In view of an express regulation: Considering the scope and soundness of a contrario reasoning © Henrike Jansen. Informal Logic, Vol. 28, No. 1 (2008), pp. 44-59. In view of an express regulation: Considering the scope and soundness of a contrario reasoning HENRIKE JANSEN Universiteit Leiden Opleiding Nederlandse Taal & Cultuur Postbus 9515 2300 RA Leiden Netherlands h.jansen@let.leidenuniv.nl Abstract: A contrario reasoning (or ‘a contrario argument’ or ‘argument a contrario’) is traditionally understood as an appeal to the deliberate silence of the legislator: because a legal rule does not mention case X specifically, the rule is not applicable to it. Modern perspectives on legal reasoning often apply this label to a broader concept of reasoning, namely the reasoning by which a legal rule is not applied because of the differences between the case at hand and the one(s) mentioned in the legal rule. This article first explains how the broader concept could have come into being, and then argues that from an argumentation theoretical point of view the modern concept makes no sense as a category of argumentation. Furthermore it is shown under which conditions the traditional concept can be sound. Résumé: On entend traditionnellement le raisonnement a contrario (ou l’argument a contrario) comme un appel au silence délibéré des législateurs : puisqu’une règle légale ne mentionne pas spécifiquement le cas X, celle-ci ne s’y applique pas. Des perspectives modernes sur le raisonnement légal appliquent aussi cette étiquette à un concept de raisonnement plus large, en particulier, le raisonnement par lequel une règle légale n’est pas appliquée à cause d’une différence entre le procès en cours et celui (ceux) mentionné(s) dans la règle générale. Cet article explique premièrement l’origine possible du concept plus large, et ensuite avance d’un point de vue d’argumentation théorique que le concept moderne, perçu comme catégorie d’argumentation, n’a aucun bon sens. En plus, on décrit les conditions sur lesquelles le concept traditionnel peut être bien fond Keywords: analogical argumentation, appeal to differences, a contrario, common ground, formulation of a legal rule, juridical starting points, legal argumentation, legal principles, interpretation, nature of a legal rule 1. Introduction In the 1950s a Dutch woman took a Swedish man to court, asking that he be declared the father of her child. The woman and the man had both spent the summer working in a Swedish hotel. It was established that they had had sexual intercourse during that summer. However, it was The scope and soundness of a contrario reasoning 45 uncertain whether the man could actually be regarded as the father of the child that was born subsequently. The man argued that the woman had also had sexual intercourse with other men. Since DNA-tests were then unavailable, the court had to examine the claim’s likeliness. It was established that the conception must have taken place between 16 July and 24 October 1951 (according to Swedish law: between the 300th and the 200th day before birth). In addition, the mother had declared before the Swedish court that from July until September the only person with whom she had a sexual relationship was the man she alleged to be the father of her child. From this statement the Dutch court inferred that, apparently, after September she had had sex with different men and therefore concluded that the alleged father could not with certainty be declared to be the father. The court described its own line of reasoning as a contrario reasoning. It had taken the woman’s statement that in a certain period she had had sex with this man alone to contain an extra meaning, namely that outside that period she had not had sex with this man alone. This precisely defines the nature of the a contrario argument: saying something explicitly about one thing is interpreted as saying the opposite of another thing. An express statement that mentions some explicit items is interpreted as a statement that at the same time conveys the opposite about the items that are not mentioned. Although the argument a contrario is also used in other contexts outside the legal field, it is known primarily as a form of legal reasoning. Most often this line of reasoning is used as a method for the interpretation of legal rules. In such an a contrario argument an appeal is made to the silence of a rule of law about a specific situation: from the fact that a rule of law only regulates one or more explicitly stated situations, it is deduced that this regulation does not apply to situations that the rule does not mention. As Pitlo (1995, p. 33) puts it:1 One finds in the law a regulation for situation A and therefore concludes that the regulation does not apply to situations B and C.2 In other words: a contrario reasoning gives an interpretation of a legal rule such that it only applies to the situation(s) that it explicitly cites .3 Latin phrases like Qui de uno dicit, de altero negat [Who speaks about one thing, denies the other thing] and Unius positio est negatio alterius 1 See also Diephuis (1869, p. 115), Van Bemmelen (1891, p. 17-18), Land (1910, p. 25), Baumgarten (1939, p. 39), Germann (1946, p. 129), Esser (1949, p. 182), Canaris (1983, p. 51), Bydlinski (1991, p. 477). 2 My own translation from the Dutch. Both ‘a contrario’ and ‘a contrario’ spellings are to be found in the literature. In quotations, the quoted author’s own spelling is retained. 