These days, I suppose that most people who think about the question of where laws and legal systems come from, have little doubt that they come from people. This view that our statutes, case law, and approved governmental procedures didn't somehow emanate from angelic legislators, but have been patched together here on Earth (like, it's often said, sausages) is called "legal positivism," and it's not only quite intuitive, it's been an extremely popular view--during my lifetime, anyhow. I suppose the "natural law" theory, though it still has its fervent adherents, had its heyday in St. Thomas Aquinas's era.
But if even if the general idea of positivism now reigns supreme, it can't be denied that there are some fairly deep differences among the positions advocated by some of positivism's most widely known champions. In particular, perhaps the movement's two greatest figures--the Victorian Benthamite John Austin and the 20th Century Oxonian H.L.A. Hart--were on the opposite sides of several central issues of legal philosophy. For example, while Austin insisted that every law is essentially an order backed by threats, deriving ultimately from the sovereign, i.e., the most powerful person or group in the land, Hart denied such a reduction. On his view, where there are laws, there must be rules, non-purely-behavioral norms that produce obligations from a specifically "internal point of view." So, while both philosophers were, let's say, Earthians rather than supernaturalists, Austin took the behaviorist (and exclusively "external") view that laws are just commands emanating from habitually-obeyed-and-subject-to-no-higher-power sources, so long as those commands reliably produce the demanded responses among those who are subject to them.
Each of those two eminent theorists required both the procedural imprimatur (to be authoritative, laws must have been created in the appropriate way) and habitual compliance among most citizens. But Hart argued not only that many laws simply aren't commands backed by threats of penalties for scofflaws, but also that, even if they were nothing but orders of that kind--each analogous to "Your money or your life!" widespread habits of compliance (even among just the right people) would be insufficient to make them lawlike. To be obliged or compelled, Hart argued, isn't the same thing as to carry out an action as a result of an obligation. And Hart argued that laws uniquely carry obligitoriness--or at least they always produce that sense among the officials who are in charge of carrying them out. So, even in those cases where a large segment of the citizenry just obeys because they're worried about trouble they'll be in if they fail to do so, the appropriate internal attitude among those in charge will make a command or power-conferring rule a law.
Like all the most world-altering and influential philosophy books (and for all I know, this is true in every discipline), Hart's book has suffered from the notorious "thousand cuts" delivered by critics. As one philosopher put it to me long ago, when Plato wrote his Republic, Spinoza his Ethics, Kant his first Critique, Rawls his Theory of Justice, Wittgenstein his Tractatus, etc., it was like a huge and majestic elephant taking a step out from a dense jungle into an open, sunlit field....only to be slowly and painfully devoured by a multitude of army ants. Whether or not that analogy is entirely fair, it can't be denied that Hart's classic has suffered that fate to some extent. In particular, numerous criticisms have been levelled at his reliance on "secondary rules" to do the work of giving putative laws the teeth of authority and legitimacy. Hart claimed that it is these second-order rules, items like To be engrossed in the House, a bill (such as a penalty for burglary) must receive three readings" which provide the authority of first-order instructions regarding how we are supposed to behave. He argued that it is the existence of secondary rules and their placement in the array of injunctions and proper procedures in which they are found that indicate when any lower-order rule is part of a legal system.
Hart dubbed the key secondary rule (or rules) by which we can tell both that some instruction is a part of a legal system and just which system it belongs to "the Rule [or Rules] of Recognition" and, unsurprisingly, the literature both supporting and criticizing this innovative concept is particularly immense. It is not my intention to rehearse or reject these critiques here. It may well be that attempts to define "law" will always mirror what Steven Sondheim once suggested to Ned Rorem is the case for operas. Sondheim said that if we want to know whether some piece of musical theater is an opera rather than a musical, the best way is to check whether or not its performances take place in opera houses. And it may be the same thing for determining what makes for the sorts of laws that relate to jurisprudence: perhaps what matters is simply the extent to which the items are found to be relevant to judges, police, governors, corrections officers, tax collectors, legislators, and so on.
No doubt this suggestion will be claimed to reflect an objectionably circular approach: knowing which are the appropriate "judges" or "legislators" to consult is no different from knowing which buildings are really opera houses. Is it enough that some building happens to be called "The Royal Opera House"? Couldn't "Pajama Game" be performed there anyhow? The point is that, since the terms "law" and "opera" are both specimens of what are sometimes called "cluster concepts," it may not be possible to do much better than arguing in a circle or producing Wittgensteinian metaphors about family resemblances or the interweaving strands in a rope. There just may not be any precise group of necessary and sufficient conditions to which one can point.
