Explanatory Non-Normative Legal Doctrine. Taking the Distinction between Theoretical and Practical Reason Seriously

THIS CHAPTER I will deal with two closely related questions. The first is the question whether legal doctrine is an explanatory discipline, the second whether it is a normative discipline. The dominant view holds that legal doctrine is not explanatory and that, or rather because, it is normative.

With respect to the first topic some legal theorists argue that although it is an aim of legal doctrine to systematise law, the claim that this systematisation is a kind of explanation amounts to a distortion or at least an unhelpful stretching of the notion of explanation. The nature of legal doctrinal systematisation is fundamentally different from the explanations of empirical sciences and therefore it is misleading to use the same term for both. More specifically, or so their argument goes, the systematising activity is not explanatory but rather justificatory in nature. To call legal doctrine explanatory is ‘to conceal justification behind a façade of explanation’.1 This brings us to the second claim.

The second claim many legal scholars and legal theorists make is that legal doctrine is normative in nature.2 Herewith they do not mean to make the trivial and undisputed claim that the object of legal doctrine is normative. They claim that legal doctrinal statements themselves are normative. In a nutshell, their argument is that legal doctrine is interpretative, that legal interpretation is inherently normative and that therefore legal doctrine is normative. Smith, for example, claims that also the most ‘neutral’ or ‘objective’ study of positive law presupposes a normative point of view.3 Soeteman argues that ‘legal answers, in easy cases as well as in hard cases, always presuppose a normative interpretation of the legal sources.’4

In this contribution I shall argue against both claims. I shall contend that legal doctrinal systematisation is an explanatory discipline and I shall contend that it is not a normative discipline, at least not in any ‘special’ or more profound sense than other, more in particular social and so-called practical, sciences might be called normative.


In this chapter I will argue that much of the confusion about both questions is due to the fact that legal scholars do not take the distinction between theoretical and practical reason seriously (enough). In particular, or so I will argue, some lessons about the distinction between theoretical and practical reason, more specifically about the claim that legal doctrine belongs to the realm of theoretical reason, can be (re-)learned from Hans Kelsen’s work.

Observant readers might object that since Kelsen has explicitly argued that practical reason is a self-contradictory concept, I cannot at the same time assentingly refer to Kelsen and also stress the importance of the distinction between theoretical and practical reason.5 This objection can be rebutted, however. Kelsen only states that reason on its own can never be practical, ie that reason alone can never prescribe what to do. In Kelsen’s own words:

Reason as a moral legislator is the central concept of Kant’s ethics. But for Kant this reason is practical reason: . . . it is both thought and will. . . . practical reason would be at once cognition and norm-positing will. But that is impossible. . . . by our reason, we cannot find the norms prescribing what we are to do.6

If we take the capacity to create norms to be the meaning of the term ‘practical reason’, I, as most other modern philosophers, agree with Kelsen’s rejection of it. However, the notion of practical reason is normally taken in a more relaxed sense, viz. as referring only to the use of reason in answering the most fundamental practical question ‘What ought I to do?’.7 To use reason for purely theoretical purposes is to use it only for the sake of understanding some part or aspect of reality. As the Dutch Nobel Prize winner in physics, Gerard ‘t Hooft, has stated, ‘My research does not result in a new product . . . However, it answers a fundamental desire. Human beings want to know how nature around them is structured.’8

Most legal scholars do not use reason only for the sake of understanding legal reality. They use reason in order to advise legal practitioners what they should do if they want their decisions to be in accordance with the legal system. However, the fact that the ultimate goal of legal doctrine is practical rather than theoretical does not imply that what legal scholars do to achieve this aim is itself practical, ie prescriptive, rather than theoretical, ie descriptive and explanatory. The ultimate goal of medicine, aerospace engineering and the science of public administration is to cure people, to build planes and to improve public administration respectively. Thus, they are practical sciences too. However, these sciences try to achieve this aim mainly by offering descriptions and explanations rather than justifications or prescriptions. The same holds, or so I will argue in this chapter, for legal doctrine.

