by Neil Duxbury. Cambridge and New York: Cambridge University Press, 2008. 208pp. Cloth $100.00/£60.00. ISBN: 9780521885799. Paper $45.00/£24.99. ISBN: 9780521713368. eBook format. $36.00. ISBN: 9780511389344.

Reviewed by Matthew J. Franck, Department of Political Science, Radford University; Visiting Fellow, James Madison Program, Princeton University. Email: mfranck [at]


Why do judges in the English common law tradition (including those in former colonies such as the United States) have a doctrine of precedent? What does it mean to say that prior rulings of a court have authority to bind judges, either on the same court or inferior ones? Are there good reasons to follow precedent? If so, what reasons will suffice for not doing so?

These are among the questions addressed by Neil Duxbury, professor of law at the London School of Economics, in this brief but densely argued and challenging book. If you finish reading THE NATURE AND AUTHORITY OF PRECEDENT with more questions and uncertainties than you previously had about its subject matter, Duxbury has been successful. For, as he argues, “no one theory can offer a plausible comprehensive or systematic explanation of why precedents constrain” (pp.ix-x). His objective seems to be to induce the reader to turn the doctrine of precedent this way and that, viewing it from multiple angles and taking none of its features as either philosophically or historically given. But his project is neither cynically corrosive nor a work of destruction and rebuilding on some novel basis; it is analytical through and through, leaving us with an established practice in place but better prepared to think about its purposes and limits.

One of the signal services Duxbury performs in his introductory chapter is to examine what the doctrine of precedent is and is not, and what it does and does not do. The doctrine involves the guidance of present action in light of past decisions, reasoning by analogy. But it is not necessarily the following of publicly accessible prior decisions; nor is it identical to reasoning from experience or to following custom or to the formulation of rules, any of which can be done without the doctrine of precedent or even in contradiction of it. And it is emphatically not the following of only those precedents of which we approve; in a helpfully tough statement of the doctrine in its most demanding form, Duxbury writes that “[t]he authority of precedent is independent of the court’s view as to its correctness or incorrectness” (p.24). As he later argues, however, the notion of precedents as “exclusionary reasons” that “tak[e] the place of individual judgments as to what ought to be done” requires “too tidy a picture” of how the doctrine can be expected to operate in practice (p.108).

When and why did the doctrine of precedent come to be what we now [*1127] know it to be – a form of constraint on judicial decision-making accomplished by tethering it to specific past decisions of sufficiently similar character that those prior rulings must either be followed, distinguished, or overruled? Duxbury argues strongly in his second chapter against the view that “it is with the advent of classical legal positivist jurisprudence that stare decisis becomes established within the common law tradition” (p.37). Indeed, to the extent that Austinian positivism is a “theory of law as coercive orders” (p.14), it lends little support to the doctrine of precedent, since judges are not punished for not conforming to it.

Duxbury’s alternative account is of a more complex confluence of historical developments. Early common law practice prized the “common erudition of the legal profession” (p.32), which came to be embodied by the late sixteenth century in abridgments and compilations of cases; this in turn led to more frequent argument from authority and the elevation of the “artificial reason” of the law as a product not just of many minds but of many minds over time. But a pivotal development was a change in trial conduct that began in the early sixteenth century: from a pattern of pleadings “advanced orally and often tentatively” (p.52) in the course of hearings, practice shifted to the determination by counsel of their pleadings before trial. Such arguments about the meaning and application of legal principles, staked out beforehand by counsel, were now formally the business of the judge to resolve by a decision. In that capacity, judges were expected to state the reasons for a decision in the form of an opinion, oral or written, about the law’s meaning. Stare decisis as a doctrine did not thus emerge full-blown at once, but grew more tightly focused on the past decisions of particular cases as law reporting (including the use of shorthand to record oral pronouncements from the bench) came into its own, particularly in the eighteenth and nineteenth centuries. Consolidating the doctrine’s evolution in England was “[t]he introduction of the hierarchical system of appellate courts” (p.56) in the late nineteenth century, for now the status of the precedent-setting court in relation to the newly deciding one entered the equation.

Leaving aside for a moment what this history cannot explain – why precedents should be raised from the merely interesting to the binding – Duxbury turns in the third chapter to the difficulties that arise within the doctrine of precedent once its authority is recognized. What binds later courts is the ratio decidendi, not the obiter dictum, and so we must know which is which, and that is not always easy. Nor does every case really have a ratio we can discern, or it may have multiple rationes. Or we may have to choose between different accounts of what it is, one offered by the court that decided the precedent and one offered by the court that must apply it. And just what is the definition of a ratio anyway, and how can we determine that we have seen it? Duxbury devotes a great deal of attention to these questions, and the reader’s attention to his discussion will pay dividends. But in any event, cases stand for legal reasons; however inefficiently they do that work [*1128] sometimes, they are hardly useless as guides to subsequent decisions, since they help courts avoid “regularly reinventing the wheel” (p.94).

