A. The Analytical Basic Norm (Kelsen)

The problem with this answer is that it includes a transition from ‘is' to ‘ought’. The ‘is* consists in the factual issuance and social efficacy of the constitution, which can be expressed in the statement,^{1 >7}

(2) Constitution C has in fact been issued and is socially efficacious.

The ‘ought’ is the legal validity of the constitution, which can be expressed in the statement,

(3^Z) Constitution C is legally valid.

Statement (3^Z) is an ‘ought'-stalement, for it implies the state­ment,¹⁵⁸

(3) h is legally prescribed that one behave in accordance with constitution C.

But not one single normative statement ever follows logic- cdly^]Si) from an ‘is*,¹⁶⁰ more precisely, from any class whatso­ever of exclusively empirical statements. So to go from (2) to (3) or (3'), an additional premiss is needed. This additional premiss is the basic norm. It can be so formulated that cither it permits (he move from (2) to (3'}—(3) is then to be derived from (3')—or it leads directly to (3). The latter formulation will be considered here. It reads,

(1) If a constitution has in fact been issued and is socially efficacious, then it is legally prescribed that one behave in accordance with this constitution.

¹⁵⁸ Sec Kelson. PTL § 34(a) (al p. 193).

¹It should be emphasized that the issue here is logical deducibility. The expression ‘follows’ is often used, albeit incorrectly, to say that something is a good reason for something else. Empirical statements can of course serve as good reasons for normative statements. But then a normative premiss is always presupposed that turns the empirical statements into good reasons.

The thesis that no ‘ought’ follows from an ‘is’ alone can be traced back to Hume, and so it is also called ‘Hume’s Law’. Sec David Hume. A Treatise of Human Nature (1st pub. 1739-40). cd. David bate Norton and Mary J. Norton (Oxford: Oxford University Press. 2000), bk. ill, pt. I. sect. 1. para. 27 (at p. 302) (emphasis in original): ‘In every system of morality, which I have hitherto met with, I have always remark’d, that the author proceeds for some time in the ordinary way of reasoning, and establishes the being of a God, or makes observations concerning human affairs; when of a sudden I am surpriz'd to find, that instead of the usual copulations of propositions, is. and is not, I meet with no proposition that is not connected with an ought, or an ought not. This change is imperceptible: but is. how­ever, of the last consequence. For as this ought, or ought not, expresses some new relation or affirmation, ’tis necessary that it shou’d be observ'd and explain'd; and at the same time that a reason shou’d be given, for what seems altogether inconceivable, how this new relation can be a deduction from the others, which arc entirely different from it.' For a statement on the logical questions arising from the ‘is’-ought’ distinction, see Rainer Stuhlmann-Lacisz, Das Sein-Solien-Problem (Stuttgart-Bad Cannstatt: Frommann-Holzboog.

The statements (1), (2), and (3) can now be brought together in a basic norm syllogism that has the following form:

(1) If a constitution has in fact been issued and is socially efficacious, then it is legally prescribed that one behave in accordance with this constitution.

(2) Constitution C has in fact been issued and is socially efficacious.

(3) It is legally prescribed that one behave in accordance with constitution C.^[159]

Scarcely any other idea in legal theory has stirred up as much strife as the idea of a basic norm. The debate is concentrated on four points: the necessity, the possibility, the content, and the status of the basic norm.

(ii) Necessity

Hart's objection to the necessity of a basic norm is that it leads to unnecessary duplication.

If a constitution specifying the various sources of law is a living reality in the sense that the courts and officials of the system actu­ally identify the law in accordance with the criteria it provides, then the constitution is accepted and actually exists. It seems a needless reduplication to suggest that there is a further rule to the effect that the constitution (or those who ‘laid it down’) are to be obeyed.^{[160] [161]}

This is a powerful objection in that it docs not apply the basic norm to things like expressions of will, behavioural patterns, and coercive measures that, with the help of the basic norm, are interpreted as a legally valid constitution, but, rather, focuses immediately on the institutional fact of a working constitution. Accordingly, the following statement could serve as the sole premiss of a justification of the legal ‘ought’:

(2') The participants in legal system S accept and imple­ment constitution C.

