© Springer International Publishing Switzerland 2015Michał Araszkiewicz, Paweł Banaś, Tomasz Gizbert-Studnicki and Krzysztof Płeszka (eds.)Problems of Normativity, Rules and Rule-FollowingLaw and Philosophy Library11110.1007/978-3-319-09375-8_10
10. Rules and Normativity in Law
University of Minnesota, Minneapolis, USA
Brian H. Bix
Two of the persistent questions in the philosophy of law concern the relationship between law and rules and the relationship between law and morality. Both topics are most sharply raised through the topic of “legal normativity.” Many contemporary legal theorists purport to “explain legal normativity,” but often fail to articulate what it means to say that law is normative or in what way that property requires explanation. As a way of resolving some problems of legal normativity, this article offers a reading of Hans Kelsen’s legal theory as a limited claim about the logic of normative claims: that when one reads the actions of legal officials normatively, this assumes or presupposes the validity of the foundational norm of that legal system, a Kelsenian “Basic Norm.” This article also looks at a different aspect of legal normativity through a focus on the work of H. L. A. Hart. Legal norms frequently prescribe what one ought to do or ought not to do. However, the rush of legal theorists to describe law as thus making moral claims seems ungrounded and unnecessary.
KeywordsNormativityRulesKelsenHartLaw and morality
The connection between law, rules, and morality, remains a central topic of modern legal theory. In this article, I will revisit these debates, and suggest some corrections and some challenges. Central to the work of many important legal theorists is the idea that law is a normative system, and that any theory about the nature of law must focus on its normativity. There are familiar questions connected with explaining legal normativity: e.g., What is the connection between legal normativity and other normative systems, in particular, morality? And there are methodological questions: when theorists claim that we need to (and that they will) “explain the normativity of law,” what is it that is being explained? In the course of looking at issues relating to the normativity of law, many of the major figures of modern legal theory (primarily English-language legal theory, to be sure) will be revisited and perhaps reconsidered, including Hans Kelsen, H. L. A. Hart, John Finnis, and Joseph Raz. Part I gives a necessarily brief overview of the relationship between law and rules, showing the tensions and connections that give rise to many of the debates in contemporary legal philosophy. Part II offers a view regarding the nature of law and legal normativity that I believe could reasonably be read off of some of Kelsen’s work. It is a slightly unconventional reading of Kelsen, but one that is grounded clearly in some of Kelsen’s work and has the support of some Kelsenian scholars. And while I recognize that a more careful analysis of Kelsen’s texts1 and their historical context might undermine the proposed reading on exegetical grounds as the best overall understanding of Kelsen’s views, I will argue that my reading has the benefit of being more defensible than alternative readings—I will defend it as right on the merits, even if I am willing to defer to others on whether it is the best understanding historically of that theorist’s perspective. That is, my discussion of the view will be offered on its merits as a legal theory, whatever its merits as an exegesis of Kelsen. In summary, the basic argument will be that the Basic Norm2 is presupposed when a citizen chooses to read the actions of legal officials in a normative way. Kelsen’s analysis should be understood as an investigation into the logic of normative thought, in particular legal normative thought. In this Kelsenian approach, all normative systems are structurally and logically similar, but each normative system is independent of every other system—thus, law is, in this sense, conceptually separate from morality. Part III will turn to H. L. A. Hart’s theory, analyzing the extent to which his approach views legal normativity as sui generis. This investigation will offer an opportunity to raise questions regarding what has become a consensus view in contemporary jurisprudence: that law makes moral claims. I will show how a more deflationary (and less morally-flavored) understanding of the nature of law is tenable, and may in fact work better than current conventional (morally-focused) understandings of law and its claims.
10.2 Law and Rules
Is a legal system primarily a matter of rules? For H. L. A. Hart, what was distinctive of a legal system (in its most developed form) was that it was a combination of primary and secondary rules, and that the system contained a rule of recognition, by which the officials determined which rules were and were not part of that legal system(see Hart 1958, 2012).
