It seems that theorists on relevancy have usually not been familiar with the publications in other disciplines or subdisciplines on the same subject. Or perhaps, they reasoned too readily that 'their' relevancy had nothing to do with the other notions of relevancy. If this were correct, however, we could not explain very well how anyone when confronted with one of these notions could ever have a less than vague, intuitive idea about what a specific kind of relevancy (such as moral relevancy) might denote -- and it is not said "connote". It is hard to imagine that someone would treat truth in this way, and suggest, for example, that scientific truth, pragmatic truth and religious truth would all be entirely isolated, primitive notions having nothing in common in denotation or connotation. For again, if this were correct, no-one could even have the vaguest, intuitive idea of what these phrases could mean if 'e were confronted with them for the first time. Therefore there is enough reason to assume some underlying structure -- a common skeleton, albeit differently fleshed out for the several forms of relevancy (and truth). Of course, we take it that the underlying unity is to embrace more than some common positive connotation of relevance and negative connotation of irrelevance eagerly exploited in academic debate.

Let us take goal-dependent relevancy as a primitive notion, as the prototype of all relevancy. Two entities are then related if they are relevant in respect of the same goal (same in the sense of identical). On this construction the notions of symmetrical relevancy or relatedness are derivative and presuppose the existence of ordinary, goal-dependent relevancy. (Theoretically the relatedness might also be taken as primitive and the common relevancy in respect of a certain goal as derivative.) This conceptual picture 'explains' the reflexivity of symmetrical relatedness, and shows also why nontransitivity should be required. Pragmatic relevancy or practical relatedness is from this point of view a particular kind of goal-dependent relevancy, whereas semantic relevancy or relatedness is a kind of (symmetrical) 'relevance-relatedness'. Pragmatic relevancy in a narrow sense differs from other forms of goal-dependent relevancy in that the fundament of the relation is a speech-act; in a broad sense, in that the fundament is an act in general. Semantic relevancy differs from other forms of relevance-relatedness in that the entities related are propositions.

Given this sketchy representation of the overall structure in which goal-dependent relevancy remains the basic notion, it is easy to locate discriminational relevancy. Because of the adjective it is the relevancy of a distinction, for example, on the basis of age or nationality. The question is every time whether the factor age or the factor nationality is relevant in respect of something valued high, such as the (best) quality of work to be done, or the (lowest possible) price of a ticket to be paid. In questions of discrimination we are not just interested in a statistical relevancy or other kind of symmetrical relatedness between one distinction we could make and another distinction we could make. We are interested in only one distinction or cluster of distinctions at a time, and want to know whether it is relevant with respect to a value which those concerned have acknowledged as something to strive for. The distinction itself is not defined as something to aim at -- otherwise the question becomes superfluous. It follows that the relevancy of discrimination is a kind of goal-dependent relevancy with a distinction or cluster of distinctions as relational fundament and a goal or other directional entity as relational terminus.

Figure I.5.2.3.1 gives a schematic representation of most of the conceptions of relevancy dealt with in the philosophical literature in such a way that the general structure shows. All these forms of relevancy mentioned in the literature have been listed in figure I.5.2.3.2, as well as a few related relevancy concepts. They are presented in such a way that it should become clear on the basis of what characteristics they are primarily (to be) classified. It goes without saying that this scheme is an idealization. The actually used notions of relevancy are often of a mixed character. Thus the primary classification of discriminational relevancy may be on the basis of the type of fundament (a distinction or set of distinctions) but this does not mean that the criterions applied to the relation of discriminational relevancy will not be equally typical. Similarly, the primary classification of moral relevancy is on the basis of the type of terminus (a moral goal or value), so it seems, but the criterions of moral relevancy probably also differ from those of all other forms of relevancy. The question of what goal(s) moral philosophers actually choose, and of what criterions they actually apply for moral relevancy (if any), is not our present concern.