—even though the attributes still depend on a locus. Yet in spite of that he went on to say, “If by ‘necessarily existent’ you mean something that possesses no receptive cause, then on such an interpretation it is not a Necessary Existent. Nevertheless it remains eternal and without any cause.” But none of this was required in order to demolish their argument. When he rehearsed their tripartite division concerning essence and attribute—(1) each could be independent of the other, (2) each could depend on the other, or (3) one could be independent while the other depends—he himself opted for the third alternative. On that assumption, they maintained: “The needy one is known, while the necessary one is the other; and whichever of the two is an effect stands in need of a cause.” This, they said, would entail that the very essence of the Necessary Existent becomes linked to a cause. He responded that the preferred view is that the essence, in its subsistence, needs no attributes, whereas the attribute does need the subject, exactly as is the case with us. He then rejected their maxim: “Whatever needs another cannot be the Necessary Existent,” and so forth.

What he quotes as their proof is plainly self-contradictory, as is his answer to it. Their statement, “Whenever the needy is an effect he requires a cause, which would bind the essence of the Necessary Existent to a cause,” is internally inconsistent. On that premise, only the subject would be the necessary being, and the attribute would not be necessarily existent by itself; rather, it would be an effect that depends on the subject, as he himself admitted. To call the attribute after that “necessarily existent,” and to claim thereby that the Necessary Existent is no longer connected to any cause, is contradictory; its upshot is that the attribute is not necessarily existent. That is why, in his reply, he was forced to proclaim it necessarily existent and then explain this by saying that it has no efficient cause, even though it does have a locus. None of this is needed. It is enough to say that, on this view, the attribute is not necessarily existent but an effect, just as they asserted. And were one to label the receptive cause an “efficient cause” in this sense, no difficulty would arise, for on this reckoning the necessary being is the essence alone; and thus the matter