<<   >>   TOC

5. Truth of a name

         If Mills’ statement that “names have denotation, not connotation” was correct, it would be utterly nonsensical to speak of truth in connection with names. But, whether logicians like it or not, the meaning of names transcends mere labelling. Speaking anecdotally, I named my daughter Isabel and not XY23FO4, even though the latter would almost guarantee unequivocal identification anywhere in the world and under any circumstance, while Isabel is a comparatively common and thus potentially ambiguous name. Nevertheless, the name Isabel has the important advantage that particular groups of people that are likely to become relevant for my daughter throughout her life immediately accepted this name as ‘normal’, ‘fitting’, ‘traditional’, but also ‘original’. Levi-Strauss (1962) provides ample evidence that names supply the individual with a place in society and that they are chosen specifically because of their connotations. For Frege (1892) it is, of course, clear that any lexical item will have connotations, since it is necessarily part of a larger language system.[1]

        Levi-Strauss (1962) argues further that the distinction between proper names and natural kind terms is artificial and shows that both form a continuum: proper names are thus used as the narrowest possible classifications, i.e. they denote natural classes with ideally one member. Since proper names and natural kind terms are conceptually related, it comes as no surprise that the latter are frequently used as proper names, while the former are often expanded into class terms. Again, examples abound and are provided in Levi-Strauss (1962).

        The Russellian consequence is to treat proper names and common nouns alike as abbreviations for definite descriptions. If this was correct, an identifying description should be substitutable for a name. As Burge (1973: 203) shows, this is not true for even Russell’s minimal solution “the object called PN”, where PN stands for the proper name in question: (14) is not equivalent to (15).

(14)             Jones is necessarily a Jones.

(15)             This entity called ‘Jones’ is necessarily an entity called ‘Jones’. 

Since a similarly contradictory statement can be produced for any substituted definite description, Burge (1973) thus opts for viewing proper names simply as predicates, leaving the exact implementation of truth conditions up to extra-linguistic factors. The name O’Hara would thus be interpreted as the predicate O’Hara(x), which has the truth condition “true of any x just in case x is an O’Hara” (206). I will assume here that this accurately describes the semantic contribution of proper names.

<<   >>   TOC

© Philipp Strazny 1998


[1]           It is often discussed whether proper names are part of a language or whether they transcend any given language system. I believe this misses the point, since any name in use will stand in a relation to the remaining language system used by the speaker. Thus, Frege’s assertion also applies to e.g. a Japanese name used by an English speaker conversing in Italian.