I. Necessary Truths in Jurisprudence and Their Critics

A cursory survey of the literature in jurisprudence shows that characterizing the central project of general jurisprudence as identifying the necessary properties of law (hence a theory of the nature of law as consisting of necessary truths) is by no means an idiosyncrasy of Raz’s theory. Jules Coleman, for example, takes the project of jurisprudence to be identifying “the essential or necessary features of our concept of law.”Footnote ⁶ Julie Dickson holds that a successful theory of law “consists of propositions about the law which (1) are necessarily true, and (2) adequately explain the nature of law” because “only necessarily true propositions about law will be capable of explaining the nature of law.”Footnote ⁷ These legal theorists all seem to share Raz’s pronouncement that “a claim to necessity is in the nature of the enterprise [of the general theory of law].”Footnote ⁸ Or, put differently (if one does not mind being a little pedantic): necessarily, a theory of the nature of law consists of necessary truths.

There are, however, multiple puzzles and difficulties surrounding this talk about the necessary features of (or truths about) law. For starters, there is something basically paradoxical about making necessary claims about law, a human institution and social practice that varies from country to country, and within each country over time. In what sense can we talk about, in law, necessity, a concept that is almost by definition insensitive to variation? What makes necessary truths about law necessary, if there are such truths? Legal theorists often hasten to leave such questions aside. Raz, after pronouncing that claims about the nature of law, if true, are necessarily true, simply “leaves the question of the kind of necessity involved unexplored.”Footnote ⁹ Dickson, after laying out her account of what a theory of law is, quoted above, notes immediately that “it would be far beyond the scope of this work to enter into a discussion of the kind of necessity which might be in play here.”

It may be open to legal theorists to say: “Fortunately, [inquiring into the nature of necessary truths in jurisprudence] is not required in order to give a presentation of the analytical jurisprudential approach.”Footnote ¹⁰ However, such a stance is hardly adequate when, in recent years, the analytical jurisprudential approach that takes identifying necessary truths about law as its central or even exclusive task has come under severe attack. Several legal theorists have argued that in doing jurisprudence, the aspiration to find necessary truths about law should be abandoned, eschewed, or at least supplemented. For example, Danny Priel suggests that jurisprudence should “abandon[] altogether the aspiration to find necessary truths about law.”^{Footnote 11} Brian Tamanaha, after examining the question whether the necessary propositions about the nature of law are a priori or a posteriori, complains:

His alternative proposal is to replace “reliance on intuitions and assumptions,” “begin with conventionalist identifications of law” and “draw on empirical studies of law to inform their analytical work.”Footnote ¹³ Lastly, Frederick Schauer has argued that an exclusive emphasis on necessary features of law, even if identifiable, may lead the jurisprudent to miss or underestimate some important features of law. For example, a pervasive feature of law (e.g., coercion) may not be necessary, but it is still a central aspect of law: “Coercion may be to law what flying is to birds: not strictly necessary but so ubiquitous that a full understanding of the phenomenon requires that we consider it.”Footnote ¹⁴

Doubts about necessary truths in law also tie into an overall critique of conceptual analysis, the traditional methodology of jurisprudence, such that the status of necessity claims in jurisprudence seems to stand and fall together with the prospect of conceptual analysis as a legitimate methodology for jurisprudence.Footnote ¹⁵ Criticisms of talking about necessity in legal theory often replicate those of conceptual analysis,Footnote ¹⁶ and the reason that the two are closely connected is evident: the necessary features of law are purportedly obtained through analyzing the (or our) concept of law.

The problem with the nature of necessary truths in jurisprudence is exacerbated by the fact that there are varieties of notions of necessity in current philosophical literature—logical necessity (as in e. g., “Necessarily, if p and q, then p”), conceptual necessity (as in e. g., “All bachelors are necessarily unmarried”), and metaphysical necessity (as in e. g., “Necessarily, water is H₂O”), just to mention three—about none of which is there consensus as to how it is to be understood.Footnote ¹⁷ Legal theorists have rarely been careful in distinguishing the kinds of necessities that might be in play in their theories.Footnote ¹⁸ Nor is it clear that the kind of necessity discussed in jurisprudence falls incontrovertibly under one of them.

What is interesting, however, is that those legal theorists who are critical of the search for necessary truths in jurisprudence, either from the point of view of a methodological critique or for the purpose of constructing an alternative theory of law, just like the theorists they are criticizing, do not usually offer reflections on the nature of the kind of necessity possibly involved either.Footnote ¹⁹ Nor do they do better than their opponents in terms of distinguishing the different kinds of necessities. This weakens the power of their critique. Their criticisms simply proceed as if there is a single, unitary, self-evident notion of necessity or necessary truth involved in jurisprudence—the only remaining question is whether such truths exist or not.Footnote ²⁰ This is far from being the case. Both Raz and Dickson, in their quotes above, indicate that it is worth investigating into the kind of necessity which might be involved in jurisprudence—only then does it make sense to “leave it aside.”

