From IVR encyclopedie
The concept of gaps
The concept of legal gap has different meanings. Lawyers often speak of a gap in a case which is covered by a general rule but where, they believe, the author of the rule has forgotten to make an exception. The author of the rule ‘if p then q ought to be done’ should have made an exception for the the case p & r but has not done so because he did not consider the possibility that q ought not to be done if p & r. We might baptize this kind of gaps, following Alchourrón and Bulygin (1971), axiological gaps. According to a second interpretation a legal gap exists in a case which is not covered by a legal rule at all. There is a gap if in some legal system L an act p in situation q is neither obligatory, prohibited, nor permitted. Alchourrón and Bulygin (1971, 20 ff) call this a normative gap. There is in such a case absence of law. This entry does not discuss the axiological gaps, which are to be dealt with under legal interpretation or judicial decision-making. In this entry ‘gap’ means ‘normative gap’. The question is: can these gaps exist in law?
Gap denying strategies
Logical analysis is needed to define the problem of the existence of gaps, but it is not exhaustive. Logical analysis can at most prove the logical possibility of legal gaps, not that legal gaps really exist within systems of law. This preliminary point will be clear if we compare legal systems with moral systems. It may be argued that morality is necessarily (the modality is not logically necessity) complete: every act is morally obligatory, forbidden, or permitted. Of course, we might disagree in our moral opinions: euthanasia is wrong according to the Pope's morality, but under certain circumstances it is not according to a more liberal conception of morality. However, if morality is not determined by whim or caprice, but is related, e.g., to human well being or human flourishing, then, in principle, every act can be judged according to its effects on human well being or flourishing: for any particular act A, does A promote well being or flourishing, does A diminish it, or is A irrelevant for it? But positive law is not morality. If, for the sake of argument at least, we follow the sources thesis of legal positivism, positive law is completely determined by legal sources, which are constituted by human authorities who have issued the legal norms. If the number of legal sources is limited, the consequence seems to be that the number of norms originating from these sources also is limited. This seems to allow for the possibility of absence of law: particular cases can arise that are not covered by one of the norms in the legal sources. One might nevertheless pursue the following strategies for denying the possibility of gaps in positive law for the following reasons. The first strategy begins with the claim that if logic cannot exclude the possibility of complete normative systems, it might still exclude the possibility of incomplete systems. In deontic logic the theorem that, for any act p, p is not forbidden if and only if p is permitted is widely recognized (i.e., the claim that p is not forbidden is logically equivalent to the claim that p is permitted). The consequence of this principle of prohibition is, it seems, that any act which is not prohibited in some normative system is, for reasons of logic, permitted in that system. This would mean that every normative system necessarily is complete. The second strategy makes use of legal rules of closure, such as the nulla poena sine lege rule, and argues that every legal system contains a rule of closure. The third strategy tackles the issue of legal interpretation. It argues that existing legal systems should be interpreted as 'seamless webs' and that they have answers for all cases, even the hard ones. This move is possible insofar as an existing legal system is considered to elaborate a particular conception of justice (or an 'integer' compromise between different conceptions), which conception can be reconstructed from its elaboration in positive law and the ideals of justice which are dominant in society and next can be applied to 'new' cases (Dworkin, 1986). The second strategy has no serious defenders. Rules of closure do exist within penal law, but there is no support for the empirical thesis that closure rules exist in all legal system. I will not discuss the third strategy: this is a topic for other entries. The problem of this entry is: does the principle of prohibition exclude the possibility that normative gaps exist.
