do not hold that they are alike, then those who hold that they constitute many different genera are even further from holding that they are alike. The disagreement over this has already been recounted, as reported by al-Ashʿarī and others.

With this established, consider his subsequent statement:

“Once this is established, we say: Each of those simple parts must touch one thing on its right and another on its left. Yet its right side is like its left; otherwise, it would itself be composite, whereas we have assumed it to be noncomposite. This is a contradiction. Once it is established that its right side is like its left, and that two like things must share all their necessary attributes, it follows conclusively that whatever touches its right side can come to touch its left side, and vice versa. If this is possible, then those parts can separate and come apart. This brings us back to the possibility of combination and separation in the essence of Allah Most High, which is impossible.”

There are two answers to this. Grant that you have posited it as a simple part, in no way composite. Yet such a part is said to touch one thing on its right and another on its left. This entails that it contains two things—a right side and a left side—which requires it to be composite. But you have assumed it to be noncomposite. This therefore combines two contradictories. This is made clear by the fact that its touching something on its right and something on its left is among