Over at
The Situationist (where I have blogged in the past), the staff has posted an essay considering the intersection of two topics: the tension between preferences and principle, as
explicated by Knowles and Ditto, and statutory construction. Knowles and Ditto argue that people's preferences override their principles, that is to say, individuals' preferred outcome trumps whatever principles they claim to adhere to. The Situationist post draws a parallel between this and Llewellyn's
famous critique of canons of construction. Llewellyn argued that each given canon had a "counter-canon", which was equally legitimate and could thus be used to reach the opposite result. Rather than using the principle to reach a conclusion, Llewellyn suggested that the judge decided on a result and chose the appropriate cannon. The Situationist post makes connections between Llewellyn's work and the more recent Knowles and Ditto piece. That post concludes:
Does this insight lead to the inescapable conclusion that the use of canons in statutory interpretation is inherently invalid? Not exactly. First, abolishing the use of canons would do nothing to solve the “problem” of results-based reasoning—judges could simply rely on other tools, such as legislative history, to provide post hoc rationalizations of their preference-based decisions. Legislative history is subject to the same critiques as the canons of construction; namely, that there is often legislative history support both sides of a debate, allowing judges to, as Justice Scalia put it,“look over the heads of the crowd and pick out your friends.” This argument is somewhat defeatist in that it seems to concede that the human decision-making process is inherently flawed. However, decision-making on the basis of intuition is not inherently invalid, and may actually “reflect adaptive insights accumulated over the course of human evolution.”
Further, our system requires judges to do more than simply declare “yes” or “no.” Federal Rule of Civil Procedure 52(a) requires judges to explicitly state their findings of fact and conclusions of law; this rule has three important ramifications to our topic. First, the very act of committing their reasoning to paper may reveal a judge’s logical inconsistency or suboptimal use of a canon to him or herself. Second, writing an opinion allows dissenting judges to file their own opinions in response; these dissents may reveal stronger arguments to judges in the majority. Finally, our appellate system allows incorrect legal conclusions made at the lower levels of the court system to be corrected by the high courts. This doesn’t entirely overcome the defeatist position, as having many biased decision-makers arguing about differing biased decisions solves very little in a search for an objective truth, but that stance presupposes the very existence of an “objective truth” – a topic far beyond the scope of this post. For now, we can take solace in the fact that even if Knowles and Ditto are correct in that the individual decision-making process isn’t quite as objective as we’d like, any judge’s individual decision must survive several rounds of checks and balances before becoming law.
I want to pick on this conclusion. First, the "several rounds of checks and balances" really provide that much "solace"? Dissenting opinions only come about (if at all) at the appellate level. Trial-court decisions are typically decided by a single judge (rare exceptions notwithstanding) and are not subject to dissent. Most decisions are not appealed, and most appeals are affirmed. The Supreme Court, as it reminds us many times, is not a court of errors and declines to hear the vast majority of cases presented to it. Even at the Supreme Court and the Courts of Appeals, the most common dispositions are unanimous decisions. (I have not dug up the citations for this post, but I make these points in
my recent article on stare decisis.) I wonder if this does not give lower court judges a tremendous first-mover advantage, particularly in state courts where the volume of cases is much higher (thus leading to the risk of less robust processes of error-correction, such that they are). More broadly, this raises the question of
procedural checks on
substantive errors. That is, we believe that imposing certain procedural hurdles (written decisions, appellate review, published dissents) will work out substantive mistakes. But it is not at all clear or intuitive that this should be the way to work out problems in the law
The response might come, "Well what is the alternative? Substantive checks on substantive errors? That simply encourages the second decision-maker substitute his (biased and subjective) judgment for that of the first decision-maker." But this position is, in the terminology of the post, "defeatist". If we are all looking out at the crowd and picking friends, then we may as well call the whole enterprise off. I'll only agree with my friends, you with yours, and we cannot reconcile our positions.
That counter-argument actually has quite a bit of purchase. The political system is constructed essentially on that premise, deferring almost entirely substantive questions, and insisting instead that meticulous procedural arrangements are met (two senators per state, two houses of Congress, judicial review, presidential vetoes, the Electoral College, etc.). It is as if the founders were aware of Aristotle's three forms of government (he called them six, but Machiavelli later explained, there are really only three; each can be either good or bad) and created each of the three branches in the image of one of those three forms (rule of the one, executive; of the few, judicial; the many, legislative).
Substantive ideals are also advanced by procedural means. Consider our criminal justice system's pro-defendant tilt. You can imagine a conversation that runs something like this: "We do not want individuals wrongfully convicted of crimes. Therefore, we will presume innocence and not require testimony from defendants. Those defendants who
are convicted will have the ability to appeal, though as a rule, the government may not appeal an acquittal. Even after meaningful judicial review, some cases may slip through the cracks. Therefore, we will allow post-conviction collateral attack on convictions by means of the writ of habeas corpus or an adequate substitute for the writ. Finally, we will even allow private actions against individual government employees means of 42 U.S.C. § 1983."
All of these elements exist in our system, but one must ask if these rights are robust or hollow. Although defendants are not required to testify, police officers routinely seek waivers of
Miranda rights. Criminal appeals are often summarily affirmed. The right to habeas corpus is limited at best; petitioners -- who are often proceeding
pro se -- must comply with labyrinthine procedural requirements or risk immediate and often-irreversible default. Although § 1983 actions are a theoretical remedy, they cannot be used to challenge the fact of confinement; any monetary awards are limited by the Prison Litigation Reform Act; and in any case government agents enjoy absolute or qualified immunity in many cases.
One could draw two conclusions from this landscape. First, we might despair that court decisions have eroded our commitment to criminal defendants' rights (this is similar to the point about procedural judicial activism I make
in this piece). But second, we might take some comfort in the fact that various procedural protections exist. Though they may be less robust than they could be, they are certainly more robust than the alternative (no protections at all!). If we as a society will be psychologically predisposed against criminal defendants, then a pro-defendant "systemic bias" may serve as an adequate thumb on the scale to correct that bias.
Wechsler's and Llewellyn's heirs will carry the debate on for years hence. Going forward, we might ask that these debates be informed by empirical data about how we humans actually think (and how
we think we think). I am reminded of Robert Maynard Hutchins's
lecture in the 1930s where he noted that the law had devoted an entire subject, the rules of evidence, to how jurors will react to different types of evidence (expert testimony, lay testimony, hearsay, etc.) -- yet no one bothered to consult psychologists and other social scientists about whether the law's assumptions had any basis in reality or not, and the psychologists had not given any study to the behaviors of juries! In a very real sense we are still only a few steps down the journey that Hutchins sought to set us on almost a century ago.
Perhaps the legal system will never perfectly embody "neutral principles" or "realism" (whatever those terms mean in a vacuum). But if the system aggressively seeks, identifies, and counteracts cognitive bias, then we may be closer to a framework that more accurately reflects society's "true", behind-the-veil preferences about how justice is best served.