The question is whether this premiss implies the conclusion of the basic norm syllogism, namely:

(3) It is legally prescribed that one behave in accordance with constitution C.

An answer in the affirmative calls for the following interpret­ation of (3):

(3") From the standpoint of a participant in legal system 5, this is true: It is legally prescribed that one behave in accordance with constitution C.

The sentence (3") follows from (2') because the fact that (he participants in a legal system accept and implement a consti­tution means that, from their standpoint, it is legally pre­scribed that one behave in accordance with this constitution. Is the basic norm rendered thereby superfluous? Is Alf Ross's rhetorical question on the mark? ‘But the norm itself, according to its immediate content, expresses what the indivi­dual ought to do. What, then, is the meaning of saying that the individuals ought to do what they ought to do!’¹⁶' The answer is ‘no, the basic norm is not rendered superfluous’. The decisive point is that one may indeed be able to move from (2') to (3") without the help of a basic norm, but (2') itself presupposes a basic norm. That the participants in a legal system accept and implement a constitution presupposes that every single one of them interprets certain facts as facts creating the constitution. That may well amount to a very complex bundle of facts. Here, the variety will be reduced to two things, namely, that the constitution has been passed by a constitutional assembly and that the other participants in the legal system accept and implement the constitution. Take now a participant in legal system S who accepts and implements constitution C. This participant is questioned as to why constitution C is legally valid, which includes the question of why it is legally prescribed that one behave in accordance with constitution C.

The participant could try to evade the question by claiming that it is senseless. He could elaborate by asserting, with Hart, that those rules of the constitution that say what valid law is (Hart terms them collectively the ‘rule of recognition’) cannot in turn be said to be legally valid themselves. They are pre­supposed as existing, and their existence is ‘a matter of fact’.^[162] A response to that, however, is that the question as to the legal validity of a constitution is both possible and routine.

(2") Constitution C was passed by the constitutional as­sembly, and the other participants in the legal system accept and implement it.

This statement is nothing other than a concretization of the second premiss in Kelsen’s basic norm syllogism, namely,

(2) Constitution C has in fact been issued and is socially efficacious.

But (2) alone implies neither the statement,

(3') Constitution C is legally valid.

nor the statement,

(3) Il is legally prescribed that one behave in accordance with constitution C.

In order to move from (2) or (2") to (3') or (3), a basic norm like (1) must be presupposed. A move from cither (2) or (2") alone to cither (3') or (3) is no more possible than a move from the statement (5) alone,

(5) Peter wants me to give him 100 marks,

to the conclusion,

(6) I am obligated to give Peter 100 marks.

What is needed in order to make the respective moves possible is, in the first case, the basic norm as introduced above, and, in the second case, a norm like

(4) I am obligated to do what Peter wants me to do.

There is no unnecessary duplication even if there is, from the issuer of a norm, not only an expression of will but, explicitly, the formulation of a norm. Assuming that Peter is speaking on his own authority—that is, without reference to law, morality, or social convention—when he declares that I am obligated to give him 100 marks, then (5') alone,

(5') Peter said to me. ‘You are obligated to give me 100 marks’,

does not imply

(6) I am obligated to give Peter 100 marks.

If it did. then words alone could establish any obligations whatsoever for any persons whatsoever. To arrive at (6) from (5'), a norm like (4') is needed:

(4') I am obligated to do whatever Peter says I am obli­gated to do.

This is duplication, but not unnecessary duplication.

The conclusion, then, is that there are two settled theses. The first thesis states that if the participant in a legal system wants to say that the constitution is legally valid or that it is legally prescribed that one behave in accordance with the constitution, then he must presuppose a basic norm. The second thesis slates that if the enquiry into legal validity is not to be cut off arbitrarily, then it must be possible for the participant to say that the constitution is legally valid or that it is legally prescribed that one behave in accordance with the constitution which presupposes a basic norm.

