in itself—which is invalid, as explained above—or it has a cause: the totality itself, some part of it, or something external to it. A parallel distinction is made between signification of a whole meaning, of an included part, and of a necessary implication.

Thus, concerning the second possibility, he said: “As for the totality requiring a cause consisting of all the individual members, it would then be caused by itself, for that aggregate and the whole are one and the same. But ‘all’ in the sense of ‘each individual member’ does not determine the aggregate’s existence.”

He means that if what determines the totality’s existence is all the individual members, including their collective configuration, it follows that the contingent aggregate is caused by itself, since the aggregate, the whole, and the totality are one and the same. This differs from taking ‘all’ to mean ‘each individual member’: the aggregate does not have its existence determined by each member separately, but by all the members together. Just as ten does not arise from each of its units separately, so too with all other composites: a composite comes into being through all its parts together, including their collective configuration, if that configuration is regarded as an existent entity. If it is not so regarded, there is no need for this distinction; rather, one simply says that the totality is all the individual members, and there is nothing here apart from those members.

Perhaps this is what Avicenna intended, and this is why they raised those objections against him,