Chapter 4 takes up the business of distinguishing and overruling precedents. If these are normal parts of judicial behavior, how do they coexist with a doctrine that regards precedents as binding? Distinguishing does not present much of a problem here, since it recognizes that a precedent controls but not here. Overruling is another matter. But in legal systems that prize precedent, overruling a precedent that cannot be distinguished from the instant case is an exercise in respect for precedent, for the overruling must not be willy-nilly: it must be explained and justified. Is there more needed to justify overruling than that a given precedent is perceived to have been mistaken? It seems there is, when a court is overruling a precedent of its own or of a court of equivalent authority. But why that should be so, or just what factors would make a difference, are matters much in dispute. It may simply be a reputational concern, a fear of certain costs to be paid in the respect paid a court that appears to change its mind too often or for reasons not perceived as extraordinary. But can such considerations justify the perpetuation of a wrong?

Duxbury’s fifth and final chapter comes to the nub of the matter: “Why follow precedent?” There are consequentialist reasons: as we have already seen, departing from precedent can appear to carry significant costs in authority or public relations. Or it may be that a doctrine of precedent imparts efficiency or strength to a court. Or that we value the settlement of legal issues, not for a day but durably. Relatedly, we may desire certainty and predictability. These outcomes are not guaranteed by a doctrine of precedent – but what else can guarantee them? Stare decisis is open enough to be flexible, and capable of change. From the standpoint of binding authority, this may seem a vice. From the standpoint of the survival of the rule of law as a viable order, it is probably a virtue.

A weaker case for following precedent exists on deontological grounds. It is difficult to say what is good in itself about treating yesterday’s decisions as binding on today’s. Placed side by side, the two of them look equal in their claim on us; why should the instance that came first in time be privileged over the later one? It does not really help the case for the authority of precedent to say that the past is inherently worth honoring, for no one honors all that is past but only that which is good in it. And the formal justice of treating like cases alike neither requires an actual doctrine of precedent, nor represents what we want to do in all instances, nor helps us in any decisive way in a world in which no two cases are alike in absolutely every respect. We must still reason our way to a grasp of those respects in which likeness is really present, in which the likeness matters to us morally or legally, and in which the institutional conditions of an earlier decision deserve so large a claim on our attention as to condition fundamentally any fresh reasoning we might apply to the situation newly before us. It is hard not to conclude that from deontology we are thrust back into the shadowy but [*1129] more palpable world of consequences, where we cannot so much see our duty as feel our way toward it.

In his abrupt conclusions, Duxbury remarks that the doctrine of precedent has a salutary “capacity simultaneously to create constraint and allow a degree of discretion” for judges, who should treat it as “not an unassailable but a strong rebuttable presumption that earlier decisions be followed” (p.183). It turns out that the authority of precedent is just one of those things: it sprang up from the soil watered by a variety of historic choices of the bench and bar, and it simply seems to work for us. It is not the only possible system for achieving the goods at which it aims, as the civil law system suggests with the much lower value it places on prior decisions. But it serves. Duxbury’s close treatment of the historic and philosophic dimensions of the doctrine reminds me of Aristotle’s remark that political science is among the practical and not the theoretical sciences, and thus has a degree of (im)precision appropriate to its workaday purposes of forging communities out of the oddly shaped human materials available to lawgivers and statesmen.

By way of criticism of Duxbury, I have little to offer. For the American reader concerned with precedent in constitutional law, there is much food for thought here, but it must be borne in mind that the book’s argument is about precedent in common law, and there are differences that make a difference when translating Duxbury’s observations to a constitutional context. There is also a long discussion of the controversy engendered when the House of Lords issued a “Practice Statement” in 1966 that “declared that it would henceforth be free, in some circumstances, to overrule its own decisions” (p.123). For American readers it will seem mystifying that so much energy was expended in debating the propriety and the impact of this statement.

I have two quibbles that I think worth mentioning. First, Duxbury remarks in passing that “more often than not the authority of a precedent will diminish rather than ripen with age” (p.63). This seems worth a more elaborate discussion, especially since it will strike many readers familiar with the older patterns of the common law as stating quite the reverse of the usual principle. After all, precedents have often been called “venerable” for a reason, and usually not because they were decided the day before yesterday.

Second, Duxbury says it is “certainly wrong . . . to define the ratio decidendi as a proposition which, were its meaning to be inverted, would have altered the decision” (p.77). This is certainly true if we seek to define what ratio is in the abstract, because in many cases there will be two or more alternative grounds of decision; inverting just one will sometimes change nothing. But Duxbury never considers a reworked attempt to define ratio that would go something like this: “the ratio decidendi belongs to that class of statements in a case which, if all its members were missing or inverted, would alter the decision.” This would account for the common occurrence of manifold rationes. Though it admittedly does not [*1130] help us dispel all ambiguity about the boundary between ratio and dictum, it has the virtue of being an acceptable definition for the uncertain and contingent world of a common law system governed by stare decisis. Duxbury’s neglect of this possible definitional maneuver was an unusual omission from a book that otherwise left very few unturned stones.

© Copyright 2008 by the author, Matthew J. Franck.