(iii) Possibility

Not only can an opponent of the basic norm make the claim just refuted, namely, that a basic norm is superfluous, he can also object that the validity or the existence of a basic norm is impossible. So it is that Ronald Dworkin countered Hart’s basic norm (rule of recognition) by maintaining that the law cannot be identified on the basis of a rule that focuses on authoritative issuance and social efficacy.’⁶⁵ This objection corresponds to the argument from principles set out above.^{[163] [164]} according to which the law also includes the totality of the standards that must be taken into account in order to satisfy the claim to correctness that is necessarily attached to the law. Indeed, these standards cannot be completely identified on the basis of a rule that focuses on authoritative issuance and social efficacy.

Nevertheless, the argument from principles does not dis­pose of the possibility of a basic norm. It shows simply that a basic norm that focuses only on empirically ascertainable facts (issuance, efficacy) cannot identify the law completely. What a basic norm like this can identify, however, is authori­tatively issued and socially efficacious law. The basic norm is to be interpreted, then, to the effect that authoritative issu­ance together with social efficacy is simply a sufficient condi­tion for belonging to the law. not a necessary one. So, on the basis of (he argument from principles, the following statement is not true:

(1) The law includes everything (hat has been authorita­tively issued and is socially efficacious—and nothing more.

Rather, the weaker statement is true:

(2) The law includes everything (hat has been authorita­tively issued and is socially efficacious.

As will be shown in (he next section, even this statement requires a still weaker formulation if the argument from in­justice^[165] is to be taken into account. Within the framework of statement (2), however, a basic norm is not only possible, it is also necessary, namely, in order to be able to move from empirically ascertainable facts to legal validity.

A basic norm limited to authoritatively issued and socially efficacious law has the disadvantage of not being a complete, supreme identifying criterion for the law. This role can be recovered by the basic norm, not completely, but in limited terms. To that end, clauses must be added to it that take into account the arguments from injustice and from principles. Only the argument from principles is of interest at this point. Building the results of that argument into the basic norm yields a non-positivistic basic norm with the following structure:

If a constitution has in fact been issued and is socially efficacious, then it is legally prescribed that one behave in accordance with this constitution in such a way as corres­ponds to the claim to correctness.

This formulation shows that a non-positivistic basic norm has only limited usefulness in identifying the law. The clause, ‘in such a way as corresponds to the claim to correctness*, refers to moral standards without naming them and without men­tioning any criterion whereby they can be unequivocally iden­tified. This openness is unavoidable. It is acceptable only because there are rules of legal method that preclude allowing openness to lead to arbitrariness.^{[166] [167] [168] [169] [170] In particular, these rules prevent issued and efficacious nonns from being arbitrarily- suppressed by appeal to the claim to correctness.169 They must do that, if for no other reason than that legal certainty is an essential element of legal correctness.}

(iv) Content

/Xccording to Kelsen, the basic norm is entirely neutral in terms of content.

The question is not what the content is of this constitution and the state legal system erected on its basis, whether this system is just or unjust; neither is it whether this legal system in fact guarantees relatively peaceful conditions within the community constituted by the system. In presupposing the basic norm, no value transcending the positive law is affirmed. ⁰ Thus, the content of the law can be anything whatsoever.¹⁷¹

This contradicts the position represented by the argument from injustice, according to which norms that are unjust in the extreme cannot have the character of legal norms.¹⁷² Not even this, however, is the downfall of the idea of a basic norm. One can add to the formulation of the basic norm a clause that lakes into account the argument from injustice. A formu­lation that corresponds to the argument from principles as well as to the argument from injustice runs as follows:

If a constitution has in fact been issued and is socially efficacious, then, if and in so far as the norms of this constitution are not unjust in the extreme, it is legally prescribed that one behave in accordance with this con­stitution in such a way as corresponds to the claim to correctness.

This formulation refers only to the constitution. I shall exam­ine norms issued in accordance with the constitution in the next chapter, on the definition of law.

(v) Tasks

That the basic norm has three altogether different tasks com­plicates the determination of its status.

(a) Transforming Categories

The first task of the basic norm is to make possible the transition from ‘is’ to ‘ought’. Because ‘is’ and ‘ought' are categories of altogether different kinds, this first task can be called 'transforming categories'In that certain facts arc interpreted as law-creating facts, the move into the realm of the law is accomplished.

