Imagine the following scenario: driving on a California highway, you are pulled over by a police officer, officiously informing you that you have been caught on his radar exceeding the speed limit and are about to be fined. You are in a philosophical mood, and the police officer happens to be very patient and ready to answer your questions. So first you ask him “Officer, have I done anything wrong?” “Of course you have,” he tells you. “You have exceeded the legal speed limit.” “No, no, that’s not what I mean,” you clarify. “Have I done something really wrong? Morally wrong, for example?” “Well, I don’t know that,” the officer replies. “I only know that you have violated the law.” “And is that necessarily wrong?” you ask. The police officer replies that it is not really for him to tell. He just knows that you violated the law. You realize that this is not leading anywhere, so you try a different tack: “Officer, what makes it the law that the speed limit here is, as you claim, 65 mph?” The officer responds by citing you the relevant section of the California Vehicle Code. “But what makes this code the law?” you ask him. “It is the law in California,” he says, “because the code was duly enacted in 1959 by the California Legislature.” “But what makes that enactment the law?” you ask. “After all, enactment, as you call it, is just an event that happens somewhere—people gather in a hall, talk, argue, raise their hands, etc.—why is that the law?” Still patient, the police officer explains to you that the California Legislature is an institution created in accordance with the Constitution of California, and granted by it the authority to enact such laws as the California Vehicle Code according to certain prescribed procedures. “Now I see,” you reply. “The California Vehicle Code is the law around here because it was enacted by a legislature, and the legislature’s authority to make such laws was granted to it by some other law, the Constitution of California. I get it. But what makes the Constitution of California the law around here?” The police officer knows the answer: “The California Constitution is the law because it was duly enacted by the authorization granted to the people of California to have such a constitution by the Constitution of the United States.” Still, you are not quite satisfied. “What makes the U.S. Constitution the law?” you ask. “The U.S. Constitution? It is the supreme law of the land!” the officer proudly proclaims. “Yes, yes,” you tell him. “I know that it is what people say. But what makes it the supreme law of the land?” you ask him again. Not surprisingly at this point the police officer loses his patience, hands you the citation and off he goes.
As a curious philosopher, you remain unsatisfied by this little encounter. The police officer dodged two crucial questions: first, you still do not know whether you have done anything wrong, though you ended up with a hefty fine that you have to pay. Is this all there is to it? Violating the law is just unpleasant, resulting in something bad that happens to you if you are caught? Mustn’t the law assume that you deserve to be punished? Second, you still have no idea what makes anything “the law.” You have learned that certain actions or events in the world gain the legal significance that they have by way of an authorization granted by some other law; but this chain of authorization comes to an end; at some point we end up with something that amounts to “the supreme law of the land,” the kind of law that authorizes the creation of all other laws. But what makes that the law? You still have no idea.
Nevertheless, this encounter was not without merits. Although we have not learned much about the answers, at least we know what the questions are. The two main questions that the police officer has dodged are the main questions about the nature of law that have preoccupied philosophers for centuries: one is about the normative character of law; the other is about law’s conditions of validity. As we shall see, both of these questions concern, though in different ways, the relations between law and morality. I will take them up in reverse order, starting with the question about the nature of legal validity.
The Conditions of Legal Validity
Whenever one says, “It is the law that X” or, more precisely, “according to legal system S at time t it is the law that X,” one expresses a proposition that is either true or false (or, perhaps in some cases, it is indeterminate whether it is true or false). The philosophical question here is: what makes it the case that the proposition is true (or false)? What does it generally depend on? Notice that it is not a legal answer that we seek here. As we learned from the encounter with the police officer, the legalistic answers are bound to run out. In other words, we are not asking what makes this or that particular requirement a legal one, but what makes anything at all legal to begin with?
Sometimes the best way to understand a philosophical question is by way of looking at the controversies that exist about its answer. The main controversy about the general question of legality is, ultimately, about the possibility of reduction. Can we fully explain the conditions of legality in terms of something else, more foundational in nature? In particular, is it possible to reduce the conditions of legal validity to social (viz., nonnormative) facts, namely facts about people’s conduct, beliefs and attitudes? Or is it the case that no such reduction is possible, partly because the content of legal norms, especially their moral content, also bears on the conditions of their validity?