Lon Fuller equated law with rules, at one point defining law as “the enterprise of subjecting human conduct to the governance of rules” (Fuller 1969, p. 96; see also Fuller 1958). The strength of Fuller’s “internal morality of law” was that it presented principles (“the principles of legality”) that were simultaneously rules of thumb for the effective guidance of behavior through rules while maintaining the moral tone of one form of justice—procedural justice. More recently, John Gardner has modified the Fullerian view as being that law is not a particular functional kind, but rather a particular modal kind: that it is a certain kind of means—governance through general rules (Gardner 2012, pp. 206–207).
In recent decades, various legal theorists have challenged the equation of law with rules. Both Ronald Dworkin (see Dworkin 1977, pp. 22–45, 71–80) and Robert Alexy (see Alexy 2002b, pp. 44–110) famously argued that, along with rules, legal systems contain “principles.”3 In Dworkin’s version, legal principles contrast with rules by giving general weight towards one outcome in a dispute rather than another, rather than being conclusive (as rules are). In most difficult legal cases, there will be principles both on the plaintiff’s side and on the defendant’s side. The weight a given principle has may vary from one area of law to another (e.g., the way the importance of certainty and predictability is stronger in commercial law and property law than it is in disputes dealing with civil liberties) and from case to case.
John Gardner has reflected that Dworkin’s idea of legal principles may be an unnecessary entity created to solve a problem that does not really exist. The problem Dworkin raised is that either judges are applying pre-existing legal norms (beyond legal rules) or they are legislating contrary to important rule of law principles when they decide cases in ways that go beyond the clear application of legal rules. Gardner’s response is that there is a kind of legal reasoning, a reasoning “according to law,” that works with existing legal rules in a way that creates new legal norms, but does not require the recognition of “legal principles” as a separate sort of entity (Gardner 2012, pp. 37–42).
Robert Alexy’s distinction between rules and principles is perhaps less sharp than Dworkin’s, and also differs on some matters of detail. For Alexy, principles are more general, more the justifications for rules rather than the specific rules themselves, and perhaps best thought of as goals to be optimized to the extent possible (Alexy 2002b, pp. 45–50; see also Jakab 2010, pp. 145–147).
A number of theorists have argued that a distinction between legal rules and principles will not hold up: that in the end purported legal principles, upon closer analysis, will be seen to be either a kind of rule or a kind of (moral or policy) argument for changing legal rules or interpreting or applying them in a particular way (see e.g., Alexander 2013; Gardner 2012, pp. 37–42; Jakab 2010; Raz 1983).
The American legal realists raised a number of hard challenges for the role of rules within law: sometimes regarding the way that general rules might fail to determine the outcome of particular cases (without being supplemented by moral or policy judgments), and sometimes regarding the way that the existence of legal rules can mislead us into believing either (a) that citizens always follow what the legal rules prescribe; or (b) that judges always cite legal rules when those rules are relevant to the cases they decide; or (c) that when judges appropriately cite legal rules it always means that those rules have in fact guided the judges’ reasoning. That is, there is almost always a gap with rules between text and behavior, whether considering the behavior of judges or of citizens; this gap (especially as it pertains to judges) Karl Llewellyn (following an idea of Roscoe Pound) called the problem of “paper rules” as against “real rules” (see Llewellyn 1930, pp. 444–457; Pound 1910).
This gap is pervasive to rule-based guidance, though it is not clear what we should take from that observation in constructing our ideas about the nature of law. Perhaps like the pervasiveness of coercion in law, the slippage between legal rules on paper and the rules “in action” is an important element to understanding law, even if it is not an “essential” or “necessary” attribute of law4.