3 For this reason Perelman (1979, p. 83) describes the argument a contrario with the phrase ‘in view of lack of an express provision’. Perelman’s description is based on Tarello’s (1972, p. 104): ‘faute d’une autre disposition expresse’. Henrike Jansen 46 [One thing being put/posed means the negation of the other thing]—often used by juridical authors to clarify the meaning of the a contrario argument—express a similar purport. I will refer to this kind of reasoning by the expression ‘linguistic’ a contrario argument. It is linguistic because the argument is based on the wording of the text: on what it mentions and what it does not mention (more on this in section 2). Confusingly, however, in legal theory the expression ‘argument a contrario’ is also used to point to a much broader concept of reasoning. According to this concept the a contrario argument boils down to not applying a legal rule analogically, because instead of similarities between the situation at issue and the one(s) mentioned in the legal rule, or notwithstanding similarities, there are relevant differences between the two that are decisive (Engisch, 1956, p. 144; Boasson, 1966, p. 75-76; Horovitz, 1972, p. 44-48; Scholten, 1974, p. 69; Perelman & Olbrechts- Tyteca 1976, p. 325; Perelman 1979, p. 23; Soeteman, 1981, p. 355, Zippelius, 1985, p. 70; Kaptein, 1991, p. 78; Tak, 1994, p. 282; Henket & van den Hoven, 1996, p. 137; Franken, 2003, p. 191): There, where the resemblance stops, where an essential difference stands out, the analogy has its boundary and the so-called […4] (argumentum e contrario) is given a chance, namely the reasoning that attributes different legal consequences to different legal conditions. (Engisch, 1956, p. 144) [Dort, wo die Ähnlichkeit aufhört, wo ein wesentlicher Unterschied hervortritt, hat die Analogie ihre Grenze und kommt u. U. der sogenannte Umkehrschluss (argumentum e contrario) zum Zuge, der Schluss nämlich von der Verschiedenheit der Voraussetzungen auf die Verschiedenheit der rechtlichen Folgen.] I will call this concept the ‘argument from relevant differences’. The argument from relevant differences resembles the linguistic a contrario argument described earlier in that both entail the same result: not applying a legal rule to the present situation, which is not cited by this legal rule. However, this result is their sole resemblance. After all, the result is reached in two rather different ways. In the linguistic a contrario reasoning the legal rule is not applicable to the situation at hand because of the simple fact that the situation at hand is just not mentioned in the legal rule. In the argument from relevant differences it is not applicable because the situation at hand differs from the situation(s) mentioned in the legal rule. Since these ways of reasoning are really different, I do not see the utility of using the expression a contrario reasoning to indicate both ways of reasoning. Moreover, taking the argument from relevant differences for the argument a contrario is 4 The German expression Umkehrschluss is virtually untranslatable; it means something like ‘conclusion based on opposites’. The scope and soundness of a contrario reasoning 47 problematic for two reasons. First, its use does not tally with the way the argument a contrario is understood in legal practice and in non-legal contexts. Second, it creates confusion about the argument’s soundness.5 In my view, much disagreement in legal theory about the soundness of a contrario reasoning can be traced back to a confusion between these two concepts of a contrario reasoning.6 In this paper I will address both problems—conceptual confusion about the argument a contrario and confusion about its soundness—from an argumentation theoretical point of view. Adopting this perspective helps produce a clearer view of some of the problems of the a contrario argument than legal theory can offer. First, I will show how the existence of two concepts of a contrario reasoning have come into being. This explanation supports the conclusion that the broader concept of the a contrario argument as an appeal to relevant differences must not be equated with a contrario reasoning. I will then discuss the argument’s soundness. Even though the linguistic a contrario argument is regarded unsound in its original understanding, in a modern perspective it may be based on implicit premises that make the argument more plausible— though not automatically valid. 2. The linguistic a contrario argument vs. the argument from relevant differences As the introduction makes clear, the linguistic a contrario argument represents a strict interpretation of legal rules. Its linguistic element is included in the appeal to the literal meaning of the words that constitute the legal rule (Aarnio, 1987, p. 105; Pitlo, 1972, p. 134-135; Kitzler, 1986, p. 99-100). In contrast to the common understanding of linguistic arguments, the linguistic a contrario argument is not about the interpretation of these words according to their meaning in daily life or their technical meaning. Instead, the linguistic a contrario argument appeals to the distinctness of the words that make up the legal rule. Those words function as an argument because they are considered to be unmistakably clear. This manner of reasoning proceeds as follows: since the legal rule is unambiguously clear about the situation(s) it mentions, it cannot be applied to the situations it does not mention. Thus, the linguistic a contrario argument is a specific way of reasoning in itself: ‘because the legal rule denotes X, it is not applicable to Y’. The absence of a certain situation in a legal rule is taken as special evidence for the conclusion that the legal rule in question does not apply to that situation. 5 ‘Soundness’ refers to the argument’s acceptability, concerning both the logical validity and the acceptability of the content of the premises. 