II. Moore and Wittgenstein on Doubt and Certainty

I would like now to start afresh (i.e., without assuming any traditional notion of necessity), and search for an alternative understanding of the nature and status of necessary truths in jurisprudence. I propose that several insights from Wittgenstein’s On Certainty can help illuminate our problems here. Let me therefore begin by expounding some of Wittgenstein’s guiding ideas in his posthumously published set of notes, On Certainty.Footnote ²⁷

The starting point of Wittgenstein’s inquiry is his interest in certain propositions that the philosopher G.E. Moore claims to know with unquestionable certainty, propositions such as “ I have two hands,” and “The Earth existed for a long time before my birth.”Footnote ²⁸ Moore’s purpose was to refute skepticism. He thought he had found instances of knowledge where no doubt was possible. Wittgenstein’s initial interest lies not so much in refuting skepticism as in understanding the status of these propositions. Some of their interesting features emerge, to which we will return later: we do not seem to learn these explicitly as propositions; we usually do not state these propositions in ordinary situations; everyone seems to know them and would agree with Moore that they are unquestionable; no special investigation is needed to know these things; and lastly, they seem to be empirical propositions in the sense that they refer to the world, but they also seem to be indubitable truths.

Wittgenstein does not think Moore’s response to skepticism has succeeded. He asks: what can possibly support my claim that “I know I have two hands?” We are tempted to say it is the evidence from our senses: we can look at our hands and see them. But under normal circumstances, we do not arrive at the conclusion that “I have two hands” by looking at them. More important, even if we were to do so, it is unclear, in this case, what is justifying what.Footnote ²⁹ As Georg H. von Wright puts it: if I am uncertain about the number of my hands, why should I trust my eyes?Footnote ³⁰ The grounds for the claim are not surer than that for which they are supposed to provide support.Footnote ³¹

Can I then be in doubt as to whether I have two hands? Might I have made a mistake about the existence and number of my hands? But what would a mistake here be like?Footnote ³² In most cases, being wrong about having two hands is impossible. It is important to see that it is impossible, not because the possibility of being wrong is reduced to zero here, but rather because being wrong or doubting it would not make sense. “For it is not true that a mistake merely gets more and more improbable as we pass from [statements about] the planet to my own hand. No: at some point it has ceased to be conceivable.”^{Footnote 33} “Doubt gradually loses its sense.”Footnote ³⁴ The point here is about the intelligibility of such doubts. That is to say, we would not understand the question, “do I/you have two hands?” in normal circumstances.Footnote ³⁵ Nor would we comprehend the statement, “I doubt it if I/you have two hands.” We could imagine a very special situation where such a statement or question would make sense: a patient is about to have both of her hands amputated but is not sure if the operation has happened or not after waking up on a clinical bed. In such a situation, she might wonder: do I (still) have two hands? But absent such background, “do I/you have two hands?” posed as a question demanding an answer or even a proof thereof, simply loses its foothold in intelligibility. Wittgenstein writes:

It would be getting beyond the scope of this essay to explain Wittgenstein’s idea of a “language game.” Suffice it to say that for a doubt to make sense, to be intelligible, it relies on a certain (conceptual) background, just like the request of cashing a check, to be intelligible, requires the institution of money and banking as its background.

Such (conceptual) background is often tacitly assumed. My friend can doubt whether I have two cars, but usually not whether I have two hands (unless perhaps as a joke). She can raise the question about the number of cars I own while having dinner with me. But if she sincerely raised the question “do you have two hands?” during our dinner gathering, I would not comprehend the question immediately; or, if I in fact did, we probably would have to imagine a very different situation (maybe a very different kind of dinner gathering) to understand it. Most of the time, the fact that I have two hands is part of the (conceptual) background that allows various other questions (e.g., those related to commuting) to be asked and discussed, intelligibly.

The background that our linguistic practice of doubting requires—just as a game requires, among any other things, a set of rules for it to be a game—must be undoubted. Doubting the background would make the very act of doubting unintelligible, just like a move violating the rules of a game will not be recognizable as a move in that game (not just that it is a bad move).

III. World-picture, Hinges, and the Shifting Riverbed of Thought

The foregoing paves the way for one central idea in On Certainty, the idea of a hinge. It begins with Wittgenstein’s considerations on the proposition, “The Earth existed for a long time before my birth,” another one Moore claims to know with absolute certainty. Wittgenstein points out that that Moore (or anyone else) knows this is uninteresting. However, “it is interesting that, and how, it can be known.” We do not, for example, need special investigations to conclude that.Footnote ³⁷ A geologist might investigate into the exact age of the Earth, but not whether it started to exist, for example, 5 years ago.Footnote ³⁸ However, “it is difficult to imagine why anyone should believe the contrary.”Footnote ³⁹ “Everything I have seen or heard gives me the conviction [that the Earth has existed a long time before my birth] … nothing in my picture of the world speaks in favor of the opposite”:Footnote ⁴⁰ that I know of my grandparents and ancestors, that I have read about very ancient historical events, that I have seen fossil records of ancient animals, and I have learned that fossil formation takes a significant passage of time, etc. The “everything I have seen and heard” forms a system. Wittgenstein calls this a “world-picture” (Weltbild). What is important about our world-picture is that it is “the substratum of all my enquiring and asserting”Footnote ⁴¹ and hence must not be open to doubt as a whole.Footnote ⁴² It constitutes a frame of reference,Footnote ⁴³ a general (conceptual) background against which our judgments become true or false, and knowledge becomes possible.Footnote ⁴⁴