(b) Setting Criteria

The move into the realm of the law could not be accomplished if the basic norm allowed the interpretation of any facts whatsoever—say. every expression of will that comes

¹ ' See Aulis Aarnio, Robert Alcxy, and Aleksander Peczcnik, ‘The Foundation of Legal Reasoning’, Rechtstheorie. 12 (1981), 133-58, 257-79. 423-48. al 142-6; Aleksander Peczenik. Grundlagen der juristischen Argumentation (Vienna and New York: Springer, 1983), at 23.

along—as law-creating facts. Thus, a second task is incum­bent upon the basic norm. It must establish which facts are to be regarded as law-creating facts. In so doing, it sets the criteria for what law is. This second task can be described as 'setting criteria. Kelson's criterion, as stated above, is that ‘of a constitution that has in fact been issued and is by and large efficacious’.^[171] Another variant of Kelsen’s criterion is that of the ‘historically first state constitution’.^[172] Criteria like this include a reference, that is, they say that the criteria of the constitution are criteria for what valid law is. Thus, Kelsen can formulate his basic norm such that it is, first, very simple and. second, applicable to all developed legal systems. In Hart's theory, by contrast, (he basic norm (rule of recogni­tion) is identified with the rules of the constitution that say what law is. Hart’s basic norm becomes thereby very compli­cated and is applicable only to the current legal system. The only thing general about it is that every developed legal system must have a norm like this. Both Kelsen’s and Hart's criteria are positivistic. As stated above, the argument from injustice requires a restriction of positivistic criteria, and the argument from principles requires their supplementation.

(c) Creating Unity

The third task of the basic norm is to create unity.

All norms whose validity can be traced back to one and the same basic norm form a system of norms, a normative order. The basic norm is the common source of the validity of all norms belonging to one and the same system, their common basis of validity. A certain norm belongs to a certain system in that the ultimate basis of its validity is the basic norm of this system. This basic norm is what constitutes the unity of a multiplicity of norms by representing the basis of the validity of all norms belonging to this system.^[173]

It could be seen as a problem that the basic norms of de­veloped legal systems are, in their content and their status, identical. Is it the same basic norm that creates the unity of each different legal system? How should that be possible? Or is it, after all. the constitution that leads to unity? These remain open questions here.

(vi) Status

The problem of the status of the basic norm concerns mainly its first task, transforming categories. As the norm that establishes the validity of all positive law. the basic norm cannot be in turn a norm of positive law.^[174] But what is it then? One might think that then it could only be a non­positive norm and, as a non-positive norm, it would have to be a norm of either natural law or the law of reason.^[175] '^s Kelsen emphatically rejects that possibility. But what should the basic norm be if it is neither a norm of positive law nor a norm of suprapositive law, that is, a norm of natural law or the law of reason?

That this is not an easy question to answer is apparent not only in the boundless literature, but also in Kelsen’s own doubts about the matter toward the end of his life.^[176] His most significant answer is found in the second edition of the Pure Theory of Law (1960). There, four characteristics of the basic norm define its status.

(a) Necessary Presupposition

The first characteristic is that, if one wants to speak of legal validity or a legal ‘ought’, one must necessarily presuppose the basic norm.^[177] In discussing the concept of the basic norm above, it became clear that this thesis is correct in so far as one must presuppose some basic norm if one wants to move from ascertaining that something has been issued and is efficacious to ascertaining that something is legally valid or legally ob­ligatory. The discussion of the possibility and the content of the basic norm has shown, however, that while this basic norm must indeed include elements of the Kelsenian basic norm, it has to be supplemented by non-positivistic elements.

Kelsen, because his basic norm is the necessary condition for the possibility of the cognition of legal validity and the legal ‘ought’, characterizes it, with a nod to Kantian termin­ology, as a ‘logico-transcendental presupposition' of legal cognition. This characterization is correct in so far as, according to Kant, the transcendental is that which is neces­sary in order to make cognition of experience possible.^{[178] [179]} Nevertheless, there is an important difference between the transcendental in Kelsen and that in Kant, a difference mani­fest in the second characteristic of the basic norm.

