We have argued that judicial reasoning is no different from the rea­soning that all actors use to make decisions.

Judges engage in uncon­strained natural reasoning, reaching empirical conclusions inductively and moral conclusions through the method of reflective equilibrium. They also reason deductively from authoritative rules.

We have also argued that the common law will be most effective if, in a range of cases in which moral settlement and coordinated action are important, current judges treat rules established in prior opinions as authoritative unless specifically overruled. This model of judicial decision-making, which we have called the rule model, assumes that judicial authority is not limited to interpretation of past decisions, but also includes the power to lay down binding rules. In Chapter 5, we addressed some of the theoretical questions that arise when judges act as rule-makers, including the scope of judicial rule-making authority, the conditions under which binding precedent rules are useful, and judicial power to overrule precedent rules. In this chapter, we consider some practical objections to the rule model, both as a prescription for judicial decision-making and as a description of judicial practice.

The most significant problem that arises in implementing the rule model of judicial decision-making is that judges are not ideal rule­makers.

Rules can be defective in several ways. They may be substantively defective, either because they aim at the wrong ends, or because the means they employ are ineffective. Problems of this kind may arise from poor design or from obsolescence. Rules may also suffer from formal defects. Because rules are general and determinate, they are necessarily over-inclusive: This is why the test of a good rule is not whether every outcome it prescribes is desirable but whether it will yield a desirable sum of results over time. At some point, however, the errors that result from application of over-inclusive rules may exceed both the errors that arise in natural reasoning and whatever benefits the rules provide through settlement and coordination. Rules may also perform poorly because they are overly complex; because the scope of their application is not sufficiently clear; or because their language generates interpretive controversies that are just as costly as the moral and practical controversies they are designed to settle.

Whether a given rule is defective for one or more of the reasons just described, or simply over-inclusive in the manner that any determi­nate rule is over-inclusive, is an empirical question. Nevertheless, judicial rules may be especially vulnerable to error. To the extent that judges have and exercise power to announce authoritative rules, they perform a dual function, as lawmakers and as adjudicators respons­ible for resolving disputes. As we will describe in the following sec­tion, the demands and distractions of adjudication pose special risks of error.

A second type of objection to the rule model of common law decision­making is that the rule model is at odds with judicial practice. As many have pointed out, judges and lawyers behave in ways that seem to con­tradict both the rule model of the common law and our more general argument that judicial reasoning comprises only unconstrained moral reasoning, empirical reasoning, and deduction from rules. The rule model assumes that judicial decision-making is constrained only by posited rules. Judges, however, often point to factual analogies between a case currently under consideration and prior decisions, and law­yers invoke factual analogies as components of legal arguments (see Weinreb 2005; Sunstein 1999; Simpson 1961). The rule model also assumes that judges have plenary power to announce authoritative rules. Yet, judges do not regularly announce general rules for future decision-making; the rules they do announce are often styled as ‘hold­ings'; and their holdings tend to be tailored quite narrowly to the dis­putes before them. Further, when precedent judges do announce rules in terms that go beyond the needs of a particular adjudication, later judges often assume that they are free either to disregard the rules as dicta or to distinguish current cases based on facts not mentioned in the rule (see Burton 1995; Llewellyn 1960a; Raz 1979; Lamond 2005; Summers 1997). The rule model of judicial decision-making, in con­trast, allows for overruling of prior judicial rules but does not recognize the practice of distinguishing precedent rules.

In the sections below, we review the impediments that judges face as rule-makers. We also suggest that various practices that tradi­tionally have played a part in common law decision-making but do not conform to the rule model may help to minimize or counteract the disadvantages that judges face as rule-makers. The conventional practices we consider do not ensure that judges will formulate good rules, but they serve, at least indirectly, to neutralize some predict­able sources of judicial error.

The common law strategies for controlling judicial error that we describe in the second part of this chapter are not deliberate, rational responses to the deficiencies of judges as rule-makers. They are custom­ary practices that counteract judicial error accidentally and indirectly. Because they depend on professional custom, they are also unreliable over time and resistant to any type of deliberate management. Yet, the accidental usefulness of these practices in controlling and improving judicial rules may be one reason why they have survived so well and continue to occupy a central place in legal education.

7.1