Is Legal Knowledge Practical?

Abstract

Given the distinction between knowing-that and knowing-how, one could claim that legal knowledge is eminently practical: One who knows the law enjoys some form of knowing-how, namely, how to exercise certain intellectual faculties, or how to perform such activities as interpreting legal texts or arguing a case. I present some arguments to the effect that legal knowledge is not practical, being rather propositional in nature, as knowing-that instead of knowing-how. This is not to deny, however, that such activities as interpretation and argumentation are extremely important in the legal domain. I also consider whether legal knowledge is practical in a different sense—namely, with a view to decision and action. I contend that it is not practical in this sense either, even if it is mainly used for practical purposes.

1. On Legal Knowledge

…it is an extremely painful thing to be ruled by laws that one does not know. ¹

What does it mean to have legal ‘knowledge’? Is this knowledge practical? Is it propositional? If you know the law in certain respects, what kind of knowledge do you have? The philosophical debate on the nature of knowledge is enormously vast. Here I will consider the distinction between ‘knowing that’ and ‘knowing how’ and will ask whether legal knowledge is of the former or rather the latter kind.

Take, as a simple illustration, the claim that a given person knows French criminal law. Under the knowing-that picture, this would mean that the person has propositional knowledge about it (e.g., that interpretation of criminal statutes must be strict, according to the French Penal Code [Code Pénal]). ² Under the knowing-how picture, it would mean that the person knows how to perform certain activities (such as finding French legal sources, interpreting legal texts, arguing a criminal case, etc.). Of course, one thing does not exclude the other, since someone can have both propositional and practical knowledge concerning the law, but the question is whether ‘legal knowledge’ properly means one type of knowledge and not the other.

The importance of the topic for legal theory is manifold. First, several activities—such as describing, reporting, teaching, and even criticizing the law—seem to rest upon the idea that they are correctly carried out when one has propositional knowledge about the law. However, in the wake of such philosophers as Wittgenstein, Ryle, and Brandom, one can claim that knowledge of the law is basically practical, being a form of knowing how to go about doing something. ³ Were this true, we should perhaps rethink and reform our systems of legal education. Second, the justification of legal action and decision is very different under the two pictures: Under the propositional one, both citizens and legal authorities like judges can justify their actions and decisions by giving legal reasons, and this is done in propositional terms; under the practical picture, legal justification depends on some rule-following ability or practical competence. Third, if a legal system embodies a publicity principle, according to which the law must be public in order to be knowable, one may wonder what this means under our alternative: If the supporters of the propositional account are right, the law is knowable only when it is publicly stated in propositional terms; if the supporters of the practical account are right, the publicity principle must instead be understood differently.

Setting aside exegetical purposes, the supporters of the practical account may leverage various points made in the literature. They may claim, following Andrei Marmor, that “[g]enerally, if N is a norm, then to know what N is, is to know how to go about doing something, or at least what it takes to do it, or such. Actually following a norm is typically a form of action, manifest in the way you do things.” ⁴ If we apply this to legal norms, we arrive at the conclusion that knowing them is knowing how to do something which is legally required.

Another possibility for the supporters of the practical account is to consider the dynamics of legal education. As Scott Shapiro has claimed, most entering students think the sole educational mission of American law schools is to teach them “the rules,” but they quickly discover that this is not the case, for “[a]lthough they learn some ‘black letter’ law in the first year, the bulk of class time is taken up not with reciting or explaining these rules, but with arguing over what they are. Students are taught to think like lawyers, which involves learning how to argue both sides of a case.” ⁵

Finding the law, therefore, involves more than just looking up a statutory provision in a legal code and reading the answer. Legal reasoning is not the same as legal research. Nor is it an impersonal, technical, scientific process. To know the law, one must exercise a considerable degree of judgment, which is a mental faculty ungoverned by explicitly specifiable and quantifiable rules and procedures. ⁶

We may well find similar claims in several jurisprudential works. ⁷ Such claims seem to convey the idea that legal knowledge (what is taught and learned in law schools and universities) is eminently practical: One who knows the law knows how to exercise certain intellectual faculties, or how to perform certain activities.

I present here some arguments to the effect that legal knowledge is not practical, being rather propositional in nature. I start by distinguishing three kinds of knowledge: direct, practical, and propositional (§ 2), and I relate them to legal knowledge (§§ 3-5). I then present three arguments against the practical account of legal knowledge (§ 6) and discuss a basic argument in favor of it (§ 7). I finish by considering what fits best with the principle of publicity of the law (§ 8) and by claiming that legal knowledge is properly propositional (§ 9). This does not exclude that legal knowledge, understood in the proper sense, can be used for practical purposes—that is, with a view to decision and action. Of course legal knowledge serves practical purposes: It is usually acquired with a view to normative compliance and application. The question is whether it is practical in itself.