(b) Possible Presupposition

The second characteristic is that one must indeed necessarily presuppose the basic norm //one wants to interpret the law as a normative system (an ‘ought’-system), but this interpret­ation itself is only a possible interpretation.™² As sociological and psychological legal theories demonstrate, it is possible even if for many purposes not very fruitful¹⁸³ to describe and to explain the law as a merely social and/or psychical concatenation of effects.¹⁸⁴ Kelsen himself emphasizes this when he comments that an alternative to the juridical inter­pretation of the law is a sociological interpretation, according to which the law is a system of ‘power relations’.¹⁸^ One can therefore say (hat the basic norm is simply a possible or hypothetically necessary presupposition.

That is of consequence to its transcendental character.¹⁸⁶ In the area of experience, according to Kant, there are no alternatives, for example, to space and time as the forms of intuition. Cognition for Kant—that is, empirical cognition— is only possible, then, in space and time.¹⁸⁷ By contrast, cognition of legal phenomena is in principle also possible without the use of the category of ‘ought'. Nevertheless, this docs not fully dispose of the transcendental character of Kelsen’s argument. While his argument cannot demonstrate, to be sure, the unconditional necessity of the basic norm and thereby the category of ‘ought’, it can show a conditional necessity. The juridical standpoint—or the standpoint of the participant in a legal system—is defined such that the law is interpreted by the participant as a valid system of norms or a normative system (an ‘ought’-system). One can of course refuse, not only in action but also in thought, to participate in the (utterly real) game of the law. But if one gets into

^ls' The interpretation of the law as a system of mere facts is not accept­able to a legal sociologist either; see Rotlleuthner. Rechtsiheorie und Rechtssoziologie (n. 144 above), at 31-61. 91-7.

^1X4 Sec Ralf Dreier, ‘Bemerkungen zur Rcchtserkenntnistheorie’, in MEA 89-105, at 95.

^1X5 Kelsen, PTL § 34(i) (p. 218).

^1X6 Sec Stanley L. Paulson. ‘The Neo-Kantian Dimension of Kelsen’s Pure Theory of Law’. Oxford Journal of Legal Studies, 12(1992). 311 32, at 322-32.

^1x7 See Kant. Critique of Pure Reason (n. 143 above), al A24/B38 (p. 158): ‘Space is a necessary representation, a priori, which is the ground of all outer intuitions’; and at A3I/B46 (p. 162): ’Time is a necessary' representa­tion that grounds all intuitions.’

this game—and there are good reasons, at least in practice, to do so—then there is no alternative to the category of ‘ought* and thereby no alternative to the basic norm. Thus, Kelsen’s argument can be called a ‘weak transcendental argu­ment*. It shows that a basic norm (not necessarily Kelson's) that introduces the category of‘ought* is the key to the realm of the law.

(c) Norm that is Thought

The third characteristic of the Kelsenian variant of the basic norm is that this norm is supposed to be only an im­agined or a thought norm)** It must be this, for the basic norm as a willed norm would have to presuppose a further norm that would first of all transform the content of the willed into the content of an ‘ought’, since no ‘ought* follows from the merely willed. And then the basic norm would not be the basic norm.

The first and second characteristics of (he basic norm can be accepted. But here the problems begin, the first of which is the concept of a thought norm. Kelsen, in his later work,^{[180] [181]} rejected the thesis that the basic norm was simply the ‘content of an act of thought’.^[182] There is ‘no “ought” without a will’,^[183] he wrote, so that ‘along with a basic norm that is thought, an imaginary authority must also be thought... whose—fictitious—act of will has as its sense the basic

norm.’^{[184] [185] [186] Kelsen himself characterizes this notion as ‘self­contradictory’, for to say (hat the highest authority is empowered by a still higher—even if only fictitious—author­ity 92 is to say that the highest authority is not the highest authority. Moreover, a further basic norm would have to be invented to empower the fictitious authority to issue the basic norm, which would amount to not only denying the original basic norm its character as a basic norm, but also—since the further basic norm, loo, could only be the content of an act of will—presupposing W infinitum further fictitious authorities and the fictitious basic norms empower­ing them. This problem is not solved by Kelson’s thesis that a ‘genuine fiction’ is precisely what is at issue and that its being self-contradictory19, is precisely what distinguishes such a fiction.}