The gap between “real rules” and “paper rules” might remind us of another gap important to understanding rules and how they operate—including legal rules, but not exclusive to them. This is the gap between the reasons that justify the rule and the rule itself, such that rules, as written, and certainly as applied, are always both over-inclusive and under-inclusive relative to the reasons that justify them (see Alexander 1991; see generally Schauer 1991). This gap leads to the idea of the equitable exception to rules (suggested as long ago as Aristotle, and exemplified in the English tradition of separate Courts of Equity and (now, in many countries) separate equitable rules and equitable defenses to the strict application of legal rules) and the idea of interpreting legal texts purposively and in ways that avoid absurd outcomes (relative to the purpose of the statute or the mischief it was meant to prevent).
This gap between rules and justifying reasons is also evident to those who are directly affected by legal rules. When people are governed by rules (to paraphrase Fuller’s definition of law), there will always be circumstances where even good citizens and officials, acting in good faith, must face the question of whether to obey the rule (as written) or not, where there are good reasons to act contrary to the rule—and these reasons often include the reasons that originally justified the rule in the first place.5 Both citizens and judges morally should try to do the right thing, but sometimes that may point in the direction of acting contrary to the rule rather than consistently with it. At the same time, there are good second-order reasons for citizens to do what the law says, and for officials to apply the law “as written,” just because the law says so; that is, there are good practical and institutional reasons for being governed by rules, and not just by reasons. We understand why there are long-term prudential reasons for why citizens and officials should not second-guess the legal rules, and substitute their judgments for those of the legislature. This is a big subject that we do not have time to explore fully here; suffice it to say that any discussion of the role of rules within law and within legal practice must pay due consideration to this problem of the gap between reasons and rules.
10.3 Kelsen and Normativity
Hans Kelsen’s jurisprudential work, through most of his long scholarly career,6 centered on the normative nature of law—that law is essentially made up of norms, and that this requires an approach to understanding law distinctively different from descriptive, empirical approaches (e.g. Kelsen 2013, p. 217). Kelsen’s approach assumes or is grounded on the view (often attributed first to David Hume, though questions remain as the best understanding of Hume’s text7) that there is a sharp division between “is” and “ought” statements, in particular, that no conclusion about what one ought to do can be derived from statements regarding what is the case. Whatever its origins, this view about not deriving “ought” conclusions from “is” statements (sometimes called “the fact-value distinction”) is generally accepted in modern philosophy8.
The importance of the Humean division between “is” and “ought” statements is the implication that for every normative conclusion (e.g. about what one ought to do), there must be at least one normative premise (e.g. about what one ought to do or what one ought to value). In the context of a normative system like law (or morality or religion), every statement of what one ought to do (or ought not to do) requires justification from a more general or more basic ought statement, leading through the normative hierarchy9 until one reaches a foundational normative premise10. Thus, the rules in a religious system that one ought not pray to idols will be grounded ultimately in the norm, “do whatever the creator God tells you to do”; one’s secular ethical rule of thumb not to lie unless there is a very good reason may be grounded ultimately on either the Kantian norm, “so act that the maxim of your will can be a universal law,” or the Utilitarian norm, “maximize the greatest good of the greatest number”; and the legal norm not to drive more than 65 miles per hour on a specified highway may be grounded ultimately on the norm, “act according to what has been authorized by the historically first constitution.” This foundational norm for legal normative systems Kelsen called “the Basic Norm” (“Grundnorm”)11.
This view regarding the separation of “is” and “ought” statements, and the hierarchical structure of normative systems, leading to ultimate norms, could lead to somewhat skeptical conclusions regarding morality (and religion and law). The reason is that under this approach every normative system is shown to be necessarily grounded on a foundational norm that is itself subject to no (direct) proof. One simply accepts (or not) the ultimate norm, whether it be “do what the Creator God commands” or “maximize the greatest good of the greatest number” or “act in accordance with the historically first constitution.” And the fact that the important normative systems of one’s life, like morality, religion, and law, may be grounded on an ultimate norm that cannot be proven, and can be accepted or rejected with seemingly equal legitimacy, seems to invite skeptical or relativistic implications. However, these implications must be left to others to discuss, or for other occasions12.