In our world-picture, the belief or proposition that “the Earth has existed a long time before my birth” occupies a special place. “The propositions describing [a world picture] are not all equally subject to testing.”Footnote ⁴⁵ Some are more resistant to doubt: they stand fast, are centrally located, and the other parts of our world-picture, as well as our activities and practices taking place against this world- picture as their general background, turn around and point to them.Footnote ⁴⁶ Wittgenstein calls such beliefs “hinges”: “[T]he questions that we raise and our doubts depend on the fact that some propositions are exempt from doubt, are as it were like hinges on which those turn.”Footnote ⁴⁷ Because of their structural importance in a world-picture, which serves as a general background for our various linguistic and epistemic practices, doubting hinge propositions would overthrow the whole world-picture and hence render the very act of doubting unintelligible.Footnote ⁴⁸ Hinges are thus the bedrock of our beliefs. We feel certain, comfortable, and satisfied about them.Footnote ⁴⁹

World-pictures, as inherited backgrounds of people’s judgments and inquiries, will not be the same across time and culture. “The propositions describing this world-picture might be part of a kind of mythology.”Footnote ⁵⁰ For example, a king in a tribe might be brought up believing the world began with him.Footnote ⁵¹ It follows that the hinges will not be the same across time and culture either. We will be able to distinguish however, at a given time, the background beliefs and those that rest on them—although there is no absolute, eternal, or intrinsic difference between them. Wittgenstein uses a metaphor to illustrate this point:

Two things from the discussion above are worth highlighting before moving on. First, our world-picture and its hinges have a practical aspect, as they are not just the background or basis for propositional knowledge, but also our “enquiring and asserting.”Footnote ⁵⁴ They figure in our various discourses and practices, through their “enacted nature”—a point that will be elaborated below.^{Footnote 55} Second, hinge propositions, such as “the Earth existed a long time before I was born” have the form of an empirical proposition,Footnote ⁵⁶ though as the scaffolding of our thoughts, they are of an indubitable character.Footnote ⁵⁷ It follows that not all empirical propositions have the same status.Footnote ⁵⁸ The propositions Moore enumerates have such “a peculiar role in the system of our empirical propositions.”Footnote ⁵⁹ This point will bear on a crucial distinction made to clarify the notion of a posteriori necessity in jurisprudence in section VII.

IV. Wittgenstein’s Foundationalism: Hinges and Their Grounding

Let us now focus on hinges and their features—ultimately, I shall suggest that thinking about necessary truth about law in terms of “hinges” of a social practice will illuminate several problems in jurisprudential methodology surrounding necessity.

To begin, the metaphor of a hinge might prompt us to think in Quinean terms: hinge beliefs are more centrally located in the “enormous system” of our knowledge,Footnote ⁶⁰ and are unlikely to be revised in the face of recalcitrant experience.Footnote ⁶¹ However, one major difference between Wittgenstein and Quine is that for Wittgenstein, those centrally located beliefs are of a different category from those surrounding beliefs, while for Quine the centrally located beliefs and those in the outskirts form a homogeneous continuum. As we shall see below, our grasp of the hinges is based more on human instinct and tacit reactions than on our discursive abilities. In other words, hinges are grounded in our action rather than ratiocination.Footnote ⁶² It is along these lines that Avrum Stroll takes Wittgenstein to be advancing a new form of foundationalism, wherein also lies one of Wittgenstein’s major contributions to Western Philosophy:

Let us unpack the idea that the base and the mansion resting on it are different in nature. The hinge beliefs are usually not objects of knowledge.^{Footnote 64} They are “perhaps not even ever formulated.”Footnote ⁶⁵ We noted earlier that our inquiries and discourses take place against certain general background that is our world-picture. The hinges, as a part of this background, are then part of the conditions making such activities possible at all and thus do not usually appear as objects of our inquiry or knowledge.Footnote ⁶⁶ They belong to a different category.Footnote ⁶⁷ We investigate whether gravitational waves exist or not, but not whether the Earth exists or not, because that the Earth exists is something that must be presupposed for knowledge and truth about gravitational waves to be possible.Footnote ⁶⁸ The background beliefs can be formulated and brought to the foreground, but only for heuristic purposes (e.g., teaching a child,Footnote ⁶⁹ conducting a philosophical analysis, etc.). Under normal circumstances, however, no one thinks it worth pronouncing that the Earth exists, and has existed a long time before one’s birth, as pieces of information. Such claims, if formulated and expressed, thus strike us as truisms—a point that we will come back to when we discuss Raz’s central claim.