Before going into that, let me clarify a couple of important points. First, by ‘legal knowledge’ I do not mean the knowledge that can be relevant in the legal domain (for instance, the knowledge of the relevant facts in a trial); nor do I mean the knowledge that may be extracted from the law (e.g., if the law regulates interstate commerce, the knowledge that there are commercial practices in the relevant context). I exclusively mean the knowledge of the law itself, as my starting example of French criminal law was intended to suggest. Second, I refer to some scholars from the Anglo-American context (Marmor and Shapiro, among others) and to some from the Continental context (i.e., Kelsen); but I do not claim that in the former context, legal knowledge is considered to be practical, and in the latter, propositional. A survey of all the positions and debates in those contexts is beyond the scope of this work. My purpose, rather, is theoretical.

2. Three Kinds of Knowledge

What is knowledge? A way to respond to this question is to distinguish three kinds of knowledge: (1) direct, (2) practical, and (3) propositional knowledge.

The first, which is also called ‘knowledge-of’, amounts to perceptual knowledge of objects. One may have perceptual knowledge of simple and ordinary objects, like a table, and also of quite complex ones, like a town. This is what William James called “knowledge of acquaintance,” ⁸ and Bertrand Russell after him, “knowledge by acquaintance.” ⁹

The second kind of knowledge, which is usually called ‘knowledge-how’, is the knowledge of the way or ways to perform a certain activity—or, to put it differently, the disposition to behave in a certain, appropriate way on certain conditions. ¹⁰ It consists in knowing how to do something, for instance, swimming or cycling.

The third kind of knowledge, which is called ‘knowledge-that’, consists in knowing that a certain proposition is true—for instance, knowing that Rome is in Italy. ¹¹ Conceptual content is taken to be a distinguishing feature of this last form of knowing.

Various questions can be raised concerning this threefold categorization of knowledge. The first is whether the three are species of the same genus. Take (2) and (3), whose features are different: One may ask in what way they would belong to the same genus—namely, knowledge. One might say that perhaps, even if they lack a common core, they have some ‘family resemblance’. Or, if one is in search of a common core—thin as it might be—one might say that they share the fact of being forms of information processing. This is not the point I want to discuss, however.

Another question about these three is how they relate to each other, and in particular, what their ordering is, if any. Their relations seem to be various and fluid. It is difficult to claim that one kind of knowledge is necessarily involved in the others; in the end, they seem to be relatively independent from one another. I can have perceptual knowledge of a certain object without knowing what kind of object it is—for example, a new technological device I know almost nothing about: In this case I have knowledge of kind (1) without a related knowledge of kind (3). I can also have some practical knowledge of which I am not able to give an accurate description or explanation—for instance, knowing how to orientate myself without being able to explain precisely what I do when I orientate myself: In this case I have knowledge of kind (2) without a related knowledge of kind (3). Conversely, I can have an accurate propositional knowledge of an activity I am not able to perform—for instance, open heart surgery: Then I have knowledge of kind (3) without the related knowledge of kind (2). Furthermore, I can have accurate propositional knowledge of an object with which I have no acquaintance at all—for instance a dinosaur: In such a case, I have knowledge of kind (3) without knowledge of kind (1).

However, even if these three forms of knowledge are relatively independent from one another, it may be asked whether one has primacy or takes precedence over the others. One might claim that direct knowledge has such primacy, for it is a necessary condition of the others. Without perception (the argument would go), no further knowledge could be acquired and developed. But one might also claim that practical knowledge comes first, since direct and propositional knowledge are the result of our dispositions, abilities, skills, and competences. Those who claim that propositional knowledge has primacy point out that only propositional knowledge has a genuinely conceptual character and is capable of growth and social transmission—the competence of which makes humans different from the other animals on earth.

I maintain that direct knowledge has a genetic primacy, while propositional knowledge has a final primacy, since the former is the most important in starting to acquire knowledge about the world, while the latter is the most important for the development and transmission of knowledge. However, this is not a point I want to insist on.

The issue I want to address here, from the perspective of legal philosophy, is the following: What kind of knowledge is legal knowledge? One may find several answers to this question in the literature, but, as we shall see, the real dispute is between practical and propositional knowledge.

