In the last chapter, we argued that there are two and only two plausible models of legal reasoning, the natural model and the rule model.

All judicial reasoning follows one or the other of these models or some combination of the two. Judges following the natural model reason engage in ordinary moral and empirical reasoning; judges following the rule model reason deductively from authoritative rules.

Our view is not the prevailing view. Respected analyses of legal rea­soning, as well as judges themselves, often claim that the distinctive reasoning process of the common law is reasoning by analogy to prior cases (see, in particular, Weinreb 2005; Levi 1948). Analogical reason­ing is praised as the means by which the common law has developed over time and adapted to changing social conditions. Closely related to analogical reasoning are a fortiori reasoning, various techniques for distinguishing precedent rules, and a reasons account of judicial rea­soning in which current courts are presumptively bound to conform to legal reasons derived from the facts and outcomes of prior cases (see Lamond 2005; Horty 2011; Lamond forthcoming). Another method thought by some to be distinctive of the common law, and to impose a form of constraint on current decision-making, is reasoning from legal principles immanent in prior decisions (see Dworkin 1986).

We assume throughout this chapter that there is no posited rule in effect that governs the facts of the current case. Instead, the preceden­tial value of a prior case or group of cases depends on factual similarity or dissimilarity between past and present cases, on ‘reasons' found in precedent cases, or on ‘legal principles' thought to emerge from a set of prior decisions. In other words, the current judge relies on inferences from the facts and outcomes of prior cases rather than the rule-making authority of prior judges. We will argue in this chapter courts cannot, logically, be doing what they claim to be doing when they find analo­gies in, or extract reasons or legal principles from, prior cases. Drawing analogies is not, in itself, a method of reasoning; precedent ‘reasons' do not determine current decisions; and legal principles turn out to be illusory.

At the same time, we are not entirely skeptical of the traditional meth­ods by which courts approach their decisional tasks. In Chapter 7, we take up the possibility that searching for analogies or reasons in past decisions, or for common principles that might link past and present cases, is a useful professional habit that can work indirectly to improve the quality of natural reasoning. Yet, to the extent that these practices play a helpful role in legal decision-making, they do so by broadening or sharpening judicial reasoning, not by constraining judicial reasoning.

We think the appeal of analogical decision-making or decision-making based on legal reasons or legal principles embedded in past cases is that these methods seem to offer a middle ground between unconstrained natural reasoning and deduction from binding rules. To the extent that judges decide cases by a process of unconstrained natural reasoning, there is no meaningful ‘common law' method that can counteract the errors or biases of individual judges. The rule of law seems imperiled. To the extent that judicial decision-making is governed by determinate and authoritative precedent rules, moral judgment plays no role in deciding legal cases. Our analysis, however, suggests that the compro­mise offered by analogical reasoning and reasoning from legal reasons or legal principles is illusory. If there is a middle ground, it must be found in reasoning to reflective equilibrium through the study of past cases, or in the power of judges to overrule faulty or obsolete judicial rules, not in spurious methods of reasoning.

6.1