If the hinge beliefs are bedrock beliefs that in some ways ground our knowledge, then they themselves are not grounded in the same manner, if at all. It is in this context that we can best make sense of Wittgenstein’s two paradoxical remarks: “The difficulty is to realize the groundlessness of our believing,”Footnote ⁷⁰ and “At the foundation of well-founded belief lies belief that is not founded.”Footnote ⁷¹ Yet if the hinge beliefs are to serve as part of the substratum of our inquiries and discourses, how is it possible that they are themselves unfounded? To understand that, we need to address two interrelated questions about hinges: first, how do hinges and those beliefs, discourses and practices surrounding them hang together—if one is to serve as the substratum for the other? Second, what makes hinges steadfast—is there anything else lying underneath that undergirds them? Wittgenstein’s answers are, in short, as follows. First, a hinge stands fast, not because it is intrinsically obvious, certain, or convincing; it is rather held fast, in part, by what lies around it.Footnote ⁷² Second, if we could talk about the grounding of hinges at all, the bedrock beliefs are not founded on any further beliefs, but rather on our ways of acting.Footnote ⁷³ These points are worth dwelling on.

V. Hinges and Necessities: Some Preliminary Reflections on New Possibilities

Let me end my discussion of On Certainty by highlighting some general philosophical points, as a way of transitioning to the alternative account of jurisprudential necessity below.

Historically, philosophers have tried to mark off a privileged category of propositions—the necessary and indubitable truths—that can perhaps serve as the paradigm or pinnacle (depending on the aims of their projects) of human thought and knowledge. Carving a cleavage between the empirical and the logical, or between the factual and the linguistic, has been one of the dominant approaches. The analytic-synthetic distinction is one such attempt, as exemplified in the works of the logical positivists.Footnote ⁹⁰ It should be obvious that necessity claims regarding a social practice such as law are not logical necessities, as exemplified by “necessarily, if p and q, then p,” nor are they conceptual necessities or analytic truths relying on the meaning of the terms involved, as exemplified by “necessarily, bachelors are unmarried men.”Footnote ⁹¹ Both of the two notions of necessity are taken to be a priori. A third approach has been to venture into the domain of metaphysics. However, as we saw above, the notion of metaphysical necessity is itself highly contested: philosophers disagree about whether there is a coherent notion of metaphysical necessity, and if there is, how it relates to, e. g., logical necessity.Footnote ⁹² Further, metaphysically necessary truths are supposed to be true in virtue of the nature of the objects they deal with.Footnote ⁹³ But the different ontological statuses of varieties of entities (e.g., between social practices and natural kinds) also make it problematic to speak of metaphysically necessary truths across the board.

We are led to wonder if these standard approaches and traditional categories exhaust all interesting and viable possibilities for our philosophical reflections on necessity. I suggest that the unique features of hinge propositions discussed above can help illuminate necessity, especially when it comes to human institutions and social practices that are constituted by and contingent upon human actions and attitudes of a particular time and community. The necessary truths about a particular social practice or inquiry could be understood as its hinges. I shall argue for this position below in more detail. But if the suggestion is plausible, some distinctive aspects of this alternative understanding of necessary truths are worth highlighting.

Such an understanding will break away from the traditional approaches, first in that the traditional distinctions employed to mark off necessary truths from contingent ones, such as the one between the logical and the empirical, or the one between a priori and a posteriori, will be blurred or redrawn.^{Footnote 94} Although removing the “experiential” components of a statement has been a typical move in philosophers’ attempts to attain necessary truths, we will see that there are empirical propositions—or, in Wittgenstein’s terms, “propositions of the form of empirical propositions”—that have a necessary character. They, together with the practice as a whole, are connected with our experience and grounded in our ways of acting.Footnote ⁹⁵

Another way this alternative understanding of necessity may break away from the tradition is as follows. While the traditional approaches often focus on certain characteristics of individual propositions—such as being true in virtue of meaning, being rules of thought, language, or logic, being of the nature or identity of an entity—in determining whether they have a necessary character, the discussion above suggests that necessity might also be a matter of the role a proposition plays in a system or network of thoughts, propositions, actions, and the relationship it has with them. Therefore, to determine whether a proposition is necessary, we need to examine a system of thoughts, propositions and actions within an inquiry or practice, and be sensitive to the contingent relations amongst them.Footnote ⁹⁶

One caveat is in order before proceeding to my positive account. Legal theorists have made numerous necessity claims when theorizing about law. It is not my intention to claim that the analysis below is applicable to each and every necessity claim found in the literature and solves every puzzle about them. What I do hope to show is that it explains some salient and interesting cases and opens up new possibilities for understanding necessity in jurisprudence.