In any event, my general sense--and I believe it's a fairly orthodox position--is that Hart's arguments to the effect that legal systems cannot be reduced to Austin's "orders backed by sanctions" are generally sound. There is evidently more to a law of the land than "Your money or your life!" even when that threat comes from someone with the clear power to act on it. To point this out is not to say much more than that a tiger compulsively circling her cage is not "following a rule": there seems little doubt that laws are some species of rule (or "norm"). Furthermore, a law declaring the cranberry muffin the official Massachusetts baked good doesn't even require a single person to do a single thing. So it's pretty clear that a behavioristic command theory can't be right.
I do think, however, that Austin and his greatest British successor were not quite so far apart as Hart believed with respect to what makes an instruction or other rule a law. This point can be gleaned from a couple of the earliest published complaints about The Concept of Law. I refer here to objections made by two of the lead "army ant" soldiers who reviewed that masterpiece in the early 1960s: George Pitcher and Rolf Sartorius.
It will be recalled that Hart claimed that the combination of primary and secondary rules provided "the key to the science of jurisprudence." This may be so, but both Sartorius and Pitcher were quick to point out that the mere existence of "a union of primary and secondary rules" could never be sufficient to produce a legal system of the sort in which Hart was interested. After all, not only does the Catholic Church have its canon law, complete both with various penalties for non-compliance and with detailed methods of adjudication, but even the NCAA has a penal code! Surely, when we want to set off the system of laws governing a particular country or political subdivision, we generally mean to exclude church or club "laws" and their own systems of governance. We may concede to the cluster theorists that we cannot provide a perfect specification of necessary and sufficient conditions to demarcate those legal systems in which we are particularly interested, but it must at least be possible to give a rough and ready way to separate such systems from the rules of the Church of Scientology and those of the NFL! The problem is that these other codes can't reasonably be denied to also be "unions of primary and secondary rules." In addition, each may have its own penalties for failures to comply, each its own methods for amendment, and each its own adjudication architecture. We can thus see that the presence or absence of some other feature must prevent them from being "real legal systems." But if it isn't those formal characteristics, what does make the difference?
To be fair, Hart was not unaware of the problem pointed out by Pitcher and Sartorius. Indeed, he had anticipated the need for an additional mark of the legal in his book when he appealed to what he called "the minimum content of Natural Law." He recognized that the purely formal characteristic of "being a union of primary and secondary rules" was insufficient and suggested that true laws must also reflect a certain (if vague) sort of humanistic goal to be part of an authentic legal system. It seemed to him obvious that "there are certain rules of conduct [like, say, not murdering anybody whose sandwich you think you might want] which any social organization must contain if it is to be viable," Hart therefore proposed that bona fide legal systems are limited to those in which the primary rules contain such "minimum content."
As may already be clear from what has been said above, this move from form to substance won't work either. Indeed it was pointed out in a particularly acute anonymous Note in the Yale Law Journal of 1975 [which I understand--thanks to Nelson Barrette--to have been written by eventual Yale Law School Dean, Anthony Kronman back when he was himself a student at that institution]. The problem is that no particular content, however laudable, can actually cut the ice as a distinguishing feature of law. After all, religious and moral codes famously express the same sort of goals, and many club rules do as well. Presumably, for example, contestants in a sanctioned chess tournament are forbidden from strangling their opponents during a match--or even from throwing captured bishops at them.
Hart suggested several other candidates for marks of true law, but Kronman's article patiently dismantles all of them....except one. And that characteristic is a matter neither of a rule's form nor its content. What makes an ostensible law part of an authentic legal system is that it does not require a dispensation from any independent system to implement physical sanctions. For consider: if the NFL wishes to enforce any of the fines on its books, it will be able to do so reliably only if such actions are allowed by the U.S. government. Similarly, if a church wants to impose corporal punishments on those it considers sinful, those actions will have to be allowed by the state. As noted by Kronman (following Max Weber), a modern state is a "community that (successfully) claims the monopoly of the legitimate use of physical force within a given territory." Thus, for any system but that emanating from "the sovereign," an explicit dispensation to apply physical penalties from that sovereign is required. It is that feature alone that makes any putative arrangement of "laws" an authentic legal system. And that is neither a matter of form nor of content.* Indeed, it seems to trump content considerations.