It is the nature of theories, in general, that they try to explain one thing in terms of another. There is, however, a type of theory or explanation we call reductive: if there is a clear demarcation of one type of discourse or class of statements, and we can provide a full explanation of that class of statements in terms of some other type or class of statements, then the explanation is reductive. For example, if we could fully explain the realm of our mental life in terms of truths about the physical aspects of the world, we would have provided a reduction of the mental to the physical realm. It is this kind of reduction that we need to consider about the legal case as well. Some legal philosophers have claimed that we can explain what constitutes legality in terms of something else, more foundational in nature. According to this line of thought, in order to answer the question about the conditions of legal validity, we need to have an account of the relevant social facts, concerning the ways in which people behave and the kind of beliefs and attitudes they share about their conduct. Other philosophers, however, deny the possibility of such a reduction. They claim that legality is not fully explicable by social facts, and mostly because legal validity is partly a matter of moral truths. At least in some cases, they claim, truths about morality determine what law actually is.
With one notable exception (Hans Kelsen’s “pure theory of law”; see Marmor 2011, ch. 1), the jurisprudential tradition called legal positivism (Postema 2012) strove to provide a reductionist account of legal validity. One of the most influential nineteenthcentury legal positivists, John Austin, propounded a theory of law that is perhaps the clearest example of a reductionist account of law. According to Austin, law consists in the commands of the political sovereign. A command is defined by Austin as the expression of a wish by one person (or group of persons) that some other person(s) behave in a certain way, backed by a threat of sanction if the addressee does not comply. “Do/don’t do this … or else …” And then sovereignty is defined sociologically: the sovereign is the person or group of persons who are habitually obeyed by a certain population and are not themselves in a habit of obeying anyone else (Austin 18321832).
Thus, according to Austin, the conditions of legal validity are fully reducible to facts of a nonnormative kind: facts about the relevant social reality constituting political sovereignty, identified by habits of obedience, and facts about actual commands issued by (or on behalf of) the sovereign. Though there is something intuitively compelling about Austin’s model, later legal positivists have subjected the command model of law to fierce criticism. In particular, H. L. A. Hart, in his seminal work The Concept of Law (1961), devoted three or four chapters to a detailed critique of Austin’s theory. But a careful reading of Hart should reveal that it is not Austin’s reductionist project that Hart objected to, since his own theory of law is just as reductive. Hart’s disagreement with Austin is about the main building blocks which are supposed to constitute the foundations.
Austin’s main insight has two components: that law always consists of commands (do this … or else …), and that it always originates with the political sovereign. Hart argued that both theses are fraught with difficulties. To begin with, only a small fraction of law has the structure of a command. The main provisions of a criminal code, or traffic regulations for that matter, do have such a structure: they impose an obligation (do/don’t do this …), backed by a threat of sanction for failure to comply (or else …). However, a great many laws, perhaps even most, do not purport to impose an obligation; they grant legal powers of various kinds, that is, the power to introduce a change in the normative relations that obtain between the relevant parties concerned. Consider, for example, the laws which determine how to form a legally binding contract; such laws do not have the structure of “do this … or else.” The law is not in the business of telling anyone to form a contract; it is up to the parties whether they want to enter into contractual relations or not. The law only provides a tool here: it designates certain ways in which people can introduce a change in their legal relations, if they want to do so. And this is more typical of most of the areas of private law and many other legal domains. Most of the law is not like the criminal law, where the law imposes an obligation backed by a threat of sanction (Hart 1961, ch. 3; for more details, see Marmor 2011: 36–40).