VI. Hinge Propositions and the Nature of Necessity Claims in Raz’s Theory of Law

I shall now examine, as a paradigm case of necessary truths in jurisprudence, Joseph Raz’s claim that law necessarily claims legitimate authority. The claim is central to an argument for a particular variation of legal positivism.Footnote ⁹⁷ I argue that this claim could be understood as a hinge of our legal practice as well as our understanding of it, a presupposition that must be in place for law’s intelligible operations.

We begin with noting an interesting correspondence, coincidental or not, between Raz’s argument and Wittgenstein’s discussion of a hinge. The beginning of Raz’s argument goes as follows: “I will assume that necessarily law, every legal system which is in force anywhere, has de facto authority. That entails that the law either claims that it possesses legitimate authority or is held to possess it.”Footnote ⁹⁸ What is striking here is that Raz could ever assume such a necessary truth about law. Indeed, commentators notice that Raz does not seem to set the ground for this claim with prior arguments, nor does he support it with any independent, further argument.Footnote ⁹⁹ Interestingly, a hinge in Wittgenstein seems to be posited in a similar way:

We might feel strange that a hinge proposition or a necessity claim could ever be assumed, and the feeling of strangeness may come from opposite directions: on the one hand, one might point out: how can we assume such a proposition without defending or justifying it? On the other hand, one might also say: why do we need to assume or even mention this, isn’t it evidently true that law necessarily claims authority?

Three interrelated points about hinges are worth revisiting here: (1) A hinge is assumed in the sense that for a discourse, inquiry, or practice to proceed, for the other judgments and propositions involved in it to be true or false, the hinge proposition has to be presupposed and thus exempted from doubt. The hinges are the bedrock of our thought and action,Footnote ¹⁰¹ but it is what lies around them that determines the immobility of hinges.Footnote ¹⁰² (2) Because it is “isolated from doubt,” and “simply gets assumed as a truism, never called in question, perhaps not even ever formulated”,Footnote ¹⁰³ a hinge appears to be not worth asserting. No further proof or argument would be given for it. In fact, no further argument could be given—other than perhaps showing its role and status in a particular kind of inquiry or practice. (3) A hinge is a foundational belief that is interwoven and shown in our basic ways of acting. It is therefore grounded and manifested in our actions and practical attitudes, on which a practice or discourse depends, forming a mutually supporting whole.

These points provide us with several helpful ideas for understanding the nature of Raz’s necessity claim. That law claims authority lies at the very foundation of our legal practice and discourse, as well as our understanding of them. The workings of our legal institutions and procedures (e.g., how courts and legislatures are organized and function), our claims and arguments about legal truth and validity (involving rules, norms, rights, precedents, interpretation, etc.), our reasons for action as affected and altered by the law (our motivations, attitudes, obligations etc. vis-à-vis the law), as well as our understanding of all of these, turn around this claim. For example, claims of substantive criminal law, criminal punishment, and the criminal process (investigation, preparation, policing, trial) of a state all revolve around that the (criminal) law of that state claims legitimate authority. That is to say, for our various institutions, procedures, discourses in this domain of law to function, in a way that is intelligible to us (i.e., we can understand and make sense of them, to a large extent), that law claims authority must be presupposed. They form a system within which that law claims authority is a hinge. Denying this claim would amount to toppling the legal practice and discourse we currently have. And consequently, our understanding of them will be derailed.Footnote ¹⁰⁴ They would make no sense to us.Footnote ¹⁰⁵

Thus, the necessity of the proposition that law claims authority comes from the fact that it is at the center of our legal practice and discourse, like a hinge that has to be assumed for what surrounds it to operate. It is in this way a presupposition of law as a social practice, not an analytic truth based on semantic meaning, nor a metaphysical assertion about the nature of an eternal entity. The existence of our current legal practice and discourse determines and points to the fact that law claims authority, while its opposite (i.e., a legal system that does not claim authority over its subjects) “has nothing on its side.”Footnote ¹⁰⁶ The claim, as a presupposition of our legal practice, is thus being held stable and immobile by the operation of our legal system.

That law claims authority simply gets assumed. It often goes unformulated and unnoticed in the daily dealings of the legal realm. Once formulated and expressed, it appears to be a truism.Footnote ¹⁰⁷ One implication of this is that we do not think that it requires a further proof or justification. However, it is also hard to see how this claim, as a presupposition of our legal practice, could be validated via a separate discursive argument or demonstration. This is not because our intellectual or philosophical powers fall short here, but because it is not something we could argue or provide justification for. These ideas don’t apply here, as there is no way to step outside of a social practice and seek justification for or demonstration of its presuppositions. We have reached the bedrock of our thinking about law: indeed, what could be more evident than the claim that law claims authority?

