著者

神崎 繁

出版者

日本西洋古典学会

雑誌

西洋古典学研究 (ISSN:04479114)

巻号頁・発行日

vol.32, pp.41-53, 1984

---

Many scholars treat the first part of the dialogue (384 c-391 b) as the exposition of Plato's own views about the correctness of names because of its similar phraseology to that of the Republic X and its critical remarks on the Protagorean background of Hermogenes' conventional theory of names. But this treatment not only drops the demarcation between Plato's position and Cratylus', but also prevents them from understanding the overall structure of the dialogue. For, as Bernard Williams recently revealed in his brief but excellent paper, it is Cratylus' natural theory of the correctness of names that Socrates attempts to elaborate with Hermogenes in the next part (391 b-427 d) and then he sets about to refute with Cratylus himself in the final part (427 d-439b). According to my interpretation, however, Plato's strategy is already mentioned to some extent in the first part of the dialogue. Because a name-giving role personified as νομοθετη&b.sigmav; is said not only to precede a name-using role as διδασκαλικο&b.sigmav; but also to be supervised in turn by the dialectician, διαλεκτικο&b.sigmav;, who uses names for discussion. And I suppose these three personified figures to represent the positions of Cratylus, Hermogenes and Socrates respectively. So the main theme in succeeding arguments of the dialogue as well as the dialogue itself seems to be under dialectical supervision. This is a noeud initial of the dialogue. I suggest, therefore, that the notorious etymological section in the second part displays various deviational forms of name-relation from such integral linguistic practice with names as the dialectic. For, though we find in the text somewhat confusing expressions with the same verb 'δηλουν' (to show) , e.g. 'someone shows something by a name.', 'a name shows something to us.', 'a name shows something.', etc., we can presume from them an underlying structure: A shows p with n to B (where A and B are persons, p is a thing, and n a name). This underlying 4-place relation is transformed into the various expressions found in the text without changing the meaning of 'show' in it. But the etymological explanation urges us to think the verb 'show' equivocal like this: A legislator as a name-giver showed^1 something with a name, from which descended name shows^2 it to us. Thus, etymological study turns out to be nothing less than research for original intentions or thoughts of the name-giver-they are usually expressed by a verb 'βουλεσθαι'-, for such intentional factors take advantage of the time- and semantic gap between giving names and deciphering their descendants. But, of course, research for such factors may be uncertain and arbitrary. To do without it, there must be a direct relation between names and things : 'a name shows^3 something'. Being interpreted by similarity, this binary relation though it is also transformed from the same 4-place relation above gets another meaning of 'show': 'a name is a copy of something'. It is this final position of Cratylus' natural theory that is to be refuted in the third part of the dialogue. If we demand the correctness of names for its own sake i.e. apart from our integral linguistic practice, we cannot but rely on either the correspondence between language and reality or tacit consent among people without reference to our understanding the reality whatever we acknowledge it to be. So we must try to understand the truth (that is not the same thing as the correctness anymore) with names and the other linguistic components in the dialectical integrality. This denouement seems to have been already anticipated in the noeud initial of the dialogue.