196. Is authority in its "basic structure" a triadic relation . . .

PDF Version of this Document

Is authority in its "basic structure" a triadic relation between the bearer, the subject, and the domain of authority (as Bochenski holds); or is it, rather, a quadratic relation between the bearer, the subject, and the domain of authority and the reason for it (as De George suggests—more clearly, however, in "The Nature and Function of Epistemic Authority" than in The Nature and Limits of Authority!)?

Perhaps in the case of de facto authority, the basic structure of authority may seem to be merely triadic. But if it is correct, as I have argued, that "de facto authority presupposes de jure authority," in that no one can either purport to be or to have authority or be accepted as such without making or implying the claim that the authority is valid, then the fourth factor that I speak of as the reason for authority can hardly be ignored.

I incline, then, to share De George's view that "[a]ll authority is ... essentially a relation among a bearer, a subject, and a field, in virtue of a particular quality, attribute, or context. This latter component supplies the justification for the legitimate use of authority, and it varies from type to type. There is no one thing or quality which makes authority legitimate, and to search for any such component is to search in vain. The core relation of authority can be put formally by saying that 'A is an authority for B over field C in virtue of D' (or in some contexts authority is the right or power of A to do B with respect to some field C in virtue of D)" ("The Nature and Function of Epistemic Authority": 77; italics added).

(It's clear enough, I take it, that De George's phrase "in virtue of" could be.replaced with "because of," or "by reason of," without in any way changing what he means to say. In fact, at least one dictionary I consulted defines "by or in virtue of' with "because of' and "by reason of. ")

It was the fourth factor of the reason for authority to whose importance I was concerned to draw attention when I wrote that "all de jure authority is by logical necessity 'rule-conferred authority'" and that, therefore, "there can be de jure authority at all only where there is a preexisting rule or rules that confer it as a right and so authorize it" (OT: 49 f.). As I now look back on such statements, I suspect they are vulnerable to the objection that they "overload" the notion of a rule (so Watt) and that their point could be better expressed by speaking simply of the reason for the authority in question. Of course, if there is a reason for the authority, one may not inappropriately say that there has to be a rule, implicit if not explicit, governing claims made or implied concerning the authority by either the bearer or the subject of it.

9 July 1996