Here we come to another important aspect of the necessity claim as a presupposition of legal practice: their enacted nature. Such a truistic claim is rather embodied or shown in the ways our legal institutions operate and the ways law is practiced, as well as in people’s actions and attitudes in relation to those institutions and practices. What we can do, to show the truth and reasonableness of such a claim, is to describe the relevant facts, based on our experience, that are manifestations of this claim. This is exactly what Raz does to support his claim:

Nowhere else does Raz seem to offer further argument or support for this necessity claim. This absence reveals that the claim is a presupposition. Even here, this paragraph is not intended by Raz to be a proof or argument, but rather a description of how legal officials behave, how legal discourses are conducted, and how legal institutions function, and above all, what they demand from us. This is because these facts, facts about the ways legal institutions work and their participants act, manifest that law claims authority. There isn’t much more we could offer to support the claim that law necessarily claims authority other than giving a description of the relevant facts about our legal practice: it is a presupposition of a social practice, interwoven into the actions and attitudes of those who are participants of this practice.

These practicable aspects of the foundational claim can mesh with others to form more complex legal actions. A criminal trial, for example, also proceeds on the assumption that law claims authority, as manifested in the ways described in the quote by Raz above. In the meantime, the contour of a criminal trial is determined in this way, by this foundational claim. As discussed above, for our legal practice to proceed intelligibly, that law claims authority must be presupposed; the presupposition is held stable by the practice as a whole, but the practice itself is also dependent on the hinge as embodied in those basic actions and attitudes of the people engaged in the practice. They form a mutually supporting whole, like the foundation-wall and the structure on top of it. This again brings out the idea of “practicable dependence”: hinge propositions structure our practice, and shape the fundamental actions and attitudes of those who participate in it.Footnote ¹⁰⁹

It is important to notice that the convincingness of Raz’s description, and hence the truth of the necessity claim, depends on our experience with regard to the law, as the description is about the actual workings of legal institutions and the actions of legal officials. Can necessary truths be grounded in experience at all? This question gives rise to three general issues in jurisprudential methodology: first, the possibility of a posteriori necessity in jurisprudence; second, whether such a necessity claim is parochial (and if so, is that a deficiency of a theory of law?), and finally in what sense a necessity claim can also be contingent? The next two sections will be devoted to discussing these issues.

VII. How is A Posteriori Necessity Possible in Jurisprudence?

Toward the end of the last section, we observed that the truth of Raz’s necessity claim is dependent on our experience of law. This brings up a crucial methodological question for jurisprudence: are assertions about the nature or essence of law, of the form of necessary truths about law, a priori or a posteriori? The traditional view about the epistemology of necessary truths is that they are obtained a priori, that is, their justification does not require any reference to experience. This is because necessity claims are often the results of conceptual analysis, which, according to the traditional view, is conducted a priori.Footnote ¹¹⁰ The view of necessity in jurisprudence explicated above implies that they are a posteriori, but not in the usual sense of this term. The pair of terms, ‘a priori’ and ‘a posteriori’, proves to be too crude for any nuanced discussion.Footnote ¹¹¹

VIII. Parochialism and Contingent Necessity in Jurisprudence

If, as argued above, necessity claims in jurisprudence are a posteriori, another question arises. In his 2017 book, A Realistic Theory of Law, Tamanaha writes: “An additional complication analytical jurisprudents must address is that a posteriori claims typically involve truths about the actual world, whereas necessary truths are about all possible worlds real and imaginary.”^{Footnote 132} The same problem can be raised from a different direction. Joseph Raz has argued that our concept of law, and hence a general theory of law, is parochial in the sense that it is produced by a particular culture or society at a particular time in history. He uses “our concept of law” to indicate this feature of the concept of law.Footnote ¹³³ But how is it possible that we obtain necessary truths from an analysis of a concept that is parochial?

This indeed is a problem, but only if we are still gripped by an often-unarticulated metaphysical notion of necessity, the kind of necessity that speaks of truths across all possible worlds. The alternative way of understanding necessity in legal theory outlined above should already help us loosen that grip. The concept of law, the one that modern Western society has, is a product of a specific culture in a specific time period. It therefore is imbedded in, and is part of, a world-picture our community currently has.^{Footnote 134} That our concept of law and a general theory of law are parochial follows from the fact that our world-picture is parochial.Footnote ¹³⁵ It follows that the claim in Raz’s (parochial) theory, “law necessarily claims legitimate authority,” is parochial. Overall, all necessary truths obtained through analysis of our concepts are possibly parochial. Why only possibly? This is because whether, and under what conditions, a theory of a supposed parochial concept is actually universal, in the sense of being present in all past and current societies, must be decided on a case-by-case basis. Raz supposed our concept of law is parochial. That is a safe assumption to make. However, there are important concepts that seem to be universal across cultures and history.Footnote ¹³⁶ We cannot rule out a priori that certain necessary claims following from analyses of the relevant concepts are universal. But even in that case, it only means that these necessary truths are less dependent on the specific culture or historical period that produce them, not that they are true across all possible worlds.Footnote ¹³⁷

But isn’t Raz’s claim also a necessary truth? We now know the sense in which Raz’s claim is necessary. As argued above, within our (parochial) world-picture, it serves as an indubitable presupposition and thus foundation of our legal discourses and practices, which turn around it and hold it stable. It interconnects with, and plays a structural role in, a whole body of our experience and understanding with regard to the law. The denial of such a proposition would overthrow a whole area of our thought and action. Unless we are in the rigid grip of the dichotomy of necessity as either analyticity/rules of language or metaphysical necessity of natural kinds, there is no reason to deny the possibility that a necessary truth can at the same time be parochial, that is, necessary only relative to a particular culture’s conceptual scheme. Once we adopt this view of necessity, the problem raised by Tamanaha dissolves.