6 See for example Horovitz’s (1972) reaction to Klug (1982), Kaptein’s (1991) reaction to Nieuwenhuis (1976) and the discussion between Henket (1992a) and Kaptein (1991, 1993). Henrike Jansen 48 Legalism embraces such a kind of argument. At least in the legalistic period in the Netherlands (the second half of the 19th century) it was good usage—in theory anyway—to appeal to the law’s express words (Fockema Andreae, 1904, p. 43; Kop, 1992, p. 40). However, since the decline of Dutch legalism, the argument a contrario has gained considerable notoriety as a form of argument.7 After all, the argument is based on the assumption that the law comprises a complete and coherent system, which is fully up to date. Only such an assumption accounts for the inference that if a legal rule does not list certain situations, the legislator must be supposed to have left those out on purpose (Schneider, 1965, p. 178-179; van Hoecke, 1979, p. 177; Fikentscher, 1975, p. 546). Nevertheless, such an assumption presents a rather idealised view of the legal system. The legislator may have overlooked a certain situation, or a certain situation may have resulted from developments that could not have been foreseen. Therefore, Schneider (1965, p. 178) calls the argument a contrario a petitio principii if the silence of the legislator is presupposed to be intentional. An example of an argument based on such a presupposition comes from the Belgian Court of Cassation: After all, it is not conceivable that the legislator had stipulated the condition under which the restriction made by him was made subject to, in case he would have wished the same with regard to situations that do not fulfil it. (van Hoecke, 1979, p. 177) [Dat het immers niet denkbaar is dat de wetgever de voorwaarde heeft bepaald waaraan de door hem ingestelde beperking was onderworpen, indien hij hetzelfde had willen doen gelden voor de gevallen, die deze zelfde voorwaarde niet vervullen.] In this argument it is said that the legislator precisely regulated what he intended to, which is taken as evidence for not applying the rule to situations not listed. Since this assumption about the legislator’s intention is not argued for, the reasoning must be regarded as circular. In legal theory, remarks about the fallaciousness of the argument a contrario must be interpreted against this background. Germann (1946, p. 129) calls the argument a contrario a formal method of interpretation, which overlooks the nature and interpretation of the law. Making use of such an argument would reveal the laziness of the arguer; depending on the need for a certain outcome one could reason a contrario in the one case and analogically in another. It would also show garrulous confidence in those ‘learned-sounding formulas drafted in Latin’ [‘die 7 For that matter, criticism on the way of reasoning of an a contrario argument already existed during the legalistic period. Opzoomer (1849, p. 729) considers: ‘If I state: No father is allowed to abuse his child, does it follow then, that a mother is indeed allowed? It would follow only then, if I had considered the mother and had intended to judge her situation as well. Thus, the rule must be confined to: qui de uno dicit, de altero, de quo etiam statuere voluit [which he also intended to judge], negat’. The scope and soundness of a contrario reasoning 49 gelehrt klingenden lateinisch formulierten Rezepte’]. Esser (1949, p. 182) refers to a ‘logical fake’, which has nothing to do with objective interpretation. In legal practice, the a contrario argument also has a bad name: ‘arguments a contrario do not enjoy great popularity’ (HR 1 December 1998, NJ 1999, 310) and ‘analogical arguments are to be preferred to arguments a contrario’ (HR 6 February 1998, NJ 1999, 478). The general contention of the above is that the a contrario argument is an unsound form of reasoning. In my view it is precisely this conclusion that has led to the conceptual development of this kind of argument. The new concept is the result of attempts to transform the unsound linguistic a contrario argument into a sound argument. The standpoint that a certain legal rule is not applicable to a certain situation not mentioned in that rule must gain better support than the simplistic reason that the situation at hand is not cited in that rule. According to the literature, better support consists of establishing the relevant differences between the situation at hand and the one(s) mentioned in the legal rule. The argumentation then consists of one or more interpretative arguments (see for a classification McCormick & Summers, 1991), on the basis of which it is shown that the expressly listed situation(s) differ(s) from the present one. The interpretative argument thought to be most suitable for this task is the teleological argument (Engisch, 1956, p. 144; Boasson, 1966, p. 75-76; Perelman & Olbrechts-Tyteca, 1976, p. 325; Perelman, 1979, p. 23; Soeteman, 1981, p. 355; Peczenik, 1983, p. 57-58; Zippelius, 1985, p. 70; Peczenik, 1989, p. 398; Tak, 1994, p. 282; Henket & van den Hoven, 1996, p. 137; Franken, 2001, p. 191)—the argument by which an interpretation of a legal rule is based on the rule’s intent. Using this kind of interpretative argument also entails that the differences or similarities (the latter suggest analogical application) are judged in light of the goal of the legal rule. But other arguments will do as well, such as the historical or contextual argument. Legal theory’s preoccupation with making the a contrario argument sound is fully justified. However, it has also engendered a conceptual confusion with regard to the a contrario argument. Formerly, it constituted a specific kind of reasoning, viz. ‘because the legal rule denotes X, it is not applicable to Y’. This reasoning is characterized by the appeal to the situation at hand not being mentioned in the legal rule. In the extended version, the a contrario argument can no longer be considered a specific manner of reasoning. That is