Tuesday, May 12, 2009
The Goals of Adversarialism
I want to start to consider how to incorporate evidence-based systematic reviews into the legal system. There are two ways to approach the problem. The first is to think about how to use systematic reviews in a supplmentary way: the parties can still retain their own experts, but the judge and/or the jury can also see what the complete picture is as well. The second is more ambitious: can we replace the partisan experts altogether with the systematic review?
In order to figure out how to incorporate non-adversarial devices into an adversarial framework, it is essential to first identify the goals of an adversarial system. Uncovering the truth is certainly one goal, but it is not the only one. Here, I'll briefly touch on the goals of adversarialism and some of the reasons why its proponents think it advances them well. If there are other goals that I am overlooking, I'd definitely like to hear about them.
Goal No. 1: Discovery of the truth. As the US Supreme Court said in Tehan v United States, 383 US 406, 416 (1996), "[t]he basic purpose of a trial is the determination of truth." And defenders of adversarialism, such as Stephan Landsman and Sheila Jasanoff, have put forth several arguments for why adversarialism advances truth-finding.
1. Parties have the strongest incentive to gather evidence and find the best expert. Since the party has the most amount at stake, it has a much greater incentive to turn up compelling evidence. But it also has the greatest incentive to turn up obscuring evidence or to bend the evidence it or the other party has.
2. Party control minimizes the risk of confirmation bias. An independent investigator may initially believe the outcome could either be x or y. But once he starts to think it is x--perhaps simply because by chance the first several articles he read favored x--he will (on the margin) approach the issue with that bias, putting more weight on pro-x evidence and less on pro-y evidence. Party control ensures that the fact-finder hears from powerful advocates for both sides. Of course, if one side has a much stronger case, confirmation bias isn't necessarily a bad thing.
3. Cross-examination reveals important biases. These biases are not just the standard conflict-of-interest or gun-for-hire concerns. There are deeper worries about methodological, ideological, or even cultural biases as well. Cross-examination exposes these biases to the light of day. Do juries, though, have the epistemic competence to know how to properly account for these biases?
4. Prevention of "undue deference." Juries may put "too much" weight on the claims of an independent expert because of this independence. At one level, this is a bizarre concern: if indepedent experts traditionally produce more reliable results, such reliance is not undue, but due. Howie Erichson, however, raises a good concern: a biased (consciously or not) judge could take advantage of this deference by appointing an independent expert who shares his policy preferences with the hope of getting the jury to reach the same conclusion.
Goal No. 2: Conflict Resolution. Party control ensures that the trial proceeds at the pace desired by (at least one of) the parties. But the bigger role of conflict resolution is a normative one. As one federal judge told Cecil and Willging, “We’re conditioned to respect the adversary process. If a lawyer fails to explain the basis for a case, that’s his problem." Cases are often viewed as local issues, pertaining to the parties before the judge. But the fact is, we heap a lot of regulatory and social insurance responsibilities on the courts, so at least in cases involving complex evidence, a lawyer's failure to explain his case could be a problem for all of us.
Goal No. 3: Individual control of a government institution. The sources of this goal are varied. Part of it is our bedrock individualism and distrust of government. By putting the parties in control, the government's job is narrowly constrained. Even as we seem to move away from agreeing with Reagan's nine terrifying words joke, we remain a country leery of the government. Another justification is that the parties themselves are paying, so they should be in control. And unless we are willing to create a large judicial bureaucracy to process complex science cases--a highly unlikely possibility that I will not even bother to consider here--the parties will likely be paying in the future as well.
Goal No. 4: Preventing arbitrary or biased outcomes. A passive and neutral fact-finder does not have the ability to impose its own policy preferences on the outcome. Perhaps tied a bit, though by no means completely, to concerns about confirmation bias, the thought here is that by cabining the deciders' actions to a narrow issue--here, determining who presented the more compelling case--adversarialism limits the deciders' ability to behave poorly. It should be noted, though, that others have suggested the epistemic limitations of the fact-finders actually cause arbitrariness, since untrained juries can do little more than guess as to who has the better claim.
Because of these values, judges are generally unwilling to appoint independent experts. In my next posts, I will argue that they should. In particular, systematic reviews not only advance goals 1 and 4, but do so in ways that respond to the very concerns judges and others have with independent experts. Systematic reviews are in tension with goals 2 and 3, but their application can be reserved for those cases where these interests are relatively weak.
Posted by John Pfaff on May 12, 2009 at 11:53 AM | Permalink
TrackBack URL for this entry:
Listed below are links to weblogs that reference The Goals of Adversarialism:
The comments to this entry are closed.