Kelsen devotes a chapter of his General Theory of Norms to the analysis of the question ‘What ought I to do?’.9 In it he not only agrees that there is nothing self-contradictory about posing this question, he also fully agrees that reason can play a role in answering it. On Kelsen’s account reason can offer information both about the question whether there is a norm which prescribes what to do, ie whether there is a valid norm, and about the question what is the content of this norm? To Kelsen’s claim I would add that reason can also be helpful in answering the preliminary question ‘What norm ought I posit?’ – either as an autonomous moral being to myself or as a legislator to others. In particular, reason can offer information about how to prevent inconsistencies and incoherencies between the existing – be it personal or positive legal – system and the new norm to be created.10 Next to that reason can also offer information about the content a norm should have if one wants to realise a particular goal or set of goals.11

Thus, my stressing the importance of the distinction between theoretical and practical reason, if we understand it as the distinction between knowing what is the case and prescribing what to do, between discovering norms and creating norms, is not in conflict but rather in accord with Kelsen’s view. Although the information about the validity and the content of norms can be used for the practical purpose of deciding what to do, the information itself is offered by theoretical reason. The central thesis of this contribution is that even though legal doctrine is a practical discipline in that it intends to contribute to answering one version of the question of practical reason, it does so mainly by offering descriptions and explanations.12


A.  Why bother about the Question whether Legal Doctrinal Systematisation is Explanation?

Section III is devoted to the question in what sense we can and should say, pace Peczenik, Van Hoecke and others, that legal doctrine is an explanatory science. Before answering this question, however, we should first briefly go into the question as to why we would want to quarrel about the question whether legal doctrine is explanatory in the first place.

As I have argued elsewhere,13 our focus should not be restricted to the internal debate among legal scholars and legal theorists, but rather on the question on how we can best explicate the nature of legal doctrine to scientists of other disciplines and, more importantly, to organisations like NWO that fund scientific research.14 If we have the latter focus, the question whether and in what way legal doctrine is an explanatory discipline seems to be important. By showing that and in what sense legal doctrine is explanatory, we can not only make it clear that legal doctrine pursues the same aim as other disciplines, viz achieving, enlarging and improving knowledge of some part or aspect of reality, but also that it does so in ways which are relevantly similar to – and therefore as respectable and as subsidisable as – the ways in which other sciences do so. As was argued in section II., the mere fact that legal doctrine aims at applying this knowledge to answer the question what to do or decide, in itself does not affect the ways in which legal doctrinal knowledge is achieved.

There is no room to give a profound explication of the features which are claimed to make sciences respectable, but a few remarks might be helpful. One way to explicate the nature of scientific knowledge is to contrast it with everyday knowledge. Scientific knowledge is claimed to differ from everyday knowledge, among others, in being more precise, more profound and more systematic than everyday knowledge.15 Some scientific systematisations are explanations. Thus, when we are unwilling to call (part of) legal doctrinal systematisation explanation we seem to downplay a fundamental and important characteristic that legal doctrine shares with other sciences.

B.  Erklären and Verstehen

Some readers might argue that my problem is merely verbal and could easily be resolved if I would not insist on using the term explanation (Erklären). When I instead use the term understanding (Verstehen) the disagreement would dissolve. Since the nineteenth century, explanation is claimed to be the aim of the natural sciences, whereas understanding is the aim of the humanities and therefore also of legal doctrine.16

One reason, however, why I believe it is unwise to stick to the more generally accepted term ‘Verstehen’ is that this notion is often used in ways which (intend to) downplay the systematic or at least the generalising character of scientific knowledge.17 It is generally argued that the focus of verstehende disciplines, such as history, is on understanding concrete and sometimes even ‘once-only’ states of affairs and events in ways which typically do not allow for generalisation. A verstehende approach is thought to focus on the idiosyncratic properties of the particular state of affairs at hand.