Odd though this may sound, this way of understanding necessary truths entails that they are not only parochial, but also contingent, within one society. Now in what way can a necessary truth be simultaneously contingent? Just as a world-picture in the primitive stage of our society might be mythologicalFootnote ¹³⁸ and thus entail that the Earth has only existed a very short period of time, a world picture very different from that of ours could entail a different concept of law (e.g., a concept of law that does not claim authority, or does not involve coercion or punishment). That world-picture and its concept of law (is that still “law?” we might ask) will be very different from ours, and in order to understand that world-picture, we will be “brought to look at the world in a different way.”Footnote ¹³⁹

Our own inherited world-picture, or our form of life, has undergone changes.Footnote ¹⁴⁰ The concept of law and thus necessary truths about law will in turn change as a result of a society’s legal, political, and moral development over time. One interesting way of change is that a necessary proposition will cease to be necessary and drop out of use, while some other propositions that used to be contingent become necessary; and when we juxtapose them, we may observe a (paradigmatic) shift in people’s way of thinking rather than that one proposition directly contradicted or denied by the other.^{Footnote 141} Law and its theory in our own world-picture have undergone significant changes. A case in point is the change from the fundamental idea that law is the command of a sovereign, to be found in the writings of Thomas Hobbes (16th-17th century) and John Austin (18th-19th century), to the basic thought that law is a system of rules, norms, or principles, to be found in H. L. A. Hart, Hans Kelsen, and Ronald Dworkin’s writings in the contemporary world.Footnote ¹⁴² The change consists not so much in falsifying Hobbes or Austin’s necessary proposition in the tribunal of empirical evidence (as in the cases where we falsify phlogiston theory or verify the existence of gravitational waves), as rather in the fact that larger social, historical changes have led us to the point where we no longer accept a picture of the society that embodies Hobbes or Austin’s understanding of law. Hart’s theory, for example, is not merely a theoretical innovation upon Austin’s version of legal positivism, but also an expression of the ideal of “the rule of law and not of men” in modern democratic societies. Necessary truths, in this sense, are contingent and they reflect the zeitgeist of the time.Footnote ¹⁴³

The metaphor from On Certainty discussed at the end of section III can help illustrate the idea of necessity that also has the dimensions of relativity and contingency: The relation between necessary propositions and other contingent ones “altered with time, in that fluid propositions hardened, and hard ones became fluid.”Footnote ¹⁴⁴ The riverbed of thoughts may shift, but at a given time, “I distinguish between the movement of the waters on the river-bed and the shift of the bed itself; though there is not a sharp division of the one from the other.”Footnote ¹⁴⁵ Thus the necessity of a necessary proposition is relative to a particular time and place, but also to those other propositions and understandings that surround it and hold it fast. That relation might change, but we are able to distinguish what is changing from what remains stable. There is no inherent difference between hinges and what lies around them. Hinges are distinguished by the roles they play in a particular kind of discourse and practice, and the fact that they are less likely to be subject to revision. The propositions “form an enormous system. And only within this system has a particular bit the value we give it.”^{Footnote 146}

IX. Replies to Objections

Some objections must be met here. The first is a problem about understanding Raz’s claim, or necessary truths about law generally, in a Wittgensteinian fashion. If they are taken to be hinges of our legal practice and discourse as I argue, and if hinges are not objects of knowledge, often unformulated, but only manifested in our actions, it would be problematic to think of them as beliefs or propositions.Footnote ¹⁴⁷ How can they then be necessarily true propositions constituting a theory of law, as Raz and Dickson, among others, envisage? Indeed, how can they be true propositions at all?Footnote ¹⁴⁸ A two-point reply can be made here. First, it is one thing for people to be engaged in a practice, without being able to state explicitly its presuppositions; it is quite another to discursively formulate those presuppositions, as a result of reflection on that practice. Second, I do not think that Wittgenstein would deny that hinges could be formulated and verbally expressed as propositions. They often are not or need not be, but certainly can be, as he did so himself. What he would deny is that their propositional truth is established through proof, demonstration, argument, etc. Their truth is acknowledged from seeing the role they play in an inquiry or practice, and how they are manifested in the relevant actions.

