9. It follows that in the cases specified above – agent, knowledge and the rest – the relation must be considered as in actual operation, and the Act and the Reason-Principle in the Act must be assumed to be real: in all other cases there will be simply participation in an Ideal-Form, in a Reason-Principle.
If Reality implied embodiment, we should indeed be forced to deny Reality to these conditions called relative; if however we accord the pre-eminent place to the unembodied and to the Reason-Principles, and at the same time maintain that relations are Reason-Principles and participate in Ideal-Forms, we are bound to seek their causes in that higher sphere. Doubleness, it is clear, is the cause of a thing being double, and from it is derived halfness.
Some correlatives owe their designations to the same Form, others to opposite Forms; it is thus that two objects are simultaneously double and half of each other, and one great and the other small. It may happen that both correlatives exist in one object-likeness and unlikeness, and, in general, identity and difference, so that the same thing will be at once like and unlike, identical and different.
The question arises here whether sharing in the same Form could make one man depraved and another more depraved. In the case of total depravity, clearly the two are made equal by the absence of a Form. Where there is a difference of degree, the one has participated in a Form which has failed to predominate, the other in a Form which has failed still more: or, if we choose the negative aspect, we may think of them both as failing to participate in a Form which naturally belonged to them.
Sensation may be regarded as a Form of double origin [determined both by the sense-organ and by the sensible object]; and similarly with knowledge.
Habit is an Act directed upon something had [some experience produced by habit] and binding it as it were with the subject having [experiencing], as the Act of production binds producer and product.
Measurement is an Act of the measurer upon the measured object: it too is therefore a kind of Reason-Principle.
Now if the condition of being related is regarded as a Form having a generic unity, Relation must be allowed to be a single genus owing its reality to a Reason-Principle involved in all instances. If however the Reason-Principles [governing the correlatives] stand opposed and have the differences to which we have referred, there may perhaps not be a single genus, but this will not prevent all relatives being expressed in terms of a certain likeness and falling under a single category.
But even if the cases of which we have spoken can be subsumed under a single head, it is nevertheless impossible to include in a single genus all that goes with them in the one common category: for the category includes negations and derivatives – not only, for example, double but also its negative, the resultant doubleness and the act of doubling. But we cannot include in one genus both the thing and its negative – double and not-double, relative and not-relative – any more than in dealing with the genus animal we can insert in it the nonanimal. Moreover, doubleness and doubling have only the relation to double that whiteness has to white; they cannot be classed as identical with it.