This may seem like an arcane matter of philosophy, but it has great practical importance: it provides us with a necessary condition for any putative law to have actual authority. As I argued both in my 3:16 AM review of the recent Huemer/Layman book on political authority and my own book on democratic theory, to have bona fide authority, a rule must have the right parentage. It must come from that entity that has the most power in any region.
So much for authority: elements of legal systems must have proper form and causal background. But what about "legitimacy"? Shall we call every authoritative law legitimate? Many will deny this. They will say that to be legitimate, a law must also be morally acceptable, that any putative law allowing for the murdering of anyone whose sandwich one covets--or, to take a more serious case, an imprecation calling for the rounding up and murdering of Jews, Gypsies or Catholics--is always illegitimate, no matter how "authoritative" its history and formal/relational characteristics may make it. I agree with that myself (at least with respect to the rounding up and murdering case). Does this mean that we must bring back content--that to be legitimate laws must reflect the appropriate ethics? Must the widespread appeal of such intuitions spell the end for legal positivism of every stripe? Certainly it has seemed so even to many of the friendliest critics of both Austin and Hart.
I don't think those critics are correct. I believe there is a way that even those who are skeptical about all claims to know the truth of some proposed ethical maxim can manage to distinguish the subset of authoritative laws that are legitimate from those that are not. In my view, this is the point at which self-governance, i.e., authentic democracy, comes to the rescue. I have argued, both in my book and elsewhere (see, e.g, my interview with Richard Marshall) that there is nothing like a plausible background theory to substantiate our moral intuitions--even if, in some cases they are warranted (by our emotional responses) and happen to be true. They are in this way like astrological claims: such warrant as is provided by our emotional sympathies may be no more reliable than induction from previous astrological forecasts that happen to have been successful. To put this in terms made famous by Susan Haack, our moral intuitions are like "across clues" in a crossword puzzle containing not a single "down clue." It is not that moral propositions are "subjective" or necessarily "true only relative to a particular society or culture." Still less am I claiming with J.L. Mackie that every moral claim is erroneous. What morality seems to me lack is neither objectivity nor truth, but a background theory with sufficient plausibility to provide anyone with knowledge of a single moral ought proposition. Utilitarians and other consequentialists have tried to provide such theories, and divine command theorists have as well, but nothing has found wide acceptance. Indeed it is my view that no proposal has made much more sense than the astrologer's admonition that "The stars impel, they don't compel." We have evidence only in the form of across answers that fit.
But if we can't actually know any ethical truths, we can take some comfort from the fact that our our epistemic situation may be in much better shape with respect to prudential values--that which makes some thing or event, not necessarily good, but good for us. This is so because it is reasonable to postulate that, at least ex ante, getting what one wants is a (prudentially) good thing. That second helping or fancy TV may not be good for us in the long run, but if we forget about consequences for a moment and just focus on whether it is intrinsically better to get what one has freely chosen than to strike out, it seems sensible (though assuredly not provable) to say that successfully obtaining what we freely choose is better than nothing (i.e., better than utter failure to acquire what we want).** Again, we need make no assumption that any particular choice, whether or not coerced, must produce any long-term benefits for the chooser. For it is of course the case that any one of them might be disastrous not only in the eyes of a third party expert, but even according to the ex post assessment of the chooser herself.
What does any of this have to do with legal positivism? In brief, it provides a rational basis for the provision of legal legitimacy in self government. This is because what is true of individuals seems to also be true of groups: getting more of the things groups or societies want is an intrinsic ex ante prudential good for groups as well. This means that a shift toward prudential values for our foundations allows political theorists, for perhaps the first time, to explain legitimacy without reference to morality. We can instead say that authoritative laws will have the additional feature of legitimacy if and only if the (monopolistically powerful) states from which they emanate are truly self-governed. Which democracies actually are the "authentic" ones? Roughly, they are the ones in which frequent and fair voting mechanisms reliably result in good faith (and transparent) attempts by sovereigns to get the majority what it wants, while also providing minorities with appropriate voice in government.***
The first and most pressing objection to this theory that is likely to come to mind may bring us right back to the threat of the carting off and slaughter of whatever ethnic, racial or social group the majority happens not to like. How can simply satisfying the desires of the majority prevent such horrific occurrences without the importation of moral values? Aren't there cruel majorities to be found in the world?
