3. The double-faced character of norms and value judgments

The controversy among criminalists is, in our view, a good starting point for the study of legal values. Let us, therefore, take a closer look at some of the questions that have arisen in that dispute.

We said that a reconstruction of the circumstances exempting from culpability in terms of permissions would not make sense, in contrast to what is the case with reasons of justification, that is, those circumstances that exclude unlawfulness. We also said that one is permitted to kill in legitimate defence, but that an insane person is not permitted to kill, although she cannot be blamed for her behaviour. In other words, the behaviour of the insane brings about a result that is a disvalue from the point of view of criminal law; but an insane person cannot be guided by norms, and that makes it impossible for us to blame her for her behaviour and, thus, to be justified in punishing her.

It is that incapacity of being guided by norms which excludes the insane from the addressees of the directive part of criminal norms. The situation we confront when there is a reason for exemption from culpability is a situation where, for a certain category of subjects (those to whom the respective reason for the exemption from culpability applies), the judgment of disvalue con­cerning a certain behaviour does not translate into a prohibitory directive (nor any other kind of directive) of that behaviour.^[60]

Earlier, with respect to constitutional liberties, we saw another case where there was no complete correspondence between a value judgment and a directive: That type of permissive sentences — we said — has directive con­sequences, that is, they translate into mandatory principles and policies, but they are not equivalent to such directives, because they are also value judg­ments whose domain — whose expansive force — is greater than that of the directives deriving from them.⁷

⁷ With good reason, Juan Carlos Baydn has pointed out to us another case of incomplete corre­spondence, located outside of the sphere of law.

The status of supererogatory acts poses very difficult questions for ethics because it is not easy to accomodate them within a consistent conception of morality: How is it possible that performance of an action which, in some context, is the most valuable of all possible actions, is not obligatory? Must we not always do what is best? Actually, some very influential moral theories leave no room at all for the concept of supererogatory acts: According to classical utilitarianism, for example, agents must always choose the course of action that produces the greatest global good, irrespective of the cost this implies for the agent himself. And if James S. Fishkin (1986) is right in categorizing them under the label of 'systematic impartial consequentialism (SIC)', the same would be true of moral theories as distant from utilitarianism as those of John Rawls or Bruce Ackerman: the repeated admission of supererogatory acts in A Theory of Justice would be inconsistent with Rawls' basic position — grounded on the equal consideration of everyone's interests — that agents may not give special weight to their own interests or those of persons close to them.

One way open to this kind of conceptions of morality for accomodating supererogatory acts is the one suggested (and rejected) by Raz (1986, 197 f.): supererogatory acts could be understood as acts that are in fact obligatory, but compliance with which requires such extraordinary personal qualities that, while their omission cannot be justified, it can, however, be excused. But it would hardly be accepted as an adequate characterization of what the soldier in the first example did to say that he complied with his duty, whereas his comrades failed to do so, although they had an excuse for it. Except for fanatics, everyone would agree that what the soldier in question did was beyond the 'limits of obligation' (for this expression, cf. Fishkin 1986 and 1982), and that such conduct could not be required, neither of him nor of his comrades.

This kind of considerations has led Bayon (1991a, 364) to reject the assertion sustained — he says — by many moral philosophers that the sentences 'There is a moral reason for performing acts of class P' and 'The acts of class P are prima facie obligatory' express exactly the same thing. In Baydn's view, with respect to the class of supererogatory acts, the first sentence is true — because "otherwise one would not understand why their performance is morally praiseworthy, and not morally indifferent" —, but the second is false, because of supererogatory acts "we don't say that they are 'obligatory', not even prima facie".

However, in our opinion there is a way that allows us to assert that supererogatory acts are prima facie obligatory as well as that they are acts the performance of which, all things considered, is beyond the limits of obligation. Let us return to the example of the soldier: If we ask what act it was the soldier performed, and we answer 'saving the lives of his comrades', then it was undoub­tedly a prima facie obligatory act, since 'saving the lives of others' obviously is a prima facie obligatory act: everyone would agree that saving the lives of others under circumstances that do not imply a substantial sacrifice for the agent is obligatory, all things considered.