because—in contrast to what is suggested in the literature—an appeal to relevant differences is not a discriminatory characterisation of a legal argument. By focussing on the actual content of such an appeal, an argumentation theoretical perspective provides two reasons why this is the case. First, an appeal to relevant differences is not an argument on its own account. According to the literature, it can be analysed on the basis of the teleological argument. In such a case the explicit argument is that since the situation at hand does not tally with the rule’s intent, it cannot Henrike Jansen 50 be brought under its scope. Consequently, the appeal to relevant differences can be analysed as an implicit element of the argumentation. Namely, it can be inferred that the rule’s intent reveals a relevant difference between the present situation and the one(s) listed in the legal rule at issue. Thus, the appeal to relevant differences is an implication of the use of an interpretative argument and therefore it is not an argument in its own right. One could object now that that this does not in itself constitute a problem, because when an appeal to relevant differences is always connected to the teleological argument, this makes it, in a way, an argument in its own right. However, it is not true that an appeal to relevant differences can only be inferred from the teleological argument. It can also be analysed when the decision not to apply a legal rule is based on a contextual or a historical argument, or any other interpretative argument. If it is decided, on the basis of whatever argument, that a legal rule does not apply to a situation that it does not mention, it is always possible to make the inference that the situation not listed must (thus) be regarded as relevantly different from the listed one(s). After all, the new situation always differs from the listed one(s) in the light of the interpretative argument that is used. In fact, just as an appeal to relevant differences is not an argument in its own right, neither is an appeal to relevant similarities—alleged to be characteristic of analogical reasoning. Accordingly, such an appeal can always be analysed when it is decided on the basis of whatever interpretative argument that the legal rule can be applied analogically.8 Second, and more importantly, it is a consequence of the above that an appeal to relevant differences is not restricted to a contrario reasoning. Traditionally, the argument a contrario is related to a context in which there is a semantic gap: the context in which the present situation clearly does not coincide with the listed one(s). After all, applying Pitlo’s argument presupposes situation B and C to be semantically distinguished from situation A, otherwise the argument would make no sense. Pitlo’s argument cannot be about interpreting B or C as A; since B and C not being A is the reason for not applying the legal rule to them. Thus, the a contrario argument presupposes well defined notions, as is also noticed by Kitzler (1986, p. 99-100). However, the argument from relevant differences is not restricted to the context of a semantic gap. The same constellation of arguments from which relevant differences can be inferred can be used in other contexts, for example the context in which the legal question is about interpreting the meaning of a vague expression (then the present situation does not clearly belong to or fall out of the scope of the one(s) listed in the legal rule). Or it can be used in the context in which applicability of the legal rule is considered 8 This shows the redundancy of qualifications like ‘appeal to similarities’ and ‘appeal to differences’ for analogical and a contrario reasoning. It is not necessary at all to deduce such an appeal in order for the argument to function: the interpretative argument will do on its own. The scope and soundness of a contrario reasoning 51 to be unacceptable, even though the situation at hand matches the cited one(s) unambiguously (a decision contra legem). I will give an example of a context where the legal question is about the interpretation of a vague expression in the legal rule. The example is taken from a trial about squatting. The legal issue concerned the interpretation of the phrase ‘being in use’, since the legal rule at issue forbids intrusion onto another person’s residential property which is in use (HR 2 February 1971, NJ 1971, 385). The prosecutor defended a broad interpretation of the phrase, which would have rendered squatting within the scope of the prohibition. The High Court disagreed with this view, thereby making use of a teleological argument, from which relevant differences can be inferred between squatting and the situation(s) aimed at by the legal rule: (…) that, even though the aforesaid article 138, insofar as it regards dwelling, in particularly intends to protect inviolability of the home, in conjunction with which in that connection the word ‘being in use’ (…) can only be understood as ‘actually being in use’. [(...) dat toch voormeld art. 138, voorzover op woningen betrekking hebbend, in het bijzonder beoogt het huisrecht te beschermen, in verband waarmede te dezen aanzien de woorden ‘in gebruik’ (...) slechts kunnen worden verstaan als ‘feitelijk als woning in gebruik’.] Since it is the legal rule’s intent to protect the inviolability of the home, squatting (in an uninhabited house) must be regarded as another activity than the ones aimed for—intruding in an inhabited house. Therefore, in the light of the goal of the legal rule, squatting is considered to be different. 