Norms and normative propositions
Alchourrón and Bulygin start their discussion of whether the principle of prohibition excludes gaps with a distinction between norms and normative propositions. Norms are prescriptive sentences, which are used to give obligations, prohibitions or permissions. Normative propositions are descriptive sentences that convey information about norms or about the obligations, prohibitions or permissions issued by the norms (1971, 121). The principle of prohibition pertains, they argue, to norms, not to normative propositions. It is perfectly possible that both the normative proposition that p is prohibited and the normative proposition that p is permitted (in some system) are false: the system may be silent. Likewise, it is possible that both normative propositions are true: the system may be inconsistent: "There is nothing paradoxical about a consistent description of an inconsistent normative system" (1971, 123). There is, however, some ambiguity in this approach. Normative propositions are taken to say something about norms (i.e. normative sentences) or about obligations, prohibitions and permissions. It is clear that normative sentences may be silent or inconsistent concerning some act p. But it does not necessarily follow that the normative system is silent or inconsistent. It is possible to attempt to explain the legal validity of some norm in terms of its being issued by a generally recognized legal source. According to such an explanation, the occurrence of inconsistent normative sentences in the legal sources of a system renders the system inconsistent if in some case q the act p is, according to the system, both prohibited and permitted. According to Dworkin such a definition would be too narrow (as it cannot account for legal principles), but it certainly is too broad. Legal systems quite often formulate exceptions to norms. These exceptions sometimes are to be found in very different parts of the system's statutes and precedents. Legal systems also usually make use of priority rules, e.g. lex posterior derogat legi priori. It is, therefore, unacceptable to hold that every norm which can be found in a legal source is necessarily valid. If we still follow the positivist approach, we should explain legal validity of norms in terms of being issued by the normative legal system. To know whether some norm is legally valid, we then have to interpret the legal system and to determine whether the norm exists in an acceptable interpretation of that system. It may be argued that some interpretation which allows for inconsistent norms within the same system does not present an acceptable interpretation. In that case, the normative propositions describing both contradictory norms cannot both be true. In the same way, if the normative sentences of some system are silent about some act, it does not follow that the system itself is silent. The principle of prohibition seems to say that according to the system such an act is permitted. In his Rechtsphilosophie, the German legal philosopher Gustav Radbruch argued that in a strict sense absence of law ("ein rechtsleerer Raum") is impossible: when a legal system is silent it regulates in a negative way, by denying legal consequences (Radbruch, 1963, p.298).
Weak and strong permission
Making use of this distinction between norms and normative propositions we can make a difference between a strong permission and a weak permission (Von Wright, 1963, p. 86). An act p is strongly permitted in the case q by a normative system means that a norm to the effect that p is permitted in q is a consequence of the system. The act p is weakly permitted in the case q by the normative system means that among the consequences of the system there is no norm which prohibits p in the case of q.
The sentence saying that p is weakly permitted can, it seems, only be a descriptive normative proposition. If a prescriptive sentence would say that an act is weakly permitted it is permittting the act. It is therefore making the act strongly permitted. The principle of prohibition is acceptable as a logical principle only if it is taken to mean that an act p which is not prohibited is weakly permitted. This is evidently true: the weak permission is defined as the absence of a prohibition. The principle is not a logical truth if it would refer to a strong permission.
It seems, therefore, that the principle of prohibition does not exclude gaps. On the contrary, if there is a gap, then there is a weak permission: a gap is a case where there is no normative solution to be found in the normative system; hence there is no prohibition either (Alchourrón en Bulygin 1991, 126). But it also seems that Radbruch is right: by not prohibiting an act the legal order is regulating the act in a negative way. Gaps have legal relevance.
The legal difference between weak and strong
The distinction between weak and strong permission is easily made and the consequences concerning the meaning of the principle of prohibition follow logically from it. But there is a question left. We have to note that weak and strong permission do not exclude one another. On the contrary, if according to a consistent normative system an act p is strongly permitted in case q, then it also is weakly permitted: a consistent system does not prohibit an act which it also permits. The reverse, of course, is not valid: according to the definitions an act may be weakly permitted without being strongly permitted. The class of strongly permitted acts (within some normative system) therefore is included in the class of weakly permitted acts (i.e.: if the normative system is consistent in its relevant part). The remaining question is: wat does the strong permission add to the weak permission? The criteria are clear, but what is the legal difference between an act which is only weakly permitted and an act which is also strongly permitted? We may compare criminal law and civil law. Criminal law is usually seen as a closed legal system, because the legal principle "nullum crimen sine lege", valid in most modern penal systems, functions as a rule of closure, excluding gaps for systems of criminal law. Civil law, on the other hand, usually is seen as not containing a rule of closure: civil law is open. What then, is the difference between a weak civil permission and a strong penal permission, when in both cases no explicit rule exists governing the case, apart from the penal closing principle? It seems that if the nulla poena principle penally permits every act which is not penally prohibited, this means that such an act is strongly permitted. When a legislator ex post facto or a judge nevertheless sanctions such an act, they violate this strong permission. It seems that here we find what the strong permission adds to the weak permission. Under both permissions the subject of the norm may perform the act. But the strong permission, in a way, guarantees this situation, by prohibiting norm-authorities to prohibit the act (or to sanction it).