3. Legal Knowledge as Direct?

It is difficult to see and even to imagine how legal knowledge might be direct—that is, of a perceptual kind, like knowing a physical object by acquaintance. Since the law does not consist of physical entities, but rather of normative entities, like rules and principles, it is impossible to have direct knowledge of it, at least of the same kind as the knowledge we have of a table or a room. “There is no phenomenology of law in this sense”: We cannot smell the law, nor taste, nor touch, nor see, nor hear it as we hear the voice of a friend—strictly speaking. ¹² Of course, we may have direct knowledge of legal texts and legal acts, but this is a different issue. I can perceptually detect some marks on the paper, but the law is neither the marks themselves nor the paper itself: it is rather their content.

A different sense in which one might try to claim that legal knowledge is direct is the intuitionist sense, according to which the essential properties of a thing are grasped by an act of direct intellectual apprehension. Along these lines, one might claim that knowing the law is grasping its principles—that is, experiencing a form of direct, non-mediated, knowledge. ¹³ Frankly, I do not see how one could prove a claim of this sort, especially if we admit that the law is, by and large, a contextual and historical matter. At most, a natural law theorist might assume as self-evident that the law has some basic principles capable of being grasped by intuition—that is, by a form of direct knowledge. But to assume is not to argue. And though important, such principles would only be a part of the law. It is certain that we do not have any intuitive knowledge of the many details of positive law. So, I find it quite difficult to argue that legal knowledge is direct; it is more promising to think of it in a different sense.

4. Legal Knowledge as Practical?

Let us move to the claim that legal knowledge is practical. We saw at the outset that one may formulate this claim by focusing on norm-following and law-teaching. Indeed, a claim of this sort can find some support in several works, from the field of legal interpretation to that of legal ontology.

Some authors claim that law itself is a social, interpretive practice. ¹⁴ In such a perspective, knowing the law means being engaged in a certain practice—that is, knowing how to do certain things, namely, detecting the sources of law, interpreting the legal texts, giving reasons to justify an interpretation, etc. All these things are supposedly instances of knowing-how rather than of knowing-that.

The same conclusion can be reached by focusing on legal argumentation and legal ontology. For example, Neil MacCormick’s legal ontology is built on the fact that the law ‘exists’ because it can enter our practical life through acts and arguments:

I suggest that we can quite acceptably treat legal texts (rule-texts) as factually existent entities constitutive in a certain sense of ‘the law’ as it stands at some moment of time. This is not, of course, factual existence as brute fact, but rather only as institutional fact. ‘The law’ in this sense is of practical concern to us only because it can be operationalised through acts and arguments, and it is not adequately understood as ‘law’ until we have a good knowledge of relevant types of act and of argument. ¹⁵

A similar idea seems to be shared by Dennis Patterson, when he claims that “understanding law is a matter of being the master of a technique, specifically, a technique of argument.” ¹⁶ Being the “master of a technique” sounds like having a form of practical knowledge indeed.

One may find this sort of claim in less recent works as well. To take an extremely clear example, Julius Stone elaborated on a thesis of Roscoe Pound, according to which the law includes rules, precepts other than rules, ideals, and “techniques of dealing with the sources of law and the literary forms in which they are found.” ¹⁷ Lawyers and judges are expected to have such knowledge, which is essentially practical:

The traditional techniques of a legal order … cannot be fully explained in words to the novice, any more than mere words can explain to a novice how to ride a bicycle. They are learned by operating or watching others operating, perhaps with ancillary verbal instructions; they are a part of “the law” which is transmitted by using it. ¹⁸

Consider the example on which the analogy is built: riding a bicycle. This is a typical case of practical knowledge that cannot be fully put into words for the novice: how to ride a bicycle. Bicycle riding is learned by practice, not by being told the rules or the propositions about it. Analogously, one learns such specific legal techniques only by practicing them.

Of course, this is not to say that all legal knowledge is practical and barely translatable into propositional terms. But, if one agrees with the premise that ‘black letter’ law is just a minimal part of the law, one is committed to the conclusion that knowing the law is for the most part a practical business.

One way to accommodate the problem is to distinguish between (A) the idea that the mastery of some technique is a necessary condition of legal knowledge, and (B) the idea that such mastery is legal knowledge. As is apparent, (A) is much more cautious than (B) and, in principle, even the supporters of the propositional account, to which we turn next, could accept this—provided it is taken as an instrumental claim. The claim would be that such practical mastery is a necessary means, with propositional legal knowledge as an end. This would accommodate our problem, but it would obfuscate some important points, as I hope to show next.