In Kelsen’s understanding of “the science” of norms13, every “ought” claim—whether legal, moral, religious, or of any other kind—implies the (presupposition of the) foundational norm of that normative system. And the corollary is that every normative system is self-contained and logically independent of every other normative system. The normative system that is law, with its foundational norm, is necessarily separate from the normative system of a particular religion or a particular moral system. However, it is important to note: this would not exclude lawmakers from in fact being influenced by the content of another normative system—e.g. morality or religion. One must distinguish the logical structure of (all) normative systems from the empirical/historical/causal claims regarding why certain lawmakers promulgated the legal norms they did.
10.3.2 Presupposing the Basic Norm
In Kelsen’s works, one can find language to the effect that the presupposition of the Basic Norm is required to make “possible the interpretation of the subjective sense of [certain material facts] as their objective sense, that is, as objectively valid norms….”14 At the same time, Kelsen makes it clear, in a number of places, that one need not presuppose the Basic Norm.15 In particular, Kelsen notes that anarchists need not, and would not, perceive the actions of legal officials as anything other than “naked power” (Kelsen 1992, § 16, 36), with the legal system being for them nothing more than the “gunman situation writ large.”16
Similarly, Kelsen writes: “For the Pure Theory strongly emphasises that the statement that the subjective meaning of the law-creating act is also its objective meaning—the statement, that is, that law has objective validity—is only a possible interpretation of that act, not a necessary one” (Kelsen 2013, pp. 218–219).17 Kelsen adds: “The Pure Theory aims simply to raise to the level of consciousness what all jurists are doing (for the most part unwittingly) when, in conceptualizing their object of enquiry, they … understand the positive law as a valid system, that is, as a norm, and not merely as factual contingencies of motivation”18.
Thus, Kelsen speaks about those who perceive legal actions as norms, in some places noting, in other places simply implying, that one can also choose not to perceive such actions in a normative way. Here, by perceiving actions as normative, what is meant is that one perceives those actions as giving reasons for action. In H. L. A. Hart’s terms, it is the difference between an “internal” and “external” view of normative systems, or the difference between “accepting” or not “accepting” the system (Hart 2012, pp. 87–91; see also Morawetz 1999). At that same time, I can understand that a legal institution purports to create reasons of actions (for me and for other citizens), even if I do not accept the system, and thus do not perceive it as (“actually”) normative.
This point can be generalized across normative systems. Some look at events in our (natural, empirical world) and perceive norms: obligations (reasons) to act according to the requirements of etiquette, the dictates of a religious system, or the norms of a legal system. Other equally competent and intelligent adults can look at the same world and perceive nothing normative: etiquette systems may seem like the trivial rules of a pointless game; religious norms may seem like the superstitions of the ignorant and the self-deluded; and legal rules may seem like just one more way by which the powerful control and oppress the less powerful. And, of course, some people may perceive in a normative way in some of these areas but not in others. One could just as easily say that some people accept certain normative systems that other people do not “accept”; these, I think, come to the same thing.
One idea that may help clarify the distinction between systems that purport to be normative (reason-giving) but one does not perceive as normative, and those one perceives as normative (reason-giving), is Joseph Raz’s idea of “detached normative statements”. Raz’s basic idea is that one can speak of what a normative rule or system requires, without necessarily endorsing or accepting that rule or system (Raz 1990, pp. 170–177). Thus, someone who is not a vegetarian can say to a vegetarian friend, “you should not eat that (because it has meat in its ingredients),” and a non-believer can say to an Orthodox Jewish friend, “you should not accept that speaking engagement (because it would require you to work on your Sabbath).” Analogously, the radical lawyer or anarchist scholar can make claims about what one ought to do if one accepted the legal system (viewed the actions of legal officials in a normative way), even if that lawyer or scholar saw the actions of legal officials only in a non-normative way, as mere acts of power.
(When one says that one can choose to view the (legal) actions of officials normatively or not, it is important to note that this does not mean that this “choice” is always or necessarily a conscious choice. The reference to “choice” indicates primarily that there is an option; one could do (or think) otherwise.)