It is true that a large part of legal doctrine is directed at understanding concrete states of affairs, viz at determining the validity, the content, or both, of specific judicial decisions. It is also true that hard cases, ie cases that are characterised by idiosyncratic features that cause disagreement about the nature of their legal consequences, play a central role in legal doctrinal analyses. However, systematisation and generalisation nevertheless are important in legal doctrinal understanding of concrete decisions, even of hard cases. This is so because both the validity and the content of judicial decisions are understood as deriving from general legal norms.18 In this respect, the legal doctrinal understanding of judicial decisions in terms of general legal norms does not seem radically different from but, on the contrary, very similar to the natural scientific understanding of concrete states of affairs in terms of general laws of nature.

C.  The Principles of Causality and Imputation

There have been some attempts to analyse legal doctrine as an explanatory discipline. In particular, there have been attempts to understand the legal doctrinal explanation of judicial decisions in terms of general legal norms as analogous to natural scientific explanation of concrete states of affairs in terms of general laws of nature.19 The locus classicus of the explication of legal doctrine as an explanatory discipline is without any doubt Hans Kelsen’s work. Therefore, I will start with an exposition of his account.20

Kelsen starts from the claim that the aim of science is to make hypothetical judgements which express a functional connection between two different states of affairs, viz between a condition and consequence. According to Kelsen there are two different types of functional connections, namely causation and imputation (Zurechnung).21 Whereas the principle of causation states, ‘If A obtains, then B is the case’, the principle of imputation says ‘If A obtains, then B ought to be the case’.22 Thus, for example, the fact that this volume of gas is heated (A) explains the fact that it expands (B). Similarly, the fact that x committed a tort towards y (A) explains the fact that x ought to pay compensation to y (B).23

Accordingly, Kelsen makes a distinction between natural sciences that explain by means of the principle of causality and normative sciences that explain by means of the principle of imputation. Legal doctrine is perhaps the most prominent of the normative sciences, but it is not unique in being guided by the principle of imputation. Ethics and theology are guided by it too.24

However, the fact that legal doctrine, ethics and theology are guided by the principle of imputation is not yet an answer to the question as to what extent they are truly scientific. Kelsen argues:

Die Rechtswissenschaft bleibt innerhalb der Grenzen der Erfahrung, so lange sie nur Normen zum Gegenstand hat, die durch mensliche Akte gesetzt sind und sie sich nicht auf Normen bezieht, die von übermenschlichen, transzendenten Instanzen ausgehen, das heisst, so lange sie jede metaphysische Spekulation ausschliesst.25

The same holds for ethics. Ethics is a science if it confines itself to the study of norms of positive morality, ie if it does not postulate or presume the objective existence of non-positive moral norms.26 Analogously, it seems that Kelsen can only allow for practical theology as a science either if the existence of God as lawgiver could be established empirically or if theology would take religious norms as acts of human will (eg of church officials).

According to Kelsen, normative scientists also stay within the boundaries of experience if the existence of a norm and its norm-giver are explicitly taken to be a fiction or a construct. Kelsen explicates the difference between fictions and constructs as follows: ‘A fiction differs from a hypothesis in that it is accompanied by – or ought to be accompanied by – the awareness that reality does not agree with it.’27 Moreover, fictions ‘are not only in contradiction with reality but self-contradictory in themselves . . . To be distinguished from them are constructs which only contradict reality as given, or deviate from it, but are not in themselves self-contradictory.’28 Thus, on Kelsen’s account, ethics and theology can even study non-positive legal, moral and religious norms, at least as long as scholars take them to be fictions or constructs.

According to Kelsen’s theory, legal doctrine, theology and ethics are normative sciences. Kelsen’s term ‘normative sciences’ is unfortunate, however. Whereas Kelsen only meant to refer to the fact that normative sciences make statements about norms, nowadays the term ‘normative sciences’ is used to express the claim that these sciences themselves make normative claims. The latter claim is explicitly rejected by Kelsen. On Kelsen’s account normative sciences are called normative sciences because of the normative character of the imputative connection these sciences study. However, for the very reason that they are sciences they can only offer descriptive statements of these connections. Due to the ambiguity of the notion, I shall not follow Kelsen in his use of the term ‘normative sciences’.