5. Legal Knowledge as Propositional?

We finally come to the claim that legal knowledge is propositional. This is, in my understanding, the standard view of twentieth century jurisprudence. It is Hans Kelsen’s view, for instance:

If jurisprudence is to present law as a system of valid norms, the propositions by which it describes its object must be “ought” propositions, statements in which an “ought,” not an “is,” is expressed. ¹⁹

Jurisprudence (in the sense of Kelsenian legal science) describes the law and transmits the knowledge of it if it provides a set of propositions expressing its normative content. When such propositions are true, the description is correct. This is not formally different from describing a natural phenomenon, save the fact that the content of jurisprudence is normative, while the content of natural science is not.

The difference between natural science and jurisprudence lies not in the logical structure of the propositions describing the object, but rather in the object itself, and hence in the meaning of the description. Natural science describes its object—nature—in is-propositions; jurisprudence describes its object—law—in ought-propositions. ²⁰

The logical structure of such propositions is that of a conditional, attaching a specific consequence to a specific condition; but the difference from natural science lies in the normative connection between condition and consequence—for instance, an offense and a sanction:

[T]he science of law describes its object by propositions in which the delict is connected with the sanction by the copula “ought.” ²¹

Being a normative science, legal science is a norm-describing science. Now, is it possible to give a description of something in non-propositional terms? I think it is not, a fortiori when we deal with non-perceptible entities like norms. If this is correct, the description of legal norms cannot be but propositional in nature. ²²

How is such propositional knowledge acquired? Not only by going through such legal authoritative texts as constitutions, codes, and statutes, but also by consulting interpretive texts—such as treatises, textbooks, and the like—and by reading interpretive and applicative texts like judicial opinions, administrative proceedings, etc., insofar as legal theory, interpretation, and application have a cognitive dimension and do not reduce to stipulation or discretionary decision. ²³ This calls for a distinction between cognitivist and non-cognitivist approaches to legal interpretation in particular, though I cannot expand on this here: If one is a cognitivist regarding interpretation, it is because one thinks that in some texts at least there is meaning to be discovered, and interpretation will bring about the propositional knowledge of that meaning; if one is a non-cognitivist instead, it is because one thinks that interpretation is rather attribution or ascription of meaning, not its discovery, and therefore not a matter of knowledge. However, in both cases there can be propositional knowledge of someone else’s interpretation of the law, which becomes second-order knowledge in the cognitivist scenario and knowledge of someone else’s attribution of meaning in the non-cognitivist scenario. By reading interpretive texts, in any case, one acquires propositional knowledge about the law.

This is an entrenched view in Continental legal theory, although it is widely admitted that legal doctrine is both knowledge of the law and a source of it. ²⁴ The propositional view is also defended in the Anglo-American world: Ron Allen and Mike Pardo, for instance, claim that the answers to legal questions are given by statements which express propositions with truth values. ²⁵

Now, if direct knowledge is not a good candidate and there are no further forms of knowledge, our question faces the following alternative: Legal knowledge is either practical or propositional. The idea that it is a form of knowing-how is philosophically attractive, and—if we consider the claims in favor of it—convincing to some extent. The idea that it is instead a form of knowing-that is more traditional and probably less exciting; is it also less convincing? In the following section I examine three arguments that provide serious objections to the claim that legal knowledge is practical in the sense of being a sort of ability or a set of abilities.

6. Three Arguments Against Legal Knowledge as Practical

6.1. The Category Mistake Argument

The supposedly practical knowledge of the law is knowledge of the ways to perform certain legal activities, not of the law itself. It consists in knowing how to persuade a jury, how to interpret a text, how to adjudicate a case, how to justify a decision, etc. These are, of course, extremely important activities in the legal domain. But they do not amount to knowing the law, strictly speaking. Claiming that knowing how to perform such activities is knowing the law is committing a category mistake.

A supporter of the practical nature of legal knowledge might reply that (i) there is no legal knowledge without interpretation of legal texts, and (ii) interpretation is an activity irreducible to, and prior to, propositional knowledge. (A similar reply is possible in terms of legal argumentation.) This reply can be refuted in two ways. First, as a counterexample, one can have a detailed knowledge of Roman law without having any specific ability related to Roman law. One need not engage in specific Roman activities in order to acquire knowledge about Roman law. Analogously, one can compare legal institutions of different times and places without engaging in the different practical activities related to them. Second, and more interestingly, one can take some techniques as instrumental to the end of propositional legal knowledge. To be clearer about this, ‘interpretation’ as an activity can bring about propositional knowledge of the law (unless the non-cognitivist is right about legal interpretation), and ‘interpretation’ as a result of that activity can be the meaning one learns or learns about. If so, the tension between interpretation being instrumental and being cognitive is overcome by resolving that ambiguity of ‘interpretation’. This would resonate with the way of accommodating the problem mentioned in section 4, and it would make sense of the idea of action as a source of knowledge (i.e., ‘learning by doing,’ wherein performing an activity not only makes one more proficient, but also generates propositional knowledge).

