I Do Analytic Jurisprudence. This Is Why It Matters

This August, law professor Lawrence B. Solum published a blog directed to law students entitled, “Why Read The Concept of Law by Hart?” The Concept of Law, for those who do not know, is a major, and perhaps the most important and influential, work of the 20^th century in what is called “analytic jurisprudence” or sometimes “general jurisprudence.” Analytic jurisprudence is a branch, arguably the core, of philosophy of law. It seeks to answer fundamental questions that apply universally to laws wherever they exist or are imagined, such as: What is law? What makes a standard, principle, or rule legally valid? What makes a group of laws a legal system? What is the nature and structure of legal reasoning? What role does language play in the law created by a statute? These are questions Hart famously addressed, at least to some extent, in his book.

I take Solum to be responding indirectly to a fairly recent notion that seems to be influencing the teaching of jurisprudence in American law schools. (See his account of the “decline of analytic jurisprudence” in legal scholarship and law professor Brian Leiter’s survey of the teaching of analytic jurisprudence in American law schools.) This idea, stated bluntly, is that it is valueless for would-be lawyers to study analytic jurisprudence. Analytic jurisprudence has no practical value for lawyers, in whatever law role they may come to occupy, including that of judge. Rather, it is thought that law schools ought to teach theorizing specific areas of local law (such as American or common law tort law) and what is called “normative jurisprudence,” which seeks to answer moral questions raised by law and roles in the legal system.

Some legal scholars have gone further and argued that analytic jurisprudence is valueless for philosophical investigation itself. More precisely, they have argued that analytic jurisprudence as a search for the correct idea of what law is, as if law has some unique nature, ought to be abandoned in favor of what might be called “general normative jurisprudence,” or moral philosophy applied to the laws (of a legal system, conceived of in several different ways) or its legal history or legal practices. I have in mind law and philosophy professor Scott Hershovitz’s 2015 “The End of Jurisprudence.” (And see also law professor Hillary Nye’s more recent “Does Law ‘Exist‘? Eliminativism in Legal Philosophy.”)

I argue, as Solum did in his blog to some degree, as Michelle Dempsey, professor of law at Villanova, did in a 2016 speech to the Future of Legal Scholarship Symposium (revised and printed in The Journal of Legal Education) and as NYU law and philosophy professor Liam Murphy did in his 2014 book, that it is simply false that analytic jurisprudence is useless and valueless for future legal professionals or for philosophical investigation.

The first reason is that it is impossible to do a (very) useful job of normative jurisprudence without a sound analytic jurisprudence. Consider theories of adjudication. These theories concentrate in common law systems on adjudication in appellate courts, where the questions raised are about the local law (and not about the facts of the case). Nye argues that a normative theory can be fashioned to recommend to judges what they ought to do in various situations without recourse to the right answers (if they exist) from analytic jurisprudence.

But one problem is that the various situations do not come readily labelled, particularly in appellate litigation where there are respectable legal arguments for both sides. Suppose you have a normative theory of adjudication saying judges ought to do x when faced with law that has property P (e.g., is unjust), y when faced with law lacking P, z when faced with indeterminacy in law, and so on. No matter how well-reasoned such a theory is, it is useless on a practical level (“academic” in the pejorative sense) and in danger of moral misuse by the well-intentioned unless judges can accurately identify when they are faced with existing law, legal indeterminacy, or outright gaps in the law. (Law is indeterminate when it is uncertain and fails to provide a single, clear answer to a specific legal question.)

Such identification isn’t always straightforward, even for legal experts. This is particularly true of appellate cases law students are apt to study, at least in the United States. Suppose that the Supreme Court of the United States says there is a right implicit but never previously declared in the Constitution, as it sometimes has, based on its understanding of some moral principle also not previously authoritatively declared. Is a subsequent appellate judge now presented with a newly discovered part of law, a newly created part of law, or something that isn’t yet law and may never be? Can there be such a thing as implicit Constitutional law to be discovered? Can there be such a thing as judge-made law in a democracy, or will it simply be judge-expressed opinion and a bald exercise of extralegal power?