A solution can be found only if one gives up the notion that every ‘ought’ must be traceable to the willed. There arc good reasons to do this. While an ‘ought’ is usually connected with the willed, there are exceptions. Thus, one can arrive at the view, based on considerations of fairness or justice, that one is morally obligated not to evade taxation, but one can at the same time continue to want to evade taxation and for that reason act contrary to one’s insight into what is morally obligatory. Cognition of an ‘ought’ is necessarily attached neither to one’s own act of will nor to another’s act of will.^[187] If this is correct, then the notion that the basic norm is simply a thought norm causes no difficulties.

A second problem is that of the normative or prescriptive character of a basic norm that is thought. Kelsen formulates the thought basic norm to say (hat one ought to do something: ‘One ought to behave as (he constitution pre­scribes.'^[188] That is one side of the matter. The other side is that, according to Kelsen, legal science prescribes nothing when it bases its cognition of the law on this basic norm: ‘It docs not prescribe that one ought to obey the commands of the framers of the constitution.’^[189] How is it possible that a legal scientist, in formulating a sentence about what is legally obligatory, necessarily presupposes that one ought to behave as the constitution and thereby the law prescribes, while, on the other hand, he does not prescribe with the formulation of such a sentence that one ought to behave in accordance with the constitution and thereby the law? The solution lies in the concept of prescribing. Person A prescribes something to person  when A demands of  that  do something. According to Kelsen, the legal scientist qua legal scientist demands of no one that he ought to behave in accordance with the constitution and (hereby the law. The legal scientist can straightaway, qua legal scientist, give information about a legal obligation and at the same time, qua human being, demand that on moral grounds this legal obligation not be met. The result is that the normativity of the law is hypothet­ical or relative in character. A legal scientist who gives infor­mation about a legal obligation does not say, ‘You ought to carry out act a.' Rather, the information he gives is, ‘If you take the standpoint of the law, then you are obligated to carry out act a.' To be able to say this, and only to be able to say this, the Kelsenian basic norm is needed: ‘One ought to behave as the constitution prescribes.’ The legal scientist in fact prescribes nothing thereby, for the decision to take the standpoint of the law is left to whatever deliberations might be entertained by the addressee of the legal scientist's state­ment. Above all, there is no prescription that the addressee take the standpoint of the law. Rather, the attitude toward the law is completely indifferent. There is a legal obligation only for one who participates, for whatever reasons, in (he game of the law. For one who does not participate, there exists only the risk of bearing the brunt of coercive acts. In this respect, the law imposes no obligations whatsoever. There can be no question that an interpretation along these lines is possible. The question is only whether it is adequate.

At many points in Kelsen's texts, there is only the most incomplete expression of the hypothetical or relative character of the normativity of the law. Accordingly, the conclusion of a basic norm syllogism is supposed to say. according to Kelsen, that one ought to behave in a certain way.¹⁹⁸ That gives the impression that the basic norm leads to a categorical obliga­tion. independent of one’s standpoint, which in turn can encourage the mistaken interpretation that Kelsen establishes with his basic norm a general obligation to comply with any and all legal norms. In the context of the basic norm. then, it is better to speak of a legal ‘ought’, not simply an ‘ought’. Even then, the question remains open as to whether Kelsen’s inter­pretation of this ‘ought' is adequate.

(d) Incapability of Being Established

The fourth characteristic of the basic norm is supposed to be that it is incapable of being established: ‘Into the basis of the validity of the basic norm there can be no further enquiry.’¹⁹⁹ This thesis is plausible on first glance. The basic norm as basic norm is the highest norm. A still higher norm would have to be presupposed if the basic norm should be established, but then the basic norm would no longer be the highest norm and therefore no longer the basic norm. On second glance, how­ever, it is clear that this thesis is easily toppled. The basic norm al issue here is only the basic norm of the law. As the highest norm of the law, it cannot in fact be established by

Sec ibid. § 34(d) (at p. 202).