6.2. The Specificity Argument

Someone who knows how to swim, knows it without regard to location or context. Potentially, that person can swim everywhere, in every pool or river or sea—at least on earth. But someone who knows the law (as an ability) does not know it without regard to context. That person does not know the law of every possible legal system. Is legal knowledge propositional then?

The supporter of the practical nature of legal knowledge might reply that legal abilities are a specific kind of practical knowledge, related to context. Knowing how to swim has no contextual component, and those who know how can swim in every possible place with water enough, while those who know the law know it in a contextual way and cannot practice it in every possible place. If I may draw a lay analogy, knowing the law would be in this sense similar to cooking: Both are contextual activities, tied to specific traditions and cultures, as is the ability to distinguish two dialects; conversely, swimming does not depend on culture. Moreover, the supporter of the practical nature of legal knowledge might add that the contextual character of legal knowledge depends on a set of institutional features and constraints structuring the specific legal practices. Therefore, legal knowledge would be a form of contextually- and institutionally-constrained practical knowledge. Surely it is not a universal affair.

6.3. The Justification Argument

Since a specific trait of practical knowledge is the impossibility to translate it into propositional knowledge (or at least the difficulty of doing so), the practical knowledge supporters are committed to the claim that interpretive and argumentative abilities are not translatable into propositional knowledge (or at least that they are hardly translatable into such terms). ²⁶ They are also committed to the claim that asking a propositional justification of what cannot be put into propositional terms is fairly close to nonsense. But most of the contemporary legal systems require judges to expose the reasons of their decisions, i.e., to give a propositional justification of them. ²⁷ Insofar as such justification is propositional and concerns the legal rules or principles applied to the case at hand, the knowledge of such rules or principles is propositional as well.

It appears the supporters of practical knowledge are ready to grant that judicial decisions must be justified according to many contemporary legal systems; they are also ready to recognize that justifying them requires referring to legal rules or principles. Nevertheless, perhaps they will stress that deciding is an activity, and justifying it is making it explicit—i.e., saying what the decision criteria are, what the interpretive canons are, what the relevant arguments are, etc. ²⁸ In this way, the relevant practical knowledge (how to decide a case) is put into propositional terms for the sake of justification. But even if it is finally put into such terms, as many legal systems require, for the supporters of practical knowledge—according to my reconstruction—it is basically a matter of practical competences and abilities, resulting in that sophisticated form of activity which we call ‘deciding a case’ and ‘justifying a decision’.

Now, if the last argument is correct, we need to notice an important point: In the end, the specificity of practical knowledge is missing from that practical account. What is specific to practical knowledge is that it is hardly translatable into propositional terms. If I teach someone how to cycle, I show that person how to do it; I do not present a list of explicit rules for doing it. It may be the case that such rules are implicit in my knowledge, but the point is exactly this: They are implicit, and it is hard or even impossible to make them explicit. Even if some of them can be made explicit, what is important in learning how to cycle is not learning such rules explicitly and repeating them by heart, but developing the ability to perform the relevant activity. Conversely, in the justification of legal decisions, what is important is not the ability to perform the relevant activity, but the reasons on which decisions are made and the fact that they are made explicit.

So, practical knowledge is relevant in the legal domain when it results in (convincing) arguments and claims from the justification point of view, with respect to the propositional content of the performed speech act, not in relation to the act itself.

Something which is connected to the argument from justification is the argument I call ‘from reflexivity’. According to many writers in epistemology, knowing that p implies knowing that one knows that p. Some think that our ascriptions of knowledge do not require the satisfaction of such a demanding condition. ²⁹ In any case, insofar as it is correct to require it, it is a requirement that only applies to propositional knowledge. Knowing-that is reflexive, knowing-how is not. It makes no sense to require knowing how one knows how do to something. By contrast, it seems a legitimate requirement of propositional knowledge, especially when some are also required to give, as judges are, a fully propositional account of what they decided and why. The same seems to be true of legal doctrine or jurisprudence. As Gray put it: “Jurisprudence is the science of Law, the statement and systematic arrangement of the rules followed by the courts and of the principles involved in those rules.” ³⁰ What knowledge is this? Definitely propositional and presumably reflexive, for in order to ‘state’ and ‘systematically arrange’ such rules and principles, one must also know what one is doing and why. (Those who think that Gray’s account is disputable because it refers to the rules followed by the courts instead of referring to the rules enacted by the lawmakers, can change that part of the definition and draw the conclusion as before: The statement and systematic arrangement of the rules enacted by the lawmakers is propositional and reflexive.)

