2. About a Definition’s Second Condition

196. One must consider then the second condition [n.187], namely per se unity, whether it prevents there being a definition properly speaking of a sacrament, or of what has the account of this sort of name. - But that a being that is not per se one is not definable can be understood in two ways, namely that it consists of beings either of the same genus or of different genus that however are not of a nature to constitute something per se one. The first has the name of an aggregated being (as a mound or heap), and the second is properly called a being per accidens, as is plain from Metaphysics 5.6.1015b16-34. But neither of these is properly definable: not the first, from Metaphysics 8.6.1045a7-25, and not the second, from Metaphysics 7.4.1029b22-30a7.

197. It is in fact said [Richard of Middleton] that a sacrament does not properly have unity, so it is not properly definable. For it includes many things (as is touched on in the first argument to the opposite [n.175]), from which something per se one cannot come to be, namely an element (as water in the case of baptism) and spoken words (and these two are material parts) and the idea of signifying (as something formal): the first of these is a real being, the second a being of reason. From such things it is impossible for something per se one to come to be.

198. But against this: for although one accident may be in many subjects, yet it would be definable properly just like other accidents, because the manyness does not pertain per se to the idea of the accident but is as it were something added. Now in the aforesaid idea of the name [n.194] it is plain that what is called ‘sensible’ is posited as an addition to the sign. So however much there is no unity in it, yet while the other things that pertain to the formal idea of the name do not prevent per se unity, a sacrament will not for this reason be non-definable. I mean that the plurality under discussion here [n.197] is a plurality of things that come together in the sensible thing as the sensible thing signifies the foundation of the formal idea that a sacrament involves.

199. I say therefore that in the aforesaid definition [n.194] the formal element is understood to be the sign and also to be the things that per se determine the idea of a sign. Of such sort are ‘by divine institution’ and ‘efficaciously’ (the other two there, namely ‘sensible’ and ‘gratuitous effect of God’, are there as additions: the first as subject or foundation, the second as correlative of the sign). But a plurality of accident with subject, or of the subject in itself, or of correlative with correlative does not prevent the relation from being definable simply. Therefore a sacrament is not excluded by non-unity from having a definition. For this concept ‘per se one in the intellect’ is as conventional and efficacious a sign as is the concept of paternity. And just as paternity could be properly defined (notwithstanding per se unity) if paternity were in two subjects and these two were posited as additions and the correlative of father were posited as an addition, so too in the issue at hand.