Yet another way to get at the general point is John Gardner’s observation that law is voluntary in a way that morality is not. Gardner argues that morality’s claim upon all of us, as human beings, is “inescapable” (Gardner 2012, p. 150)19. According to Gardner, one cannot reasonably ask whether one should follow the dictates of morality20. But one can reasonably ask that question of law (Gardner 2012, pp. 160–176)21.
However, it may be that the reference to “inescapability” is too vague to be useful here. One might argue that the sanctions pervasively and importantly present in all (or almost all) legal systems (past and present)22 make law, in a sense, “inescapable”23. One might choose not to perceive the actions of legal officials as creating valid norms, but law (at least in systems that are generally efficacious) is not something that a practically reasonable person could ignore, the way that she could ignore (say) fashion, etiquette, or chess. Still, while one may be unable to “escape” or ignore the coercive power of the State, one can choose not to think of the State’s actions in a normative way.
Also, I am not sure that the Kelsenian approach (as I am interpreting it) would go even as far as declaring morality to be “inescapable,” for morality (or one’s moral system) would be, under this analysis, just one more normative system that one could choose or not choose, internalize or not internalize, assert or not assert. Certainly, we perceive around us a wide variety of (secular and religion-based) moral systems being advocated or assumed—with, for example, a broad range of variations on consequentialism, deontological ethics, and virtue ethics (and mix-and-match combinations of the three), just among the secular approaches to morality, even without noting the approaches to morality that are more theologically based.
The general view of normativity underlying the present analysis is often explained in analogy to games. For example, one might say to people playing chess that they ought not (e.g.) to move the bishop a certain way. However, those same people could decide never to play chess, in which case prescriptions about how one ought to move the bishop would have no application.24 (Of course, one might make an all-things-considered judgment as to whether it is right to play chess on a particular occasion, or whether it is wise to devote significant time to chess as a hobby, but these are very different inquiries, and, in any event, few would argue that everyone has an unconditional (moral?) obligation either to play chess or to avoid playing chess.)
However, the game analogy does not entirely work. If after we gave people advice regarding the rules and restrictions of chess, those people responded that they were not playing chess, we would accept that the chess rules and reasons did not apply to them. With etiquette, though, we would (or could) insist that the rules and reasons applied even if our listeners stated that they did not “accept” or “participate” in etiquette. (Foot 1978, p. 160)
The voluntariness of affiliation with religions is somewhat more complicated. On one hand, in many societies today the normative rules of a particular religion are not thought to be binding on those who are not members of that religious group. Of course, the way we think about religion today is far different from the way people thought about it in the past. As Jacques Barzun points out, “in earlier times people rarely thought of themselves as ‘having’ or ‘belonging to’ a religion.…Everybody ‘had’ a soul, but did not ‘have a God,’ for God and all that pertained to Him was simply what is, just as today nobody has ‘a physics’; there is only one and it is automatically taken to be the transcript of reality” (Barzun 2000, p. 24). And similarly, true believers even today (especially in countries in which fundamentalist views have greater social and political influence) perceive the dictates of their religion not as something chosen, but as “the Truth,” binding on all.
Back to law: if one views legal rules and official actions as things that people may or may not view in a normative way, this understandably affects how one views Kelsen’s Basic Norm—the role it plays and how it is justified. As Stanley Paulson and others have pointed out (e.g., Paulson 1992b, 2000, 2012, 2013), it is common now to view Kelsen’s argument for the Basic Norm as a neo-Kantian version of the Kantian transcendental deduction. A transcendental argument (to simplify) goes from a conclusion of what must be true, lest the conclusion be false, or, at any rate, unsupported. Kant’s transcendental deduction (again, to simplify) went from the unity of our experience to the conclusion that certain categories of thought (e.g., time, space, substance, and causation) are projected by us onto sense data25.