In order adequately to understand this kind of situations (which also per­haps helps us clarify the question of the nature of criminal norms), we think it can be very useful to return to the consideration of legal norms as reasons for action. Because when one says that a norm is a reason for action, one actually says two different things: that a norm is a guide for behaviour, and also that it is a criterion for the evaluation (that is, for the justification or critique) of be­haviour. These two elements normally overlap, and that is why their duality usually goes unnoticed — or rather, in most cases, does not need to be noticed. In a way, they are like the two sides of one and the same reality. But — as we have seen — there are occasions when the two sides seem to separate. Thus, we could say that two aspects can be distinguished in a norm: the directive, or — if you prefer — normative (in the strict sense) one, that is, the one that guides behaviour; and the evaluative one that contains a criterion of evaluation (that is, of justification or critique).

Interestingly, that very same duality has also been said to characterize values.

act performed by the soldier supererogatory is that, in this case, saving the lives of others did imply a very substantial sacrifice on the part of the agent. To see things in this way implies that supererogatory (individual) acts are cases of prima facie obligatory (generic) acts, performed under circumstances — as in the case of the soldier — or through activities — as in the case of the daily visit to the sick friend — such that performing the action goes beyond the limits of obligation (on the distinction between act or action and activity, which comes from von Wright, cf. Appendix to ch II, n. 2). In this light, we think that Rawls is right when he says (1971, 439) that "supererogatory actions are ones that would be duties were not certain exempting conditions fulfilled which make allowance for reasonable self-interest".

If we go back once more to the example of the soldier, the condition for reconstructing things in this way is, of course, that the act carried out by the soldier is described as ’an act of saving the lives of others'. If we would describe it as 'an act of sacrificing one's own life in order to save the lives of others', the supererogatory would be located not in the circumstances of performance of the individual act, but in the generic act itself. The relevant difference between the first — 'an act of saving the lives of others' — and the second description — 'an act of sacrificing one's own life in order to save the lives of others' — is that the first, but not the second points to the con­sequence (the change in the world) constituting the agent's intention. And it is justified to point to the consequence constituting the intention of the agent — as a criterion for the description of an action —, because it is the intention that makes the act morally valuable: what the soldier did was morally valuable because it was done in order to save the lives of others and not, for example, in order to commit suicide or to be admired posthumously.

On the other hand, we agree with Bayon that the reason why supererogatory behaviour cannot be required is the moral relevance of the agent's autonomy, that is, the weight an agent may legiti­mately give to his own personal projects: the value of personal autonomy is the barrier which in the case of supererogatory acts prevents complete correspondence between the evaluative and the directive spheres, because in the contexts here described it does not allow one to pass from the assertion that 'X is the most valuable action, all things considered' to the assertion that 'X is obligatory, all things considered'.

According to Nicholas Rescher, to subscribe to a value is manifested in two different ways:

"First, on the side of talk (or thought), in claiming that TV subscribes to a value, we give grounds for expecting a certain characteristic type of verbal action, namely, that he would 'appeal to this value', both in the support or justification of his own (or other people's) actions and in urging upon others the adoption of actions, courses of action, and policies for acting.

But second, on the other hand, we also expect the value to manifest itself on the side of overt action. We would draw back from saying that 'patriotism' ('financial security', 'the advancement of learning') is one of Ns values unless he behaves in action — and not just at the verbal level — so as to implement the holding of this value by 'acting in accordance with it' himself, by en­deavoring to promote its adoption by others, etc. In saying that prudence, for example, is one of Ns values, we underwrite the presumption — this does not, of course, guarantee — that N be­haves prudently (is a prudent person) [...] A value is thus bound up with a Janus-headed disposi­tion cluster, and we expect it to orient itself in two directions, that of discourse and that of overt action [...]

Subscription to a value is consequently a two-sided affair, and value imputations have a double aspect: both verbal and behavioral. When we impute to the person N subscription to the value x, we underwrite the grounds for expecting from N a reasonable degree of conformity with the cha­racteristic manifestation patterns of subscription to x both in discourse and in action. We thus im­pute a value to someone to characterize his vision of 'the good life' or at any rate his vision of how life ought appropriately to be lived" (Rescher 1969, 2 f.)