Judges need an answer that is more helpful than “it depends on which way you conceive of law.” To answer these questions correctly, we need analytic jurisprudence, and, in particular, that part of it that is legal metaphysics. By “legal metaphysics” I mean the investigation of what kind of thing law is and what kinds of facts can give rise to it.

Recourse to analytic jurisprudence will be needed even if you avoid using the term “law” in your theory of adjudication and couch this theory in terms of what judges should do when presented with “legal materials,” “legal practices,” or “legal history.” For practices (and the like) don’t come labelled as “legal,” either, that is, as the right sort to be the subject of whatever theorizing, metaphysical or moral, legal scholars should be doing. What makes practices count as legal practices is a question for analytic jurisprudence. Hart’s idea is that practices cannot become legal practices without laws creating their framework and their significance in the legal system. No one has come up with a better explanation.

We need legal metaphysics for another reason, too. A normative theory of adjudication that presupposes the existence of a situation—gaps in the existing law, for example, or unjust law—that is impossible in the legal system in question or impossible in any legal system is defective in an avoidable way.

And what should we make of a normative theory that holds that judges ought never to make law (in a democracy, at least) if we discover from analytic jurisprudence that this activity is inescapable? If all holders of a certain legal role ought not to do something that they cannot avoid doing in that role, the next moral question is whether the role ought to be abolished. And if the answer to that question is negative, the moral claim that judges ought never to make law is only pro tanto at best. That means that when all relevant moral considerations are taken into account, it would not change the moral bottom line about what judges should do. This pro tanto claim from normative jurisprudence would not then be of any particular practical interest to would-be lawyers—or would-be judges, for that matter.

The second reason is that the right answer to what law, precisely (if at all), an appellate court decision has laid down affects more than the application of a theory of adjudication. In common law, appellate court decisions are often precedent; and unfamiliarity with various tools of analytic jurisprudence has led in legal circles to the inability to reliably identify when some precedent is followed or distinguished. The right answer to what law, precisely (if at all), a precedent has laid down depends heavily upon correct answers in analytic jurisprudence. (For an overview of leading views, including mine, see “Precedent” by Grant Lamond.) And that informs, or should inform, not only judges seeking to follow precedent, but criticism of specific judges and courts for deviating from established precedent or established law.

Third, analytic jurisprudence, though not Hart in particular, has much to contribute to avoiding unintentional equivocation and confusion with terms that have several possible related meanings: e.g., “legislative intent,” “legal interpretation,” “the meaning of a statute.” Too often in the literature and the manuscripts I am asked to review, legal scholars’ lack of clarity on these matters, even in normative jurisprudence, has led them to mistakes in reasoning and thus to undefended (if not implausible) conclusions.

Fourth and equally important, judges’ assumptions on analytic jurisprudence have influenced their reasoning in appellate cases directly and, thus, determined the fates of real people. While such events are rare, there are, for example, cases in which some legal philosopher’s analytic jurisprudential view was relied on to justify a legal decision. (See, for example, citation to Ronald Dworkin to explain the task of statutory interpretation in the Canadian cases of R. v. Paré, [1987] 2 SCR 618, para. 16; the use of Hart’s contrast between mere habits and rules in the Australian case of Members of the Yorta Yorta Aboriginal Community v Victoria, (2002) 214 CLR 422, para. 41 and 42; and a citation to Hart’s explanation of what to look for to determine legal sovereignty in Joose and Anor v. Australian Securities and Investment Commission [1988] HCA 77, p. 16.) But as this blog is partly about American legal education, the case I will highlight is American: the US Supreme Court case of Erie R.R. v. Tompkins, 304 U.S. 64 (1938). 

In it, federal courts were directed by a federal statute to apply state law to a dispute between a seriously and permanently injured man and a railroad corporation whose employees had been careless in not fastening a railcar door. Judges did not agree on what the state law said because they did not agree on where to look for the applicable state law on a key legal issue in the case. There was no state statute on the matter. More to the point, judges did not agree on what kind of thing existing state law would be. And, for those of you who do not know the legal background, they had no option to declare that state law did not exist on the matter.

