4 Interpretation and modi fi cation
4.3 Language, facts and beyond – further confusion?
4.3.1 Language and law – semantic uncertainty
If we proceed from the assumption that written norms are language – even if we only claim that language is the manifestation of norms – the question of the limits of a natural language’s performance becomes relevant in ascertaining the limits of interpretation. Language is not something having inherent value or giving a fixed reference to extrinsic concepts. It is not an unchangeable entity. A word can mean anything we choose it to mean. After all, it is only a word, as a classical passage from Through the Looking-Glass (1871) illustrates.
‘I don’t know what you mean by “glory,” ’ Alice said.
Humpty Dumpty smiled contemptuously. ‘Of course you don’t – till I tell you. I meant
“there’s a nice knock-down argument for you!” ’
‘But “glory” doesn’t mean “a nice knock-down argument,” ’ Alice objected.
‘When I use a word,’ Humpty Dumpty said, in rather a scornful tone, ‘it means just what I choose it to mean – neither more nor less.’
‘The question is,’ said Alice, ‘whether you can make words mean so many different things.’
‘The question is,’ said Humpty Dumpty, ‘which is to be master – that’s all.’160
Language can only do so much. Scholars – in particular some analytical philo- sophers161 – argue that its powers are limited. ‘In all fields of experience . . . there
160 Lewis Carroll, Through the looking-glass and what Alice found there (1871), in: Martin Gardner (ed.), The annotated Alice. The definitive edition (2000) 224.
161 Bernd Schünemann, Die Gesetzesinterpretation im Schnittfeld von Sprachphilosophie, Staats- verfassung und juristischer Methodenlehre, in: Günther Kohlmann (ed.), Festschrift für Ulrich Klug zum 70. Geburtstag (1983) 169–186 at 170.
4.3 Interpretation and modification 117
is a limit, inherent in the nature of language, to the guidance which general language can provide.’162 Indeterminacy of language results in uncertainty of law – ‘the uncertainty which is inherent in certain legal concepts is due to the uncertainty of their expression’163 – and thus we cannot know law because language cannot be ultra-precise.
What are the reasons for indeterminacy? Michael Thaler distinguishes between two phenomena: ambiguity or equivocation (Mehrdeutigkeit) and vagueness (Vagheit). Ambiguity is a lack of determination of a term’s connotation (Bedeutung).
The connotation is determined a priori by a language’s rules of semantics.164 A word can connote two entirely different things: ‘star’ is a person and a heavenly body, and its context cannot always determine the ‘correct’ meaning for us.165 In contrast, vagueness is the lack of determination of a term’s denotation (Bezug).
The denotation is determined by the class of objects properly signified by the term, based on empirical experience. Thaler argues that few scholars distinguish between the two phenomena,166 even though the connotation–denotation distinc- tion is as old as semantics itself. Among jurists it seems that linguistic indeterminacy is identified exclusively with vagueness. Herbert Hart, for example, writes that ‘[p]articular fact-situations do not await us already marked off from each other, and labelled as instances of the general rule’167 which clearly refers to a term’s denotation only.
The particular problem of vagueness will be discussed below (Section 4.3.2), while linguistic indeterminacy on the general plane will be discussed further here.
The notion of ‘connotation’ is moot if ‘the meaning of a word is its use in the language’,168 as Ludwig Wittgenstein held – though he did not define meaning exclusively by reference to use. Is a ‘norm of meaning’ thus established by mere fact, i.e. by mere behavioural regularity, or is it perhaps a customary norm? Is the meaning of a word established not as norm, but as mere induction from instances of use, which would change with changed use? This would do two things: words
162 Hart (1961) supra note 1 at 123 (emphasis added).
163 Hummer (1975) supra note 94 at 88.
164 Thaler (1982) supra note 114 at 2 uses the following example: ‘We can say that the terms “bachelor”
and “unmarried man” are synonymous without an empirical enquiry whether any bachelors are married, merely by knowing the rules of semantics. This argument is not quite convincing, since the equivalence here is based on logic. “Unmarried bachelor” would be a mere analytical statement, because the “bachelor” logically encompasses “unmarried man”. But Thaler is correct, for the definition of a word (which is his Intension or Bedeutung) is defined not by logic; however, how are we to hinder the connotation’s collapse into the denotation?’