¹⁹⁹ Kelsen. ’The Function of the Constitution' (n. 192 above), 112 (trans, altered); see Kelsen. PTL § 34(a) (at p. 195). another norm of the law. That does not preclude, however, its being established by norms or normative points of view of a different kind, say, by norms of morality or through expe- diential deliberations. Kelsen could object that then these norms would be the basic norm of the law or these deliber­ations would have to be reformulated into a basic norm of the law. But that is not necessarily so. One can say that the move into the realm of the law is accomplished with the basic norm, and that there are moral or other extra-legal reasons for making this move.

To appreciate the error of Kelsen’s thesis that there can be ‘no further enquiry'^[190] into the validity of his basic norm, one need only ask why every coercive order that is by and large efficacious should be interpreted as a legal system. Kelsen is right when he says that only if one presupposes his basic norm can one interpret as a legal system every coercive order that is by and large efficacious. But why should one interpret as a legal system every coercive order that is by and large effica­cious? A reference to the basic norm is inadequate as an argument, for to presuppose the basic norm means precisely, after all, to interpret as a legal system every coercive order that is by and large efficacious. Since interpretation as a legal system and presupposition of the basic norm are two sides of the same coin, the one cannot be used as an argument for the other.

The question of why every coercive order that is by and large efficacious should be interpreted as a legal system, that is. why the Kelsenian basic norm should be presupposed, can be answered in very different ways. One answer has it that this is a matter of making a determination, a mere decision. That is not an argument, however. A second answer says that the presupposition of the basic norm is expedient. Individuals singly and collectively (say. as slates) could orient themselves better and so take action more successfully if they adopted this interpretation. That is an argument, but one might ask whether, of all the possibilities, the best presupposition for success is Kelsen’s basic norm. A third answer is that the basic norm is required on moral grounds, say, that civil war is to be avoided. Here again the decisive question is whether the best moral argument truly leads to the Kelsenian variant of the basic norm. The argument from injustice, as discussed above within the framework of the critique of legal positivism, showed that there are good moral reasons not to ascribe legal character to everything that has been issued and is efficacious, and the argument from principles led to the con­clusion that law· is not only what has been issued and is efficacious. That will be taken up again in discussing Kant’s basic norm. A fourth answer has it that Kelsen’s basic norm expresses what always underlies the jurist’s work in the law·. This is an empirico-reconstructive argument that Kelsen him­self approaches when he writes: ‘[The basic norm] simply raises to the level of consciousness what all jurists are doing (for the most part unwittingly)..But then he immediately takes leave of this view by adding: ‘when they understand the law exclusively as positive law’.^[191] The empirical question of w hether or not jurists understand the law exclusively as posi­tive law is not pursued by Kelsen. His thesis, therefore, to the effect that the basic norm simply raises to the level of con­sciousness what jurists are doing when they understand the law exclusively positivistically, is not an empirical claim. Rather than reconstructing the jurist’s endeavour empirically, Kelsen’s thesis explicates or defines the standpoint of the legal positivist. Not only is the question as to the correctness of this standpoint left open, the question as to the correctness of (he description of the work actually done by jurists is not addressed either.

In summary, then, the following may be said about Kel­son’s theory of the basic norm. Kelsen is right when he says that one must presuppose a basic norm if one wants to move from ascertaining that something has been issued and is effi­cacious to ascertaining that something is legally valid or legally obligatory. This basic norm, however, need not have the content of the Kelscnian basic norm. It may include moral elements that take into account the argument from injustice. Kelsen is also right when he says that while one must neces­sarily presuppose a basic norm if one wants to interpret the law as an ‘ought’-system, one may forgo this interpretation. Thus, the character of the basic norm is only weakly transcen­dental. Finally, Kelsen is right when he says that the basic norm is simply a norm that is thought. He is not right, however, when he claims that the basic norm is incapable of being established. On the contrary, the basic norm needs to be established. And that leads to the problem of a normative basic norm.