Some of the judges, both on the Supreme Court and on the lower appellate courts, thought that law could exist without legislation and in unwritten form. It could exist on the federal level as a generalization of principles implicit in the law of the entire common law system (which includes all the states but one, England and the rest of its former colonies). These are claims in legal metaphysics. According to these same judges, the opinion of a particular state court or courts on what that unwritten law says does not amount to law and can be disregarded by a federal judge.

Judges who held this view thought, further, that the federal general or common law became state law in the absence of a state statute. Applying it would have given the injured man monetary compensation, which he likely needed, having lost an arm in an era when his best chance of employment would have been manual labor. But the majority of the Supreme Court rejected these judges’ ideas. Justice Brandeis, writing for the majority, thought that courts in the U.S. were not free, as a Constitutional matter, to apply this alleged federal general law when state law existed. And he thought state law on the matter did exist.

But Brandeis’s reasons went further, a point it seems to me too often overlooked in the teaching of this case. He quoted a famous former Supreme Court Justice, Oliver Wendell Holmes, saying that law does not exist “without some definite authority behind it,” and that the legal authority in a state is the “voice of the State.” In other words, either a state’s legislature or a state’s highest court is the proper origin of a state’s laws; so state laws include its judicial decisions. A state’s laws do not include general principles drawn from other jurisdictions, no matter how many external jurisdictions seem to converge on a particular principle. There is no such thing, Brandeis said, as federal general law.

These claims by Brandeis are independent of any interpretation of the Constitution. They were accepted by concurring judges who refused to engage in Constitutional interpretation. They are claims about the kind of thing a law is. In accordance with them, Brandeis looked to the rule implied or expressed in a previous court opinion of the highest court in the state in which Tompkins’ accident occurred. That rule’s import: the railroad corporation didn’t owe Tompkins a cent.

Whatever moral injustice this Supreme Court decision may have represented, it would have been worse if the Supreme Court got the law wrong. For if a court decides that the applicable law is against a plaintiff, the applicable law should really be against the plaintiff; the court should have correctly identified existing law. But it takes legal metaphysics to determine if a state’s law must always have a definite state authority behind it within that state positing it at a dateable time, or if a state’s law might be created indirectly as a result of court decisions in related but different states that converge on a single rule or imply a single principle. For this reason, training in sound analytic jurisprudence, for the judges and for the attorneys arguing before them, would have had clear practical value.

The last point, which I will address much too briefly, is the case for abandoning the philosophical pursuit of analytic jurisprudence. That case has not been made. One cannot assess the moral impact of laws (or legal practices) without first making some assumptions about what kinds of things laws are; generalized moral assessments are useless in practice unless they can be applied; and the tricky part of applying any such assessment is when it is unclear that something is a law because of its kind or type (e.g., an unwritten principle never declared by any local legal authority in the past, an order by a high legal official that conflicts with the plain meaning of a statute or constitutional provision, a long-standing practice).

Whether or not the things I just mentioned are laws makes a difference to the moral position of a judge presented with them. For it takes more or somewhat different considerations to morally justify deviating from laws than from nonlegal standards or orders, given judges’ oaths and some rule of law values. Some deviations from standards or orders that are laws may not be justified when they would be justified if the standards or orders were not law. And the fact is, judges aren’t the only legal officials who take oaths to uphold, apply, or enforce the local laws.

I deny, therefore, Hershovitz’s claim that “jurisprudence should drop the metaphysics and take up morals.” He reminds us that metaphysics of law is hard, we haven’t reached agreement, and there seems to be something wrong with everyone’s answer so far (though his arguments do not depend on any of these points). Nye adds that questions in analytic jurisprudence cannot be resolved empirically (that is, from experience), a point that surprises no philosophers, and that intuitions clash on the nature of law. None of these, however, are reasons to abandon a philosophical inquiry. They are not reasons to conclude, as Nye does in her article, that “there is no way to settle who is right.” If they were, much, if not all, of Western philosophy should have ended centuries ago. Similar complaints apply to moral philosophy, including the part that addresses the question of what it is, exactly, that makes actions right or wrong. The answer to that question is essential to doing normative jurisprudence well.

So I do analytic jurisprudence, and I encourage others with philosophical training to do it. It is needed—in legal scholarship, in litigation, and, sometimes, even in life.