Now, what is the conclusion to be drawn from these three (or four) critical arguments? Perhaps the supporters of legal knowledge as practical may resist the first and the second, claiming that knowing the law requires in any case some contextual ability; but resisting the third is more difficult, since, granted that some basic ability is necessary to perform any legal activity, what is specific to legal practice is the requirement of making it explicit—that is, exposing in propositional terms why a certain claim or a given decision was made, according to what legal criteria, on what rules or principles, and so on. Giving reasons of this kind is stating the knowable norms upon which the activity of decision-making was based. Here, legal knowledge is not practical in the strict sense of hardly translatable into propositional terms. On the contrary, it is essentially propositional, and, notwithstanding the different roles of judges and legal scholars, it is not different from the knowledge displayed in the works of jurisprudence according to Kelsen, among others.

7. An Argument Against Legal Knowledge as Propositional

There is, however, a very basic argument on which the supporter of legal knowledge as practical may rely. It concerns the practice-based nature of language in general and of legal language in particular. Marmor has presented a position of this sort, relying on the idea of meaning as use:

A very important and fundamental aspect of language, as Wittgenstein famously argued, does not consist in knowledge that such and such is the case, but in knowledge how; that is, when we use language correctly, we know how to go about doing something. The use of language is a move within a social practice. Furthermore, the idea of ‘use’ cannot be reduced to some form of knowledge that such and such is the case, which would not in itself involve knowledge of the meaning of the expressions of that language. ³¹

Wittgenstein on rule-following is a mandatory source here; following a rule is a practice, under this framework, and to understand a word or a rule we should look for its use. The point is whether the knowledge required for the purpose of following a rule is practical itself, and whether it is so for any kind of rule, including legal ones, or just for some of them.

In a later work, Marmor says that it is not part of his “understanding of what following a norm consists in that knowing how is categorically different from knowing that.” ³² He grants, as a plausible account, that knowing how is a species of knowing that; but he adds that, in any case, “it is a unique species and one which is probably not reducible to knowledge that.” ³³

With such precautions, Marmor contends that following a norm typically involves a form of knowing how. This is not only true of legal norms, under this picture: It is true of any norm, in particular of norms governing language use and concept formation. According to Marmor:

[I]t would be difficult to deny that a significant aspect of our language use and concept formation is governed by norms. In other words, it is plausible to maintain that some of the concepts and categories we use in a natural language manifest a know how that is, indeed, analogous to playing a game. ³⁴

Building on similar assumptions, Robert Brandom sees the understanding of a sentence as a matter of inferential role:

From the point of view of inference, understanding a sentence, associating a propositional content with it, is having practical mastery of its inferential role: knowing what follows from it, and what would be evidence for or against it. The talk of ‘knowledge’ here is very different from that involved in knowledge of truth conditions. For it is a kind of knowing how rather than knowing that: knowing how to do something, namely distinguish in practice between good inference and bad inference in which the sentence appears as a premise or conclusion, rather than knowing that the truth conditions are such-and-such. Understanding shows up on this account as a practical ability, a kind of skill: sorting possible inferences into good ones and bad ones, endorsing or being disposed to make some of them, and rejecting or being disposed not to make some others. ³⁵

So, taking inspiration from such views, one could try to build an argument to the conclusion that ‘knowing that’ is basically dependent on ‘knowing how’. I will try to put the argument in a very basic form, and will call it the ‘argument from the nature of concepts’. The argument goes as follows. Concepts are inferential roles, and mastery of concepts is mastery of inferential moves. Mastery of inferential moves is clearly an ability, and those who exhibit it manifest a form of practical knowledge (they know how to apply the relevant concepts). Now, since propositional knowledge involves concepts, such knowledge involves mastery of concepts. Therefore, propositional knowledge involves practical knowledge. To put it differently, practical knowledge is a necessary condition of propositional knowledge, while the converse is not true. This would be true of any kind of concepts, including the legal ones.