A similar idea, in the sense that acceptance of a value judgment cannot be se­parated from acceptance of standards of behaviour implied by that value judg­ment, can be found — referring especially to moral value judgments — in the work of R. M. Hare:

"[...] a moral judgement has to be such that if a person assents to it, he must assent to some im­perative sentence derivable from it; in other words, if a person does not assent to some such im­perative sentence, that is knock-down evidence that he does not assent to the moral judgement in an evaluative sense — though of course he may assent to it in some other sense [...] But to say this is to say that if he professes to assent to the moral judgement, but does not assent to the im­perative, he must have misunderstood the moral judgement (by taking it to be non-evaluative, though the speaker intended it to be evaluative). We are therefore clearly entitled to say that the moral judgement entails the imperative; for to say that one judgement entails another is simply to say that you cannot assent to the first and dissent from the second unless you have misunderstood one or the other; and this 'cannot' is a logical 'cannot' — if someone assents to the first and not to the second, this is in itself a sufficient criterion for saying that he has misunderstood the meaning of one or the other. Thus to say that moral judgements guide actions, and to say that they entail imperatives, comes to much the same thing." (Hare 1952,171 f.).^

Carlos Nino, however, has criticized Hare's view of the relationship between value judgments and directives for behaviour (especially when these directives — as is the case with most legal norms — originate in prescriptive acts by an authority). Nino stresses that, once the distinctive characteristics of evaluative judgments have been explained "in terms of reasons, one can clearly see how they differ radically [...] from commands and legal norms" (Nino 1985b, 116). The essential difference is said to be the following: Evaluative judgments indi­cate the existence of a reason for acting other than the formulation of the judg­ment itself, whereas imperatives, of which commands and legal norms are typical examples, "do not indicate the existence of reasons for acting; rather, they are assumed to constitute such reasons themselves" (ibid., 117).

In our view, however, Nino’s disagreement with authors like Rescher or (explicitly) Hare rests on a misunderstanding provoked by the ambiguity of ex­pressions like 'imperatives' or 'legal norms'. Here, we will concentrate on the case of legal norms. The following two things — among many others — can be understood by such a norm: on the one hand, a prescription itself, as issued by an authority; on the other, the ought judgment reproduced by the content of such a prescription. Consider the following argument:

1) What A commands ought to be done.

2) A has commanded: 'In circumstances C, X should be done'.

3) In circumstances C, X ought to be done.

An explanation may be useful for a better understanding of Hare's conception. In the text, Hare speaks of 'evaluative moral judgements' and 'non-evaluative moral judgements' (or, as he calls them elsewhere, 'inverted-commas moral judgements'). With this somewhat strange terminology — 'non-evaluative moral judgements', 'inverted-commas moral judgements' — Hare refers to sentences concerning either the standards accepted as moral in a certain period or social group, etc. (for example, 'In Spain, in the forties, it was immoral for girls not to be virgins when they got married'), or the moral feelings of some individual or set of individuals (for example, 'The socialist representatives' consciences told them that they should vote against their colleague's petition'). Obviously, sentences like those contained in these examples do not express genuine moral value judgments: as Hare himself says, they are rather statements "of sociological fact" in the former and "of psychological fact" in the latter case. That consent with a moral judgment logically implies consent with the standard(s) of behaviour or, as Hare says, the imperative(s) deriving from it applies, of course, only to genuine (or, in Hare's terminology, to evaluative, not inverted-commas) moral judgments.

Obviously, if by a 'legal norm' (or 'imperative') we understand the prescription contained in 2) (or the fact that that prescriptive act has been performed), Nino is right. And the fact that he identifies them with 'commands' indicates that Nino is, indeed, speaking of 'legal norms' precisely in that sense. But it is not in the sense of 2) that we speak of 'legal norms' when we regard them as operative reasons or when we say that in a 'legal syllogism', the major premise is con­stituted by a norm.^[61] In this context, the relevant sense of 'legal norm' (or 'im­perative', if we want to stick with Hare's terminology) is that expressed in sentences like 3). Such sentences are ought-judgments deriving a) from the rec­ognition of some person (or organ, or procedure, etc.) as an authority — which is expressed in sentences like 1) —, and b) from the fact that that person (or organ, or procedure, etc.) has issued a prescription. And ought-judgments like that contained in 3) have the 'double-faced features' of value judgments under­scored by Rescher and Hare; that is, they operate as guides of behaviour as well as criteria for the evaluation of behaviour.