9 This example shows an inferred appeal to relevant differences in another context than the context of a semantic gap. The foregoing has shown that the broader concept of a contrario reasoning does not consist in a specific way of reasoning. First, a focus on the actual content of the argumentation showed that an appeal to relevant differences can be deduced from any kind of interpretative argument. As a result, arguments are labelled twice: the a contrario label plus the label of the interpretative argument(s). Second, the appeal to relevant differences can also be analysed in other contexts than those of a semantic gap. In light of the legal perspective of finding solutions for certain types of interpretation problems, it is understandable that the context of a semantic gap is distinguished from a context like the one in which a rule of law contains a vague expression. However, an argumentation theoretical perspective shows this distinction to be 9 That the legal rule’s intent reveals an important difference between squatting and intrusion in an inhabited house is explicitly remarked by Solicitor General Langemeijer (same source). Henrike Jansen 52 irrelevant with regard to the content of the argumentation. So, the broad concept of a contrario reasoning entails a demarcation problem. Interpretative arguments like the teleological, and thus the inferred appeal to relevant differences, can be used in all kinds of contexts: those of a semantic gap, those in which a vague expression has to be interpreted, or even those situations in which the case at hand literally matches the one(s) listed in the legal rule (and in which the decision not to apply the legal rule is a decision contra legem). From an argumentation theoretical point of view no difference exists between reasoning in different contexts, which implies that the argument a contrario cannot be restricted to one of those contexts.10 In my opinion, it is a result of the problems sketched above that the expression ‘a contrario’ is meaningless in the broad conception of the a contrario argument. Literally, ‘a contrario’ means: from opposites (compare note 4). With regard to the linguistic a contrario argument, the reference of this phrase is clear: the legal rule p → q is taken to mean ¬p → ¬q. ¬P is what the legal rule does not mention and ¬q means that the legal consequences of this rule do not follow then.11 With respect to the broader conception of the a contrario argument it is hard to imagine what ‘a contrario’ actually refers to. It could be said to refer to the opposition between applying and not applying the legal rule. However, if we accept this line of reasoning, then every argument put forward for the decision not to apply a legal rule, in any context, can be referred to as a contrario reasoning. Moreover, types of argument are never denoted by their result, but instead they are denoted by the pragmatic content of the inference license that forms the bridge between argument and standpoint. After all, the argument from authority derives its name from someone’s expertise being presented as an argument for accepting his claim. Likewise the linguistic a contrario argument derives its name from the opposition with regard to its being mentioned or not mentioned in the legal rule and the opposed legal consequences related to these possibilities. 10 Earlier Scholten (1974, p. 69-70) reached the same conclusion with regard to the argumentation put forward to defend analogical application or extensive interpretation of a legal rule: no essential distinction exists between the way of reasoning in the one situation or in the other: ‘(…) whether one pays attention to the method or to the data, in both cases analogy and extensive interpretation fully coincide. Analogy as well as interpretation involve seeking a decision by establishing the legal rule’s intent, the higher principle to which the rule of law can be reduced (…).’ Accordingly, Scholten calls the distinction between analogy and extensive interpretation ‘scientifically untenable’ and ‘theoretically wrong’ (p. 72). Also see Groenewegen (1997). 11 The logical problem of this reconstruction and possible solutions are discussed in Jansen (2003a, 2003b) and in the authors mentioned in note 6. The scope and soundness of a contrario reasoning 53 3. Implicit premises in an a contrario argument It should now be clear that the original, linguistic a contrario argument— the argument that the legal rule does not apply to Y, since it applies to X—is unsound, because it is based on the assumption of a complete and coherent legal system. However, I will argue below that a linguistic a contrario argument need not always be rejected as unsound, because it may be supported by certain underlying assumptions (which can, but don’t necessarily, make the argument sound). The literature shows two kinds of situation that encourage a strict interpretation of a legal rule. One of these may very well function as an assumption grounding a linguistic a contrario argument. Or, to put it in terms of argumentation theory: one of these may function as the implicit premise of an argument a contrario. (For that matter, it is possible as well that an arguer makes an explicit appeal to factors that urge a strict interpretation, factors that I will discuss below. This will be shown later with an example. For now, I’m dealing with implicit premises of a linguistic a contrario argument.) First, the implicit premise may be induced by the nature of the legal rule that is at stake. According to legal theory, it is a principle of law that legal rules dealing with matters that make legal security highly important need a strict interpretation. The rules that this principle involves are penal regulations, procedural regulations, regulations that stipulate a time limit and legal rules that constitute an exception to another rule (Baumgarten, 1939, p. 39; Polak, 1953, p. 36; Engisch, 1956, p. 44; Peczenik, 1983, p. 56 ff.; Aarnio, 1987, p. 