Is that a sound argument? There are reasons to doubt it. In the first place, saying that a certain kind of practical knowledge (mastery of concepts) is a necessary condition of propositional knowledge is not saying that the latter is merely—or can be reduced to—the former. In the same sense, one could say that direct knowledge is a necessary condition of both propositional and practical knowledge. But this seems quite innocuous. In the second place, the argument turns out to be about inferential mastery. What kind of mastery is this? Inference involves not just concepts, but concept application. So, it involves sentences expressing propositions where concepts are applied. So, it presumably involves some sort of propositional knowledge. Hence, the argument from the nature of concepts does not arrive at the conclusion it was intended to, but rather almost to the opposite one. If there are any inferential rules, they tell us what to do, not how to do it.

In any case, the philosophical discussion on these points is definitely open. ³⁶ For our purposes, what is important, in my opinion, is the following: (i) knowing-how and knowing-that are presumably two irreducible forms of knowledge, with specific features and problems; (ii) trying to say which comes first is probably pointless; (iii) in our cognitive development we acquire and employ both of them; (iv) in the legal domain, in the light of the preceding arguments, knowing the law amounts to propositional knowledge; (v) interpreting, arguing, and similar activities are extremely important in legal practice, but strictly speaking they do not count as legal knowledge. ³⁷ The results of the latter activities can amount to knowledge in fact but, as the category mistake argument has it, one should not take an activity for its outcome (note again the ambiguity of ‘interpretation’ denoting an activity and its outcome). One can, of course, use a broad notion of legal knowledge that includes all of such activities. But the considerations I will make in the following section suggest that we will not do that.

8. What Fits Best with the Principle of Publicity

Famously, Langdell claimed to teach law “as a science”: If law were not a science, “there could be no justification for making it an academic subject of university teaching; if it were merely a ‘species of handicraft,’ it would be a skill best acquired, not from lectures and books, but by apprenticeship, and ‘a university [would] consult its own dignity in declining to teach it.’” ³⁸ If the law is just a skill, why teach it in law schools and universities? I think that Langdell had a point in making this claim, but he was known to be a scholar with little interest in such legal activities as interpretation and argumentation. Unfortunately for scholars of that inclination, such things are tremendously important in the “life of the law” (to echo Holmes’ well-known dictum). ³⁹ Why not use a broad notion of legal knowledge that includes these activities? I suspect it would be a confusing notion in various important respects. Let us consider the matter more closely.

The rule of law requires that law be knowable, and knowable in advance. What ought to be done in order to comply with this requirement? First of all, the law has to be made public. Secret law does not comply with that requirement. “Law as a guide to conduct is reduced to the level of mere futility if it is unknown and unknowable,” as Cardozo put it. ⁴⁰ Prospective legislation, instead of retroactive, is also essential in this respect, for only prospective legislation can be known in advance. So, my suggestion is to see what fits best with the knowability requirement, whether practical knowability or propositional knowability.

Among the requirements defining the “inner morality of law,” Lon Fuller indicated the need for promulgation. ⁴¹ The rules should satisfy a requirement of publicity in order to be legal rules. They should be knowable, and knowable in advance. The rationale of the requirement is that they should be knowable if we want their addressees to comply with them. ⁴² Aharon Barak calls it the principle of publicity: “The principle of publicity requires, in addition to the publication itself, that the public be able to know, by reading the text, what is permitted and what is forbidden, to plan its activities accordingly, and to have its expectations of legality met.” ⁴³

In his theoretical understanding of law, Scott Shapiro insists on knowability as a necessary condition of legality:

If a system of norms were unknowable, then that system would not be a legal system. One important reason why the law must be knowable is that its function is to guide conduct. Legal institutions exist in order to secure compliance through the provision of rules. If its rules were in principle unknowable, then these rules could not possibly guide conduct. And if these rules could not possibly guide conduct, then they could not be legal rules. ⁴⁴

What kind of knowability is this? Suppose it is practical knowability. What should the public institutions do in order to provide citizens with it? They should provide them with certain abilities and skills, training them in arguing, interpreting, judging, etc. Is this what our public institutions do? Certainly not, or certainly not for the laymen. This is evidence of the conclusion that it is not what the principle of publicity is about. ⁴⁵ Of course, our knowledge of the law depends, inter alia, on such basic abilities as reading and understanding a text; but it does not identify with them. Knowing that the interpretation of French criminal law must be strict depends on the ability to read and understand Art. 111-4 of the French Code Pénal, inter alia; but it does not amount to this ability. One may also observe that some public institutions, such as law schools and universities, do provide their students with such skills and abilities. This is certainly true. But is this what the principle of publicity is about? It is a principle such that the public be able to know what is permitted and what is forbidden, in order to make plans accordingly. Simply training a minority in such professional and academic activities is not enough to satisfy the principle. The law’s knowability is of a different kind.