For Kelsen, the relevant transcendental deduction is something along the following lines: since law is (experienced as) normative, the Basic Norm must be presupposed. The difficulty, as Paulson has pointed out (e.g. Paulson 2012, 2013), is that Transcendental Arguments depend on there being only one available explanation for the matter being examined (in Kant’s case, the unity of experience; in Kelsen’s case, the normativity of law), and that Kelsen did not come close to proving that his approach was the only available explanation26.
The approach discussed in this article does not require the full machinery of a Kantian Transcendental Deduction: it needs only belief in the basic and generally accepted Humean division of “is” and “ought,” combined with a comparably conventional idea that law is a normative system. Where one asserts the validity of any lower-level norm in a legal system27, one implicitly asserts or presupposes the validity of the foundational norm of the system.
In an earlier work (Paulson 2012), Paulson expressed concerns about the sort of reading of Kelsen’s work I am offering here28. His primary worry is that this reading leaves the Basic Norm in particular, and Kelsen’s Pure Theory of Law in general, doing little work, and not the important task that Kelsen seemed to set for himself. Kelsen’s offers the Basic Norm (and its presupposition) as the key to explaining the objective meaning of norms generally, not just for those who happen to choose to interpret official actions in a normative way.
I disagree that my proposed reading of Kelsen leaves Kelsen’s theory unimportant, and the reading has the distinct benefit of being more defensible than more ambitious readings of Kelsen’s aims29. Kelsen’s Pure Theory, as I read it, is offering important insights about the logic of norms, about what follows from the fact that someone perceives the actions of officials normativity, and it offers related insights regarding the connections (or lack thereof) between law and morality, and regarding whether (or not) one has an obligation to accept or presuppose the Basic Norm of one’s legal system.
10.4 H. L. A. Hart and the Relationship of Law and Morality
10.4.1 Hart and the Internal Point of View
H. L. A. Hart, like Kelsen, emphasized the normativity of law in his criticism of earlier legal theorists (particularly John Austin), and in the development of his own, more hermeneutic theory of law. Hart argued that Austin’s command theory did not sufficiently distinguish a community acting out of fear, the “gunman situation writ large” (Hart 1958, p. 603), from a community where the officials and at least some portion of the citizens treated the law as giving them reasons for action—what Hart called “the internal point of view.”
As part of the legal positivist separation of law and morality that he advocated, Hart is careful (a) not to claim that citizens must accept the law as giving them reasons for action (he does not even discuss the circumstances under which citizens should do so); and (b) he offers a broad and open-ended set of reasons for why citizens might accept the law as giving them reasons for action. Hart writes that a citizen “may obey it [the law] for a variety of different reasons and among them may often, though not always, be the knowledge that it will be best for him to do so” (Hart 2012, p. 114). And later: “[A]dherence to law may not be motived by it [moral obligation], but by calculations of long-term interest, or by the wish to continue a tradition or by disinterested concern for others” (Hart 2012, p. 232).
10.4.2 Hart’s Legal Normativity
The question still remains for Hart: what is the nature of this normativity of or in law? The law prescribes behavior—to act in certain ways, and to avoid acting in other ways—and also empowers citizens to use legal institutions and processes for their own purposes (through wills, contracts, incorporation, and the like). If under a Hartian analysis someone accepts the legal system as giving reasons for action, what kind of reasons are those? Is there any alternative to understanding these reasons as moral reasons?
One alternative that comes to mind is that people often obey the law for purely prudential reasons: to avoid the financial penalties, potential loss of liberty, or public humiliation that can come from being adjudicated as a law-breaker. However, Hart builds his theory of law from a critique of Austin’s command theory of law, and a key part of Hart’s critique is that for many people law is more than (that phrase again) the “gunman situation writ large”—that a perception of (legal) obligation can frequently be something different from merely feeling obliged (coerced) (Hart 2012, pp. 82–91). Hart clearly intends an understanding of legal normativity where legal reasons are something distinct from (mere) prudential reasons.