106; Peczenik, 1989, p. 397 ff.). In those situations, when it is decided that such a rule does not apply to a situation it does not mention, and the formulation of a linguistic a contrario argument is used, the implicit argument may be that the nature of the legal rule allows for such a meagre motivation. A second interpretation of the implicit premise can be based on the fact that the legal rule contains a formulation that urges a strict interpretation of the cases to which the legal rule is applicable. The most obvious indication that the legal rule may not be applied beyond its literal scope are restricting expressions like ‘only’, ‘merely’ and ‘just’ (Klug, 1982, p. 137; Alexy, 1989, p. 280). Fockema Andreae (1904, p. 196), Baumgarten (1939, p. 39) and Soeteman (1981, p. 356) mention the case in which the legal rule at hand contains an explicit enumeration of situations it applies to. Such a detailed circumscription of the conditions under which the legal rule is to be applied might give rise to the assumption that the legislator had the express intention of showing the limitation of its scope. Why is it only these two and no other considerations that may arouse an interpretation of the implicit premise grounding a linguistic a contrario argument? After all, one could ask whether the appeal to relevant differences could also be regarded as the implicit premise. The answer has to do with common ground between the discussion parties Henrike Jansen 54 and with the juridical starting points of the discussion. When the situation is such that the legal rule contains an expression like ‘only’ or when it is of a nature as described above, this is apparent to all discussion parties and can therefore be taken as common ground. Moreover, the legal principles that urge a strict interpretation in those cases are also known to the discussion parties: these are part of the common juridical starting points. In the terminology of argumentation theory we can then say that when a linguistic a contrario argument is used in such a situation, both the content of the premise (e.g.: ‘this is a penal rule’, or: ‘this rule contains the indicator “only”’) and the inference license (e.g.: ‘penal rules should not be applied analogically’, or: ‘rule’s that contain a restrictive expression should be applied restrictively’) are information that is available to both discussion parties. Now my point is that information that is apparent to the discussion parties may play an implicit role in the argumentation. Therefore, both the wording of the legal rule and the rule’s nature, as well as the legal principles stating that those features of a legal rule urge a strict interpretation, may function as an implicit premise in a linguistic a contrario argument. In contrast an appeal to relevant differences may not be taken as a premise in a linguistic a contrario argument. Of course, the principle that similar cases should be treated alike (and dissimilar cases should be treated differently) is a juridical starting point and could therefore obtain the function of an implicit inference license. Nevertheless, that the situation at hand differs from the one(s) mentioned in the legal rule (information that then would have to function as the content of the implicit premise) cannot be taken to be obvious and can thus not be taken to constitute common ground. After all, an appeal to relevant differences could only be inferred when the arguer has explicitly made use of an interpretative argument, like an appeal to the rule’s intent, its position in the legal system, etc. Since the legal rule itself does not provide a clue, it cannot be known which method of interpretation the arguer has chosen. Instead of being common ground, there being relevant differences is something that has to be argued for explicitly. For that matter, the choice of a specific interpretative argument is not a common starting point either. Although it is a common starting point that with regard to the interpretation of legal rules use can be made of interpretative arguments, the choice of a specific one in a specific situation is not. That the two juridical principles discussed above—the legal rule’s restrictive wording and/or its nature—do indeed play a part in legal practice is shown in a decision of the district court of Zwolle (Arr.- Rechtbank Zwolle, 7 January 1970, NJ 1970, 159). In this case such considerations were actually put forward explicitly. In this lawsuit the legal question concerned the competencies of the different ecclesiastical organs and their office holders, all belonging to a certain strict branch of a Dutch Reformed Church in Zwolle (a provincial town in the Netherlands). Its cause was a rift between the church council, consisting The scope and soundness of a contrario reasoning 55 of the vicar(s) and the elders on the one hand, and the diaconate on the other hand. The disagreement centred on the latter’s power of decision. The diaconate was allowed to attend meetings of the council at least once a month and its members held the opinion that they were allowed to vote there. However, this was disputed by the council. In its decision the court referred to an ecclesiastical regulation in which the diaconate’s tasks are listed expressly. Since power of decision is not mentioned, the court took this as a reason not to count that among these tasks: Art. 6 of the “Church Council Regulation” describes the activities in which the diaconate participates in a limited way: [enumeration of activities]. The sole conclusion to be drawn from this is that they lack power of decision in other cases. [Art. 6 der “Regeling Kerkeraad” omschrijft limitatief de werkzaamheden waaraan de diakenen deelnemen: het beroepen van dienaren des Woords, het verkiezen van ouderlingen en diakenen, alle