165 Thaler’s example in German ‘der Star’ has three partially differing meanings to its English pendant: (a) the eye diseases ‘glaucoma’ or ‘cataract’ are called ‘grüner Star’ or ‘grauer Star’, respectively, (b) a starling or (c) a film or pop star. When one says, thus: ‘Der Star singt’ (‘The star is singing’), the ambiguity is not context-sensitive as it would be in English. Thaler (1982) supra note 114 at 5 (FN 31).
166 Thaler (1982) supra note 114 at 8.
167 Hart (1961) supra note 1 at 123.
168 ‘Die Bedeutung eines Wortes ist sein Gebrauch in der Sprache.’ Ludwig Wittgenstein, Philoso- phische Untersuchungen (1953) 20 (para 43) (translation in the text: G.E.M. Anscombe at 20e).
Uncertainty in International Law 118
would only have a denotation (Begriffsumfang), no connotation (Begriffsinhalt) and there could not possibly be one or more correct or possible meanings.
[Wittgenstein] seeks to transform semantics from the science of meaning or meanings into the science of signifying activity. He rejects the traditional idea of linguistic forms and their meanings as classes of entities which are correlated among themselves . . . He seeks to impose a new vision in which linguistic forms have the meaning they do because they are used by man, the guarantee of their validity being found only in their use.169
The question is what ‘meaning’ means. What does the text of a statute refer to?
Does it refer to facts? – How can it? A norm refers to an Ought, not an Is!
How can it not? All norms themselves refer to human activity, which is an Is! Does the text of a norm refer to a norm as its transcendent ‘meaning’, as Guastini would argue (Section 4.2)? But how can this be, if the term’s extension is classes of facts and hence presupposes a subsumption, which can only happen after (in a logical, not temporal sense) the norm is cognised (Section 4.3.2)?
This indeterminacy is the result of our use of natural languages to construct norms. Waldemar Hummer argues that the ‘uniform designation of legal con- cepts as a result of a standardized colloquial usage undergoes continuous change through the influx of subcultural neologisms and other semantic “deviations” ’.170 Indeed, ‘experience shows that the context may determine a complete change in the meaning of a word.’171 Unlike pure logic and unlike the unattainable goal of a
‘mathematical language’, the meanings of words in colloquial language are flex- ible.172 To summarise: language has limits; it is arbitrary, conventional, fluctuating, subjective and imprecise and hence it produces uncertainty.
Do the arguments of analytical philosophy regarding linguistic indeterminacy square with the Vienna School’s theory of interpretation? Kelsen explicitly rec- ognises the weakness of language as a factor leading to the denial of one correct meaning.173 Hans-Joachim Koch writes that ‘[t]he limits of linguistic precision of positive law postulated by the Pure Theory of law can be refined by analy- tic philosophy by using the theory of semantic indeterminacy’,174 hence the Pure Theory remains current even taking into account later developments in philosophy.
169 Hummer (1975) supra note 94 at 146.
170 Hummer (1975) supra note 94 at 88.
171 Hummer (1975) supra note 94 at 137.
172 Larenz (1991) supra note 3 at 312.
173 Cf. Bulygin (1995) supra note 141 at 13–14; contra Luzzati (1995) supra note 153 at 130–132.
174 ‘[d]ie von der Reinen Rechtslehre postulierte Grenze sprachlicher Bindungskraft des positiven Rechts lọòt sich sprachanalytisch durch die Lehre von den semantischen Spielrọumen prọzis- ieren.’ Hans-Joachim Koch, Die Auslegungslehre der Reinen Rechtslehre im Lichte der jüngeren sprachanalytischen Forschung, 17 Zeitschrift für Verwaltung (1992) 1–8 at 7.
4.3.1 Interpretation and modification 119
Analytical philosophy, however, can only provide an approach to explain uncertainties of interpretation as a result of linguistic indeterminacy,175 not as a result of other causes for the uncertainty of the cognition of law. Not all uncertainty results from the open texture of language itself. The law-maker may have – even inadvertently – included elements making cognition uncertain, e.g. discretion by organs, dilatory formulas particularly popular in multilateral treaties, or a simple lack of text (Chapter 2).