition, live hearings often feel fairer to participants than paper-based decisions, due in large part to the desire to have expressive input in decisions that affect us. And sometimes, a live hearing may be preferable for reasons of cost or practicality.



Recent Scholarship | Permalink

TrackBack URL for this entry:


Listed below are links to weblogs that reference Spottswood on Hearings:


The alternative to a live hearing is often not a transcript of testimony. Instead, it is frequently an affidavit, drafted by a lawyer for a party in the case, and not subject to cross-examination. There are a variety of reasons why affidavits are less reliable than live testimony. Distrust of trials by affidavit is why the hearsay rule and confrontation clause are central parts of Anglo-American juriprudence.

There are also frequently good reasons to prefer live hearings to deposition testimony. Parties and lawyers routinely behave better in front of judges than they do outside a judicial official's presence, in part, because "gray area" conduct is far more likely to be sanctioned. For this reason, testimony at live hearings may be more likely to be truthful than testimony via depositions.

The only place where the research on inaccurate credibility determinations is clearly applicable is to cast doubt the appellate procedure rule that calls for deference to factual findings regarding credibility on appeal.

Posted by: ohwilleke | Mar 16, 2010 3:09:18 PM

Thanks for commenting! I agree that affidavit evidence leaves much to be desired; in the paper, I discuss some examples where affidavits are the only form of evidence that is available or cost-effective, and argue that in such scenarios live hearings may be preferable on truth-seeking grounds.

I doubt, however, that "better behavior" at hearings is necessarily equivalent to "more truthful" behavior. Hearing testimony is more polished, in part, because it is rehearsed and because hostile questions can be better anticipated. These features actually make it less likely that a fact-finder will detect deception or errors. The idea that a witness will somehow be intimidated by the presence of a judge and therefore refuse to persist with a perjured account is an old one, but there is no empirical evidence supporting it, so far as I know, and certainly we have many available accounts of perjury committed in court. I suspect that the number of witnesses willing to lie during a deposition but unwilling to repeat that performance in court is quite small, so that any advantage gained through lie-deterrence is outweighed by the hearings defects as a lie-detection procedure.

Finally, I doubt that the demeanor-problem has broad implications for appellate review of credibility determinations. There are strong justifications for deference to trial courts on fact decisions that do not relate to credibility; trial courts tend to know the entirety of a record better than appellate courts and there is value in finality on these matters even if trial courts do not have any metaphysical advantage due to the presence of live witnesses in court. Currently, I think the best answer is that suggested by Judge Easterbrook in Mitondo, 523 F.3d 784: Defer to credibility findings if they are supported by a reasoned basis in evidence, but do not extend this deference to decisions based on inferences from demeanor evidence, due to its misleading nature.

Posted by: Mark Spottswood | Mar 18, 2010 6:00:11 AM

Post a comment