zaken die de stoffelijke zijde van het kerkelijk leven betreffen, en de zaken der zending voorzover die niet behoren tot zaken van opzicht en tucht. Hieruit kan slechts worden afgeleid dat in andere zaken hun geen medebeslissingsbevoegdheid toekomt.] An extra reason for sticking to this restrictive literal interpretation was the procedural status of the regulation: In the event of opinions having altered in this respect, the Ecclesiastical Regulation or the Church Council Regulation should have been amended, yet on an important matter such as the power of decision, deviant custom cannot deprive provisions their force. [Indien de opvattingen hieromtrent veranderd zijn, hadden de Kerkenordening of de Regeling Kerkeraad gewijzigd moeten worden, doch op een zo belangrijk punt als de bevoegdheid van bestuursorganen van een rechtspersoon kan een afwijkend gebruik niet aan bepalingen op dit stuk hun kracht ontnemen.] Having argued that other implicit premises are possible in a linguistic a contrario argument, I will now address the issue of their soundness. Although an assumption based on the wording of a legal rule or on the principle that certain legal rules require a restrictive interpretation may be more rational than an assumption of a complete and coherent legal system, neither of them unconditionally constitutes sound reasoning. Both a certain wording as well as regulatory status may urge a strict interpretation, but they do not forbid analogical interpretation. As for the formulation of the legal rule: even expressions that are supposed to reflect the legislator’s intentions may be overruled. And it may even be questioned whether wording shows the legislator’s intention at all in the Henrike Jansen 56 case of an enumeration, since, as van Bemmelen (1891, p. 17) indicates, these can be meant to be limitative but illustrative as well. With regard to a strict interpretation of a legal rule caused by its regulatory status: only when penal law is in discussion is analogical application never allowed. In all other cases it is mainly careful interpretation that is incited. With regard to analogical application of procedural rules and exceptional provisions Aarnio (1987, p. 505) talks of ‘great care’ that has to be aimed for, not about it being disallowed. According to Engisch (1956, p. 148) and Peczenik (1989, p. 398-400) strong reasons may allow for analogical application in case of a rule that is an exception to another rule. In conclusion: the formulation and the nature of a legal rule require notice of the legislator’s intentions and the principle of legal security. The next step is to consider these in the light of other arguments: ‘It depends on the priority order one assumes in the concrete case between justice and legal certainty’ (Peczenik, 1983, p. 57). The decision of the district court of Zwolle evidently favours legal security. However, in other cases the decision may be the opposite (see, for example, HR 13 March 1992, NJ 1993, 96, where, according to the Solicitor General, the High Court favours an analogical application, as the legislator had not closed a gap in the law even after 20 years). 4. Conclusion A contrario reasoning has a notorious reputation. In this article I have shown that the argument’s bad name originates in the way the argument was originally understood. The a contrario argument used to be a linguistic argument, based on the unmistakably clear words of a legal rule. It reasons that because the situation at hand is not listed in those words, it does not belong to the legal rule’s scope. The argument is based on the assumption, which functions as its implicit premise, that the legal system is complete and coherent—a view that no longer commands support. Because of this, legal scholars wish to warn their readers not to use the unsound a contrario argument, and they do so by encouraging an arguer to provide additional evidence. However, the problematic side- effect of this—otherwise justified—goal, is a new conception of a contrario reasoning, which is so broad that it no longer applies to a specific way of reasoning. As a result, the expression ‘a contrario’ has become a meaningless label. From the point of view of argumentation theory, it is irrational to distinguish the new concept of the a contrario argument from argumentation that is brought forward in order to defend a restrictive interpretation of a vague expression or from contra legem- reasoning. In this article, I propose to use the label ‘a contrario argument’ in its original understanding: an appeal made to the explicit wording of a The scope and soundness of a contrario reasoning 57 legal rule that does not cite the situation at issue. Such an argument is not fallacious per se, although it would be if the assumption of a complete and coherent legal system is the only possible interpretation of the implicit premise. However, in two kinds of situations an a contrario argument may rest on another implicit premise. First, if the legal rule contains a restricting indicator like ‘only’, or a detailed enumeration, the implicit argument may be that these reveal the legislator’s intent to restrict the rule’s scope. Second, the legal rule’s procedural status may allow for a strict application of its scope. Unfortunately, neither interpretation of the implicit argument renders the a contrario argument automatically sound, for these incitements of a strict interpretation must be weighed against other arguments that favour analogical interpretation. Without additional evidence, the decision not to apply the legal rule therefore remains a meagre proof. Nevertheless, if we want to prevent any misunderstanding, such additional evidence must not be referred to as an a contrario argument. References Aarnio, A. (1987). The rational as reasonable. A