Furthermore, H. L. A. Hart and other contemporary legal positivists have argued that early positivism exaggerated the centrality of sanctions, and more generally, law’s coercive powers, to an understanding of what the law is, and what its main functions in society are. No doubt, the law is a coercive order (Edmundson 2012), and many laws are backed by a threat of sanction for failure to comply. But it is very doubtful that this element of force or law’s ability to coerce its subjects is as central to law’s main functions in our lives as people tend to assume. In order to see this, consider a thought experiment suggested by Joseph Raz (1990/1975: 158–60): imagine that there is a world inhabited by creatures who are exactly like us, with one important difference, namely, that whatever it is in human nature that requires the law to threaten us with sanctions, those people do not have. They just do not need to be threatened with sanctions in order to comply. Now ask yourself, would this world have need for various norms and institutions that are similar to what we call law in our world? The answer is clearly yes. Even in this world without need for sanctions, people need rules to solve large-scale coordination problems, and rules and institutions to determine what needs to be done when people have reasonable disagreements about their collective endeavors. They would need institutions to determine matters of fact in conflictual situations, and so on and so forth. In short, the law’s ability to use force and compel compliance, important as it is, is actually much less important than usually thought. Many of the functions that law serves in our public lives have very little to do with the need for coercion (for some reservations about this argument, see Marmor 2011: 42–44).
Hart’s second main disagreement with Austin’s reductionism consists in his argument that we cannot explicate the sources of law in terms of political sovereignty because the very idea of sovereignty is a juridical one. Law partly constitutes our conception of sovereignty; it cannot be reduced to it. We can only identify the political sovereign on the basis of the constitutional legal structure that prevails in the relevant community, not the other way around. Austin’s idea that habits of obedience are sufficient to constitute political sovereignty, Hart argued, is not a workable idea (Hart 1961, ch. 4).Think, for example, about this question: who is the sovereign in the United States these days? That is, think about the question: Who is the person, or group of persons, that U.S. residents habitually obey, and itself is not in the habit of obeying anyone else? Mindboggling question, no doubt, but whatever answer you come up with would have to be based on facts that you know about U.S. constitutional law. In short, we just cannot suggest that law is whatever the sovereign commands, because the very identity of the sovereign is law dependent. First, we must have some legal regime in place, then we can come up with some notion of who the sovereign is (that is, at least as long as we seek to identify the sovereign as a source of law).
The failure of Austin’s command theory of law does not necessarily doom the reductionist project. In fact, Austin begins with a sound intuition. Even if laws are not necessarily commands, as he thought, they are, by and large, products of deliberate human creation. Laws are created by human agents, enacted by institutions or persons in various institutional roles (Kelsen 1961/1945: 110–11). We need not assume that this is necessarily or always the case; suffice it for now to admit, as one must, that it is typically the case that laws are made, that they are products of an act of will. But of course not everyone who might want to enact a law or change one actually can. My proclamations to the world, wise and commendable as they may be, are not going to have any legal effect whatsoever. I am not in a position to make laws or change existing ones. But then, who is? And more importantly, what makes it the case that anyone is in such a position? Hart’s answer boils down to this: certain agents get to make (or modify) the law in a given population in virtue of the fact that most people in the relevant society regularly behave in ways which assume that these agents get to do so, believe that to be the case, and share certain critical attitudes that are in line with this shared belief. In every legal system in place, there are some social rules followed by the relevant population determining who gets to make law and how it is to be done. Hart called such rules the rules of recognition (1961, ch. 5). The social rules of recognition constitute what we take to be the conditions of legality in the relevant legal system. (So now, if you want to chase down the police officer who got tired of your questions and tell him what makes it the case that the U.S. Constitution is the supreme law of the land in the United States, you could tell him that according to H. L. A. Hart, it is the case simply because most Americans just take it to be the case. It is the main rule of recognition we follow around here; certainly the rule that judges, lawyers, legislators and other officials follow, which is what really matters.)
Notice that this is just as reductionist an account of legal validity as Austin’s, albeit based on different building blocks. Legality is fully determined, according to Hart, by the social rules of recognition. A social rule consists of actual patterns of conduct accompanied by certain beliefs and attitudes. We have a social rule when there is a regularity of behavior in the relevant population, attended by a complex idea that Hart called “acceptance” of a rule, which consists of (1) a belief shared by most members of the population that the existence of the rule provides them with reasons to comply and (2) a shared attitude of positive endorsement of the rule that is manifest in its use as grounds for exerting pressure on others to comply and criticizing them if they do not (Hart 1961: 82–86).