Suppose now it is propositional knowability. What should the public institutions do in order to provide citizens with it? As we said, they should enact public law according to some promulgation criterion and, to comply with the rule of law, such law must be prospective; moreover, institutions should provide citizens with the very abilities of basic education, in order to put them in a position to understand what the law requires of them, in order to use it as a guide and to plan their lives accordingly. This is what our public institutions do indeed, even if at various degrees and not always well. When the public is not able to learn the contents of the law, the principle of publicity is not satisfied. When the principle of publicity is not satisfied, an important component of the rule of law is not satisfied.

I do not claim that practical knowledge is private and propositional is public. It would be an exaggeration and would probably be wrong. But, in some sense, propositional knowledge better fits the principle of publicity and the rule of law. It does not require any specific ability on the citizens’ part, save for the basic capacities required to understand a publicly enacted text. ⁴⁶ As a specific illustration of this, consider the classic maxim Ignorantia legis non excusat. It makes sense in a propositional perspective, not in a practical one: The idea that a citizen is to be sanctioned because they were not sufficiently skilled to perform a certain activity is hard to digest.

9. Conclusion

A broad conception of legal knowledge would be confusing in some respects. It is important to keep forms of knowing distinct from one another. An ecumenical conclusion to our discussion would read as follows: Legal knowledge is practical in part and propositional in part; for one thing, knowing the law requires having some competences and basic abilities; for another, one cannot be said to know what the law is if one is not able to give a propositional account of it, and, in particular, an account of why a certain legal decision was made, a certain inference was drawn, a certain argument was accepted, and so on. Furthermore, in an ecumenical spirit, one might also reconsider the hypothesis of legal knowledge as direct, not only in the intuitionist sense favored by some natural law theorists, but also in the empirical sense cherished by the law and economics supporters and by legal sociologists as well. As a social phenomenon and a set of institutions and social rules, they may say, the law can be known with the methods of empirical social science. ⁴⁷ On this perspective, the law amounts to a set of empirical social phenomena that can be categorized in the light of our concepts according to the constitutive rules of institutional reality. ⁴⁸

So far so good. Economic analysis of law and legal sociology provide us with empirical knowledge of the law; interpretation and argumentation theory with practical knowledge about it; legal theory, jurisprudence, and philosophy with the propositional aspects of it. It is good to have a complex picture like this, where many elements interact. But remember what the rule of law requires and the principle of publicity serves: something simpler and at the same time more important from a political and moral point of view—namely, that the law be propositionally knowable.

The theoretical compromise that accommodates our starting problem (see sections 4 and 6.1) articulates some relevant differences and prevents the conceptual distinction of knowing-that and knowing-how from being taken as a dichotomy. Knowing-that requires some abilities, for one needs to know how to form propositions and how to handle them, by connecting them and drawing inferences. ⁴⁹ However, to restate the point, the arguments considered above support the view that legal knowledge is, properly speaking, propositional.

Last but not least, one may wonder whether legal knowledge can be practical in a different sense from the one we discussed. Until now we have dealt with the meaning of ‘knowledge’, somehow neglecting the meaning of ‘practical’. It was taken in the knowing-how sense. Is this the only possible meaning of it? Far from it. In a different sense, ‘practical’ means “with a view to decision and action.” ⁵⁰ In this sense, practical reflection is thinking about what (one ought) to do, and practical knowledge is knowledge about what (one ought) to do. If now we ask, “Is legal knowledge practical?”, the issue in this sense is not about our dispositions, abilities, or skills: It is about the relationship between legal knowledge on the one hand and decision and action on the other. Assume that the right answer to our question is ‘propositional’: Now the question is whether propositional legal knowledge is knowledge with a view to decision and action; whether propositional knowledge of the law is nevertheless practical in this different sense. I think that the answer is quite easy: It is certainly practical for legal participants, and officials in particular, but not essentially so for observers; it is not essentially so if you consider the fact that you can have a very detailed knowledge of legal systems different from your own, even of legal systems very distant in space and time. Suppose one is an expert in Roman law: Such knowledge will hardly be related to practical deliberations, decisions, and actions; perhaps it will animate one’s intellectual life, but for anything more than that, in making practical decisions, one will eventually refer—and defer—to their actual legal system, not the Roman one. Therefore, essentially, legal knowledge is not practical in this sense either.