The answer to these questions that are provided by what I have called "distilled populism" is that such values as are needed to prevent those sorts of atrocities can be derived from the concept of "fair voting" alone. That's because there can be no authentic democracy unless every citizen is treated with equal respect. Thus, we don't need to say that unequal treatment is evil. We need only understand that we cannot actually discern what the people want if unfair discrimination is not prohibited. Since real democracy absolutely requires both the equal treatment of votes and equal protection of individuals, even if it arguably cannot without majority disapprobation prevent a system of (legitimate) laws allowing for whimsical murders by sandwich coveters, it must always prohibit a Holocaust. Moreover, I believe that no legitimate legal system can even allow for such anti-democratic atrocities as exclusive reliance on first-past-the-post electoral schemes; indeed, any system including the U.S. Electoral Collage or the U.S. Senate's Filibuster rule cannot enact rules that will pass a legitimacy test requiring authentic democracy. And majoritarian strictures go beyond equal treatment and protection. They will also limit restrictions on free speech, association, and assembly--all, again, without appeal to what may or may not be ethical or "endowed by our creator." They are simply needed to determine what the people really want. What's more, for elections to be fair, "political rights" would seem to have to go farther than the U.S. Constitutional prohibitions, which say only what sorts of rules our government may not enforce.
I won't try to restate my entire theory of distilled populism here: the full picture may require reading my book. But I do want to close this precis by turning back to Hart's version of positivism--and the youthful Anthony Kronman's note on it. Good as the later piece is, I find that it did overstep a bit. The problem involves its concluding lapse into Austin's behaviorism. One can--indeed one must, to get this right--accept Austin's causal theory without adopting any picture that entails that a successful Skinnerian training of the citizenry to respond appropriately to governmental commands would be a legal system. Donald Black's quite extreme 1976 book, The Behavior of Law, put it that "Law is governmental social control....a quantitative variable [that] increases and decreases." Well, sort of, but nothing prevents a government from closely controlling the activities of its citizens without having a single genuine law on its books. Kronman claims that Hart and Austin ultimately agree that "A legal system is a set of rules laid down for the guidance of human beings by a determinate person or group of persons having power over them by virtue of their effective monopolization of the physical sanctions within a particular jurisdiction." This rule-containing description is better than Black's reliance on behaviors alone, but if we are interested in legitimacy as well as authority, not every "determinate person or group" (i.e., government or sovereign) can make legitimate laws by promulgating rules and having the power to enforce them without dispensation from any other person or group. Authentic democracy is also needed.
To conclude, I believe that Austin, Hart and Kronman, while generally on the right track regarding what makes for a legal system, get some important details wrong. If the U.S. has laws--and certainly it would be a gross violation of common usage to deny that!--their existence, as summarized above, is a result of the formal and causal properties necessary for authoritativeness to emerge. But there are problems. For one, as authentic laws may fail to be regularly enforced or even fall into complete desuetude, social habits can't even provide necessary conditions. Furthermore, since even cranberry muffin designations may be correctly enacted by the right entities, full-blown positivism cannot place its faith in "minimum content" or "inclusionary" provisions either. Not only can genuine (i.e., authoritative) laws be completely ignored by both citizens and relevant officials, they can be utterly stupid, pointless, or worse. Thus laws can't depend solely on either societal habits or ethical characteristics for the sort of propriety natural law theorists have sought.
But natural lawyers and other anti-positivists seem to have been correct when they have complained that the conjunction of the required formal properties and appropriate causal background is not sufficient for the provision of legitimacy. Where I differ from those critics is that I believe that it is not morality that is needed anywhere to fill that gap, but democracy, which relies on prudential, not moral values.
Being no expert either in comparative law or history, I must admit to not being entirely sure whether a single legitimate law has ever been created anywhere. I am fairly confident, however, of the non-existence of any legitimate Federal law in the history of the U.S., simply because of the many defects in what has been so generously called "democracy" since this nation's birth. But if this conclusion does not entail either that we have no legal system in America or that no ostensible Federal laws need be obeyed, we are left with a difficult parting question. Which ones should we follow and which ones may--or even should be--ignored? Given a prudential-value-based view like mine, are there any authoritative laws that we ought not--or at least need not-- follow? Those are extremely tough nuts to crack, and I shall postpone any attempt to shell them for another day.
*Thus, an acceptable theory of laws, like the Kripke/Putnam theories of proper names and natural kinds, will be essentially causal.
**For a more detailed look at my attempt to provide a plausible background theory for prudential values (based on an idea put forward by mid-20th Century value theorist, Everett Hall), see my book Democratic Theory Naturalized.
***I have tried to summarize my thoughts on the essential features of authentic self-government here, but there is, again, considerably more detail in my book.