Weak and weaker; strong and stronger
Considered from this point of view, the strength of a permission is not an all-or-nothing characteristic. Some strong permissions may be stronger than other ones. The strong permission issued by the nulla poena principle prohibits judges to make new crimes and it prohibits legislators to make new crimes ex post facto. But it does not prohibit legislators to issue new future crimes. A fundamental human right, guaranteed in the constitution or in the European Convention on Human Rights, is stronger: it also prohibits legislators to issue prohibitions for future acts which are permitted according to these constitutional or conventional provisions. This difference in strength not only pertains to normative propositions, it also pertains to norms. Even the legislator's silence may differ in strength. The legislator can be silent because he did not consider the case. But it also is possible that he considered it but thought it to be more sensible not to issue a norm. In the last case the permission might be stronger, as this ‘explicit silence’ may imply that lower authorities do not have the competence to reconsider the case. A judge can refuse to give his considered judgement on some matter because he believes the matter to be not relevant for the case at hand: he does not judge whether act p in case q is forbidden by the normative system, as act r in case s is forbidden, and the issue to decide is whether the act, which is p as well as r, in the given situation, which is q as well as s, is prohibited. Or he can refuse to give his judgement because some other judge is competent. Or because the case is not within judicial competence at all: the judge refuses to decide which is the true religion or which solution to some scientific problem is right. In all these cases the judge does not give a decision, he is silent, but the strength of his silence differs. In the first case it only pertains to judges who are asked to decide in the same type of cases (where the two prohibitions coincide), in the second case it pertains to all judges who have the same competence and in the last case it pertains to all judges. The implicit message of all the refusals is that different judges in different situations also have to refuse. (Whether this implicit message is legally valid depends from the ranking of the court within the judicial system and the role of precedents in the normative legal system: there is a difference between the (implicit) precedent-meaning and the legal force of the precedent.) If "an authority weakly permits an act" means that he does not, on consideration, prohibit the act, then these weak permissions do not imply that there is no valuation. In the judicial cases there is, e.g., a valuation on the point of the admissibility of the case. As this valuation may have a wider or less wide range its relevance for other judges in other cases may be different and the weak permission may be relatively stronger or weaker. In some situations mentioned above it is so weak that ordinary legal language does not even call it a permission. In other cases it is stronger, as the (potential) parties receive the message that no judge should admit cases like this one (or as the judge admits the case and decides that the act is not prohibited). In all these cases the judge, as normative authority, does not give a prohibition, just as the legislator who is silent does not give a prohibition. It is perhaps overstressing my point to say that the judge issues a non-prohibition, as he issues no norm at all, but what he does is on the level of norms and resides on valuation.
The possibility of legal gaps reconsidered
The analysis of the last section supports the view that the principle of prohibition only is valid for weak permissions. If an act is not prohibited it is at least weakly permitted, but from the absence of a prohibition nothing follows as to the strength of this permission. Another important point is that the principle of prohibition differs from rules of closure, as the nulla poena rule. This last rule has the effect that it gives some strength to the permissions of all the acts which are not prohibited by other rules of the system. Rules of closure are normative (legal) rules, not logical principles. However, rules of closure, even if present, are irrelevant for the problem of legal gaps. That is because rules of closure issue no norms for the subjects. They are secondary norms for legal authorities, prohibiting these authorities, under given circumstances, to issue a prohibition. Weak permission, it seems, is all that is left of the permission on the primary level of the normative subjects. Which is the consequence of this conclusion for the possible existence of legal gaps? Legal gaps may exist if this is taken to mean that there is no explicit legal rule covering the case. Legal gaps do not exist if this is taken to mean that in some legal system L an act p in situation q is neither obligatory, nor prohibited nor indifferent. If there is no rule in L prohibiting p in case q then, necessarily, for the time being the normsubject is (weakly) permitted to do p is in case q. It may be the case that p is also permitted strongly. But this strong permission does not change the answer which should be given to a normsubject’s question: is it according to L permitted for me, at this moment, to do p in case q.
Alchourrón, C.E. and E.Bulygin, Normative Systems, Wien, 1971
Soeteman, A, On Legal Gaps, in E.G. Valdés, W.Krawietz, G.H. von Wright and R. Zimmerling, Normative Systems in Legal and Moral Theory (Festschrift for Carlos Alchourrón and Eugenio Bulygin), Berlin, 1997
Wright, G.H. von, Norm and Action, London, 1963
Dworkin, Law’s Empire, London, 1986
Radbruch, G., Rechtsphilosophie, 6th edition, Stuttgart, 1963