The second objection concerns whether necessity claims in jurisprudence are open to doubt.Footnote ¹⁴⁹ One upshot of the argument presented above is that Raz’s view about law’s necessarily claiming to be a legitimate authority is not the sort of statements that can be doubted. But some theorists do seem to doubt it. Are they then making a category mistake? Matthew Kramer, for example, has argued against Raz’s idea that law necessarily claims legitimate authority.Footnote ¹⁵⁰ Though the focus of Kramer’s disagreement with Raz seems to be on law’s claiming to be a moral or legitimate authority (Kramer denies that judges’ statements of legal obligations are perforce moral obligations as well), as he grants after all that “law does necessarily claim authority in its workings,”Footnote ¹⁵¹ the question can still be raised generally: could necessary claims in jurisprudence, understood in the above fashion, be doubted? In the discussion above, the concession was made that hinge propositions could be sensibly doubted, though under very special circumstances (e.g., as a heuristic or a joke perhaps). But it is worth asking, what would a doubt about law’s claim to legitimate authority amount to here? The point, again, is about the intelligibility of such a doubt and of our concepts. It would amount to not merely entertaining the possible negation of such a claim, but imagining a very different world that comes with the negation of this claim. And that only means that we are conceiving very different concepts of law and authority if we were to doubt this necessity claim. In a different context, Raz shows exactly what we would have to imagine about a society if its law does not claim legitimate authority, that is, “political authorities of a country do not claim the inhabitants are bound to obey them”:

Raz points out that, “if such a society were to exist we would not regard it as being governed by authority. It is too unlike the political institutions we normally regard as authorities.” Raz takes this point to be a conceptual one.Footnote ¹⁵³ Doubting a necessity claim would thus amount to changing the concept involved in the claim. To Kramer’s argument, a similar line of response could be made: a system of “stark imperatives” never purporting to make any claim of moral worthiness,Footnote ¹⁵⁴ would not be instances of law and authority,Footnote ¹⁵⁵ or at least not law and authority in their fullest sense, because they are so distant from what we would recognize as law and authority, thus from the concepts of law and authority we have. We have our concepts of law and authority that we do “because in our world societies have governed by institutions claiming and being acknowledged to have the right to bind their subjects;” our concept of authority “is a part of our cultural tradition.”Footnote ¹⁵⁶

This point about concepts as being rooted in a cultural tradition connects to two further objections, arising from the above discussion of necessity in jurisprudence as historically contingent and as reflecting the zeitgeist of the community. Can we not say that, e. g., Austin’s proposition about the nature of law was wrong in his own time (hence had no status as a necessity claim), when we think about Hart’s famous criticism of Austin’s theory of law?Footnote ¹⁵⁷ That such a criticism is possible does not affect my thesis that Austin’s proposition is necessary relative to his time yet contingent. That laws (“properly so called”) are commands of a sovereign was necessary to sustain and make intelligible the practices of Austin’s time. Generally, we, standing where we are, are able to criticize those strongly held beliefs and propositions most fundamental to a specific time and culture, i. e., those moral beliefs, assumptions of the society, theories of sciences, etc., that were considered necessary. Slavery in ancient Greece was thought of not as an issue of justice, but as a necessity of that kind of social structure.Footnote ¹⁵⁸ We, of course, no longer accept slavery, and we can criticize and condemn slavery in an ancient Greek polis, as well as any theory trying to justify its necessity. However, that does not undermine the point that the belief that slavery is an indispensable part of society was the fundamental assumption of the time that cohered with many other political, social, and moral settings of ancient Greece, and was something around which many of these other assumptions and practices turn.Footnote ¹⁵⁹ In the (inconceivably) distant future, with radical changes in social and historical conditions (e.g., if we began living in a world as imagined in the quote from Raz in the previous paragraph), we might be able to criticize Raz’s fundamental claims about law.

My discussion of parochialism and contingency also seems to have assumed that Austin’s concept of law, as embodied in his theory of law, and the concept of law in early 19th century England (assuming there was a single, coherent concept that is different from our concept), are one and the same. Is this assumption justified? Is the assumption that a theory or a concept of law a particular theorist has often reflects or expresses the views or concept of law of her time and culture sound, absent further historical, sociological, and anthropological arguments? Might there be distances and differences between them? There might very well be. However, the assumption that a theory or a concept (of e. g., law) reflects or expresses the views (on e. g., law) of the time, at least in some important ways, has a lot of plausibility. People and their theories are products of their time in varying ways, and often are so to a greater extent than they themselves might recognize or acknowledge. In Thomas Hobbes and John Austin’s times, the idea of the Rule of Law was not as prevailing, and what they offered was a theory that fit more with the rule of men. It was a theory that emphasized the government’s use of law as a form of social control and instrument of power, a truth about law that still holds today, though widely thought to be inadequate, distorted, and require qualification and supplementation to be a theory of the nature of law. While it is true that in our own time, we have Hart, Dworkin, Kelsen, etc. offering differing views on the nature of law, we see that they systematically differ from Austin’s theory in that, rather than seeing law as a pure exercise of power, they emphasize to varying degrees, among other things, the importance of rules, principles or norms in understanding the nature of law. There has been a change of paradigm in discussing the nature of law, so to speak.Footnote ¹⁶⁰