Kelsen carefully distinguishes between the objects of legal doctrine, legal norms, and the norm-descriptions that legal doctrine offers. In particular, Kelsen stresses that the fact that norms themselves lack truth value does not imply that we cannot make any true or false statements (descriptions and explanations) about them. The legal doctrinal statement that a particular norm is valid has a truth value and the same holds for the legal doctrinal statement that this norm has a particular content. To make the distinction between legal norms – that are valid or invalid – and descriptions of legal norms – that are true or false – more explicit, and probably also to stress the analogy between natural laws on the one hand and legal doctrinal statements about legal norms on the other, Kelsen proposes to use the term ‘legal norms’ for the objects of legal doctrinal investigation and the term ‘legal laws’ for the legal doctrinal descriptive statements about these norms.29

As we have seen in section II., Kelsen also explicitly argues that it is not the task, or rather not in the power, of legal doctrine to create relations of imputation. On the contrary, just as natural sciences can only discover and describe causal relations that exist in nature, legal doctrine is to discover and describe the general legal norms which are found in the particular positive legal order that the legal scholar studies.30 Just as the relation between, for example, the heating of a gas and the expansion of the gas is not created but discovered by natural scientists, legal scholars do not create but discover the relation between, for example, committing a tort and paying compensation.

In a scheme we can summarise the foregoing as follows:






States of affairs

Causal relations


Legal obligations

General legal norms


Moral obligations

General moral norms


Religious obligations

General religious norms

Level of science




Principle of Causality

Facts (that state of affair p obtains)

Natural laws

Principle of Imputation

Facts (that legal obligation q exists)

Legal laws


Facts (that moral obligation r exists)

Ethical laws


Facts (that religious obligation s exists)

Theological laws

Many legal theorists do not so much reject, but rather ignore the possibility of the Kelsenian imputative type of explanation. They seem to believe that legal scholars only have two options, viz either to offer causal explanations or to offer normative claims. Soeteman, for example, explicitly states that the question of legal scholars is not ‘How can we explain our norms?’ but rather ‘What should we do?’31

Soeteman is right in claiming that the aim of legal doctrine is not to offer causal explanations of the validity or the content of legal norms.32 However, when Soeteman argues that if the task of legal scholars is not to causally explain norms, their task must be to answer the practical question ‘What should we do?’ He then too hastily concludes that this turns legal doctrine into a non-explanatory and normative discipline. In doing so he overlooks the possibility that the task of legal doctrine is only to answer the theoretical question whether and to what extent the validity and the content of a particular legal norm can be explained by reference to other legal norms. In other words, he ignores Kelsen’s claim33 that legal doctrine can help to answer the practical question ‘What should we do?’, not by offering normative claims, but by offering descriptions and explanations.

D.  Explanation of Individual and General Legal Norms

We have seen that Kelsen argues that the aim of science is to make hypothetical judgements that express a functional connection between two different states of affairs, viz between a condition and consequence. In this section we will look in more detail at the similarities between causal explanations in the empirical sciences and imputative explanations in legal doctrine.

From the outset it should be noted that the aim of the natural sciences is not only to explain the individual fact that a particular state of affairs obtains by means of a general law that describe a general fact, but also and more importantly to explain the general fact itself. Kelsen himself has only compared the natural scientific explanation of individual states of affairs to the legal doctrinal explanation of judicial decisions. In this section I shall expand on Kelsen’s analysis and argue that not only the explanation of individual states of affairs and judicial decisions is alike, but that the same holds for the explanation of the causal and imputative connections.

In the foregoing sub-section we have seen that the individual fact that a particular volume of gas expands (the effect) is explained by reference to both the individual fact that this particular volume of gas is heated (the cause) and the general fact (described by the gas law) that ceteris paribus all gases which are heated expand.

In turn, the general fact that the causal relation itself obtains is explained by means of a theory according to which the heating of a gas is nothing but the increase of the mean kinetic energy of molecules in a gas. Thus, the explanation of the general fact that all gases expand when they are heated is that when temperature rises the mean kinetic energy of the molecules in the gases increases, which means that the molecules bump into each other faster and push each other further apart. That is why volumes of gases expand.

