An empirical supposition underlying our interest in relevancy is that people did, do and can make distinctions which are irrelevant and that they could or can decide not to make these distinctions, or to make only distinctions which they believe to be relevant (and which, ideally speaking, are relevant). The endeavor to analyze relevancy presupposes furthermore that for quite a number of people relevance actually does play a role as a principle, not only in conversational cooperation, but also with respect to the making of distinctions (assuming that the former would not amount to the making of distinctions as well). That is why we cannot divorce the question of the meaning and criterions of relevancy from the formulation of its principle. Two extremes to be avoided are to require too little, so that everything is relevant or can deceitfully be 'made' relevant, or to require too much, so that no material distinction is or can be taken to be relevant. This would contradict our basic, existential postulate and annihilate the principle of relevance altogether. With it, it would exhaust such notions as those of equality and fairness or justice of all meaning insofar as they depend on the relevancy of making distinctions.

Two general criterions playing a role in any criterial theory or formulation of a definition are consistence and noncircularity. Taking consistence as a test of relevancy, a distinction on the basis of a certain factor which is claimed to be relevant has to be made consistently all the time in respect of the same focus. For example, if income is a relevant factor with respect to the amount of taxes to be paid, then it is a relevant factor at all times for all taxpayers (possibly together with other factors). Consistence itself, however, does not prove relevance. A government, for instance, may consistently exclude all those who do not speak and write the 'standard' language from official positions, yet this in itself does not make the distinction of language (dialect, sociolect, idiolect or spelling) relevant with respect to the work to be performed in these positions. This limitation is not different from that of a coherence test of truth. In other words: consistence is not a criterion of relevance in the sense of a sufficient condition; it is only inconsistence that is a criterion of irrelevance in this sense.

Many a traditional theorist has believed that the dictum that 'equals should be treated equally and unequals unequally', or that 'similar cases should be treated in a similar fashion' would express a principle of justice. This is a grave mistake, unless justice does not mean more than consistence. But if justice is defined in such a way that the distinctions drawn must also be relevant, or that exceptions made must be made on relevant grounds, treating equals equally and unequals unequally does not guarantee relevance and justice by any manner of means. Only the reverse may be true: if the making of the distinction or exception is relevant, then those who are equal on each side of the divide should be treated equally. A classical philosopher may thus treat all 'er compatriots (the equals) in exactly the same fashion and all 'barbarian aliens' (the unequals) also in exactly the same fashion, yet this does not prove that the distinction between compatriots and aliens would be relevant, and that it would be just to treat aliens in another way than compatriots, even when done consistently. Such reasoning is fallacious, because it confuses consistence as a necessary with consistence as a sufficient criterion of relevance, and indirectly of justice.

Whereas it is relatively clear what consistence and inconsistence mean, the notion of 'circularity' is a much vaguer one. Strictly speaking, we are faced with circularity when the relevancy of a distinction depends truth-conditionally on the relevancy of this distinction itself (or on the belief in the relevance or irrelevance of this distinction). We will deal with this case when the difference is made within the focus of relevancy itself. However, if the relevancy of a distinction is dependent on another kind of irrelevant distinction drawn within the focus, or drawn in a (discriminatory) attitude or practise with which it is correlated, the circularity is there from the point of view of relevancy in general but, perhaps, not from the point of view of one person making or not making a distinction. Circularity is still too vague a concept in a disciplinary environment which is not purely truth-conditional to be used itself as a criterion of relevancy, altho it plays a role in a number of criterions which are of a more substantive type than inconsistence.

A criterion of irrelevance is a means to falsify judgments of relevance. The five criterions to be proposed in this division for the falsification of judgments of discriminational relevance are:

1. the distinction made is not consistently applied;
2. the focus (goal or other directional entity on which the relevancy depends) is fake;
3. the focus is genuine but the relevancy with respect to this focus is based on a correlation or difference in correlations only (like in the case of statistical relevance), and therefore shows at the most the possibility of discriminational relevance with respect to groups or categories, and the chance of this relevance with respect to persons or members of these categories;
4. the focus is genuine but the relevancy with respect to this focus depends on the existence (past/present/future; real/expected) of an (other) attitude or practise in which a nonrelevant distinction is made by the person 'imself or the persons themselves;
5. as 4., but now by others.

The first criterion does away with cases of partial relevancy. The second criterion should do away with cases of fake relevancy, the third with cases of pseudofactual relevancy, the fourth and fifth (if accepted) with cases of circular relevancy. By nonrelevant we shall mean what is obviously irrelevant, partially relevant, purportedly relevant in respect of a fake focus, pseudofactually relevant or circularly relevant. Moreover, we shall use the word determinant as a generic term for criterion (of relevancy), focus (of relevancy) and factor (of distinction).