Now, if all that is right, we must conclude that the difference between value judgments and norms is one of degree, or emphasis. The question then is how the two aspects — the evaluative and the directive — can be understood and combined. Let us look at the following sentences:

1) 'Life is a good.'

2) 'Thou shalt not kill.'

3) 'It is prohibited to kill, unless there is a justificatory reason.'

4) 'Judges must sentence those who have killed another person

to a minor prison term, unless there has been some justificatory reason or the author is a subject not responsible for his actions.'

Provisionally, we can say that 1) is a sentence expressing a value judgment; 2) is a statement of a principle (a principle norm); and 3) and 4) are mandatory rules with different addressees: those of 3) are citizens in general, and those of

4) are judges. But let us look at things more closely.

To say that 7) is a value judgment is not, of course (in contrast to what Mir Puig seems to suggest), the same as expressing a wish. According to Re­seller, we would say that for a person who accepts it, this value judgment im­plies that, in principle (let us assume that the assertion is not that life is an — or the — absolute value), he regards actions and omissions intended to main­tain people's lives (again, let us assume the statement refers exclusively to the life of persons) as justified, and those provoking death as unjustified, and that he himself is disposed to perform or expects others to perform (also, in prin­ciple) actions and omissions intended to save life.

Now, 2) does not seem to be very different from 1). To accept 2) as a principle of behaviour also means to be disposed to preserve the lives of others, and that actions consistent with this principle would be considered justified and those in conflict with it unjustified and blameworthy. The two sentences stress different aspects, but other than that they say the same thing. Besides, in 7) as well as in 2) the axiological side has, so to speak, priority over the directive one: it makes sense to say that one should not kill because life is a good, or a value, but not that life is a value because one should not kill.^{[62] [63] If we accept the convention that in a value judgment [like 7)] there is special emphasis on the value element, and in a norm — a principle — [like 2)], on the directive element, then we can say that 1) — value judgment — has justificatory priority over 2), i. e. over principles.11}

3) and 4) are clearly rules. Although in the conditions of application as well as in the legal consequences there are terms that may be vague, the con­figuration is of what we have earlier called ’closed form’. Thus, for those who accept those norms, 3) and 4) entail protected or peremptory reasons — and 3), besides, with respect to the reasons of justification, the absence of a peremptory reason — for doing what is stipulated, and for regarding behaviour that is in accordance with those norms as justified.

From 1) to 3) and 4), there is a process of specification; but in J) as well as in 3) and 4), the two aspects of guiding and justifying behaviour are both present. The difference is that the guidance and justification offered by 3) and 4) is more precisely defined than in the other two cases, and that that guidance and justification is peremptory.

Now, the examples can be generalized in the sense that the two elements are not only present in (primary or secondary) criminal norms, but in all the kinds of norms we have distinguished: not only in mandatory rules (including — as we will see later — the rule of recognition) and in principles in the strict sense, but also in policies and in power-conferring norms. For example, a poli­cy stipulating the goal of full employment is, prima facie, a guide for the be­haviour of the public powers which, at the same time, serves as a (prima facie) justification of actions that have been or could be undertaken in that direction. In the case of power-conferring norms, the guidance and justification is indirect or hypothetical: if it is justified to bring about result R (either because a mandatory norm stipulates that bringing it about is obligatory, or because bringing it about corresponds to a wish or interest of the agent and is not subject to normative constraints), given circumstances X, Y 'should' be per­formed. Here too, it makes sense to say that — given the circumstances — Y is a justified action, but here we have another kind of 'justification' than before. The reason is that the values implied in the different types of norms are not of the same kind. So we should examine this somewhat closer.