treatise on legal justification. Dordrecht: Reidel. Alexy, R. (1989). A theory of legal argumentation. The theory of rational discourse as theory of legal justification. Oxford: Clarendon Press. Baumgarten, A. (1939). Grundzüge der Juristischen Methodenlehre. Bern: Hans Huber. Boasson, Ch. (1966). The use of logic in legal reasoning. Amsterdam: Noord-Hollandse Uitgeversmaatschappij. Bydlinski, F. (1991). Juristische Methodenlehre und Rechtsbegriff. Wenen/New York: Springer. Canaris, C.W. (1983). Die Feststellung von Lücken im Gesetz. Eine methodologische Studie über Voraussetzungen und Grenzen der richterlichen Rechtsfortbildung praeter legem. (2nd Rev. Ed.) Berlin: Duncker & Humblot. Diephuis, G. (1869). Het Nederlandsch Burgerlijk Regt, 1e deel. Groningen: Wolters. Engisch, K. (1956). Einführung in das Juristische Denken. Stuttgart: Kohlhammer. Esser, J. (1949). Einführung in die Grundbegriffe des Rechtes und Staates. Eine Einführung in die Rechtswissenschaft und Rechtsphilosophie. Wenen: Springer. Feteris, E. T.: (1999). Fundamentals of Legal Argumentation. A Survey of Theories of the Justification of Legal Decisions, Kluwer, Dordrecht. Fikentscher, W. (1975). Methoden des Rechts. In vergleichender Darstellung. Band I: Frühe und religiöse Rechte, Romanischer Rechtskreis. Tübingen: Mohr (Paul Siebeck). Henrike Jansen 58 Fockema Andreae, J.P. (1904). Tien jaren rechtspraak van den Hoogen Raad. Bijdrage tot de leer der wetsuitlegging. Leiden: Van Doesburg. Franken, H. et al. (2001). Encyclopedie van de rechtswetenschap. (9th Ed.) Deventer: Gouda Quint. Germann, O.A. (1946). Methodische Grundfragen. (6th Ed.) Basel: Verlag für Recht und Gesellschaft. Groenewegen, F.T. (1997). ‘Analogie’. In: E.T. Feteris, H. Kloosterhuis, H.J. Plug & J. Pontier (red.), Op goede gronden. Bijdragen aan het tweede symposium juridische argumentatie. Nijmegen: Ars Aequi Libri, pp. 108-114. Henket, M. (1992a). ‘On the logical analysis of judicial decisions’. International Journal for the Semiotics of Law 5, p. 154-164. Henket, M. en P.J. van den Hoven (1996). Juridische vaardigheden in argumentatief verband. Groningen: Wolters-Noordhoff. Hoecke, M. van (1979). De interpretatievrijheid van de rechter. Antwerp: Kluwer. Horovitz, J. (1972). Law and logic. A critical account of legal argument. Vienna/New York: Springer. Jansen, H. (2003a). ‘A contrario reasoning and its legal consequences’. In: F.H. van Eemeren, A. Snoeck Henkemans, J.A. Blair & Ch.A. Willard (Eds.), Proceedings of the fifth conference of the International Society for the Study of Argumentation. Amsterdam: Sic Sat. Jansen, H. (2003b). Van omgekeerde strekking. Een pragma-dialectische reconstructie van a contrario-argumentatie in het recht [Inverted Pupose. A pragma-dialectical reconstruction of e contrario argumentation in law] PhD-thesis University of Amsterdam. Amsterdam: Thela Thesis. Kaptein, H. (1991). ‘Logica in rechtspraak, in het bijzonder in analogie- en a contrarioredeneringen: vroom bedrog?’ In: F.H. van Eemeren & E.T. Feteris (Eds.), Juridische argumentatie in analyse. Opstellen over argumentatie en recht. Groningen: Wolters-Noordhoff, pp. 75-90. Kaptein, H. (1993). ‘A contrario arguments in law: From interpretation to implicit premises’’. International Journal for the Semiotics of Law 6, p. 315-324. Kitzler, A. (1986). Die Auslegungslehre des Anton Friedrich Justus Thi- baut. Berlin: Duncker & Humblot. Klug, U. (1982). Juristische Logik. (4th Ed.) Berlin etc.: Springer. Land, N.K.F. (1910). Inleiding tot de verklaring van het Burgerlijk Wetboek. Haarlem: De Erven Bohn. MacCormick, D.N. & R.S. Summers (Eds.) (1991). Interpreting statutes. A comparitive study. Dartmouth etc.: Aldershot. Meier Hayoz, A. (1951). Der Richter als Gesetzgeber. Eine Besinnung auf die von den Gerichten befolgten Verfahrensgrundsätze im Bereiche der freien richterlichen Rechtsfindung gemäss Art. 1 Abs. 2 des schweizerischen Zivilgesetzbuches. Zürich: Juris. The scope and soundness of a contrario reasoning 59 Nieuwenhuis, J. (1976). ‘Legitimatie en heuristiek van het rechterlijk oor- deel’. Rechtsgeleerd Magazijn Themis 38, p. 494-515. Opzoomer, C.W. (1849). ‘Verdediging der aanteekening op de algemeene bepalingen der wetgeving van het koningkrijk tegen J.W.A.R.’. In: Nieuw Jaarboek XI, Amsterdam: Müller. Peczenik, A. (1989). On law and reason. Dordrecht etc.: Reidel. Perelman, Ch. & L. Olbrechts-Tyteca (1976). Traité de l'argumentation. La nouvelle rhetorique. (3rd Ed.) Brussels: Université de Bruxelles. Perelman, Ch. (1979). Juridische logica als leer van de argumentatie. Antwerp/Amsterdam: Standaard Wetenschappelijke Uitgeverij. Pitlo, A. (1995). Het systeem van het Nederlandse privaatrecht naar het Nieuwe Burgerlijk Wetboek. (11th Ed., rev. by P.H.M. Gerver, H. Sorgdrager & R.H. Stutterheim, ass. by T.R. Hidma) Arnhem: Gouda Quint. Schneider, E. (1965). Logik für Juristen. Die Grundlagen der Denklehre und der Rechts-anwendung. Berlin/Frankfurt am Mein: Franz Verlag Vahlen GmbH. Scholten, P. (1974). Mr C. Assers handleiding tot de beoefening van het Nederlands burgerlijk recht. Algemeen deel. (3rd Ed., suppl. by Mr. G.J. Scholten; 1st Ed. 1931) Zwolle: Tjeenk Willink. Soeteman, A. (1981). Norm en logica. Opmerkingen over logica en ratio- naliteit in het normatief redeneren met name in het recht. Zwolle: Tjeenk Willink. Tak, P.J.P. (1994). Rechtsvorming in Nederland. Een inleiding. Een inleiding. (4th compl. rev. ed., ass. by F. Kusters) Alphen aan den Rijn: Samsom Tjeenk Willink. Tarello, G. (1972). ‘Sur la spécificité du raisonnement juridique’. In: Die juristische Argumentation ARSP, Beiheft 7. Wiesbaden: Franz Steiner, pp. 103-124. van Bemmelen, P. (1891). Regtsgeleerde opstellen 1. Leiden: Brill. Zippelius, R. (1985). Einführung in die juristische Methodenlehre. Munich: Beck.