Analogously, legal doctrine can explain both individual and general facts. Kelsen compares judicial decisions, ie individual legal norms, to concrete empirical states of affairs and general legal norms to causal mechanisms. We have seen that both the content and the validity of the individual norm ‘x must do B’ (eg x must pay compensation to y) are explained by reference to the individual fact that state of affair A (x committed a tort) obtains and the general fact that the general norm with the content ‘In case A obtains, all must do B’ (eg all who commit a tort to someone else must compensate this person for the damage done) is valid.

Also, just as natural scientists can explain why causal relations obtain, ie why general facts are true, so legal scholars can explain why the general legal norm with that particular content obtains, ie is valid. In order to explain the validity of the general norm, legal doctrine will refer to one or more higher norms which give this particular general norm its validity.34

Van Quickenborne, however, objects to the analogy.35 For one thing he argues that whereas laws of nature have to be discovered by scientists and are a result of experimentation, general legal norms are pre-given. In other words, the job that legal scholars do in establishing legal laws which describe legal norms and in applying these laws in explaining judicial decisions is not as impressive as the job natural scientists do in establishing natural laws which describe causal relations and which explain the obtaining of individual states of affairs.

Van Quickenborne is right to claim that there is a difference here, but it is too simple to say that legal norms are pre-given. In fact, it is not true at all that legal norms are pre-given. In the case of written law, the linguistic formulation of norms is pre-given, but that does not mean that the norm, being the content of the formulation, is pre-given too. On the contrary, it is not always clear that some linguistic formulation is the expression of a valid norm and often its content is not pre-given with the formulation, but needs further scrutinising. Thus, although most norms – unlike causal mechanisms – have an explicit verbal formulation, it does not follow therefore that legal doctrine cannot make discoveries about their validity and their content.

Van Quickenborne downplays the extent to which legal doctrinal statements about general norms too can be the result of respectable scientific activities such as hypothesising, testing and falsifying.36 Moreover, he does not explicate why and to what extent these kinds of scrutinising legal regulations differ from, and are less impressive than, the scrutinising of causal relations.

We can further clarify this point if we compare legal norms to causal mechanisms that are already known to exist, ie that already have a linguistic formulation. For example, when the relation between the heating and the expanding of gases was discovered, further investigations showed that the relation does not hold for very high temperatures. The consequence has not been that natural scientists have completely rejected the gas law. They rather had to reinterpret the causal relation and therewith to reformulate the gas law. The same holds for legal norms. Sometimes the norm turns out to have another content than legal scholars thought it had. Accordingly, legal scholars have to reinterpret the norm and reformulate their description of it.37

There is, however, another and much more important difference between explanations of causal relations and those of general norms. Higher general norms, ie rules of recognition, normally do not prescribe the content of lower norms, but only the procedure by which lower general norms can be created.38 Therefore, the explanation of the content of a lower norm in terms of one or more higher norms is quite ‘thin’.39 However, this does not imply that no explanation whatsoever can be given of the content of the lower general norm. In particular, the content of the lower general norm should not conflict with and preferably both entail and be entailed by the content of other legal norms, principles and underlying legal values, at least within the same statute, within the same legal domain and to some extent within the legal order as a whole. Therefore, the content of general legal norms can at least partially be explained in terms of the content of other general norms, principles and values.

E.  Explanation or Interpretation?

Again, some readers might want to object to my use of the term ‘explanation’ for the legal doctrinal systematic analysis of the content and validity of legal norms in terms of coherence and consistency with other legal norms, principles and underlying values. In particular, some readers might want to argue that such an analysis is more aptly called ‘interpretation’. In fact, this seems to be one of Van Hoecke’s main objections to Kelsen’s view of legal doctrine.40 He argues that ‘Kelsen strongly underestimates the importance of interpretation in law and the influences of non-legal elements through such interpretation.’

Van Hoecke claims ‘The main reason for this unrealistic view is Kelsen’s theory of “meaning”, which he limits to the psychological sender-meaning, to the intention of those having issued a rule or a command.’41 To substantiate his claim Van Hoecke quotes Kelsen who states that the meaning of an act of will is the intention that some other person is to behave in a certain way.42