be called rational evidence, and to evidence that signifies through the intention and will of the one providing it—the latter being called transmitted, conventional, or intentional evidence. In both categories, one thing often entails another: its existence indicates the existence of its necessary concomitant, while the absence of that concomitant does not indicate its absence. Thus, the absence of an entity indicates the absence of the attributes subsisting in it, and the absence of any substantive condition indicates the absence of that which depends upon it. For example, the absence of life indicates the absence of knowledge, and the absence of disorder indicates the absence of divinity in anyone other than Allah, and so forth.

As for the second category, which signifies through intention and choice, consider how people write the letters of the alphabet: they distinguish some by a dot and others by the absence of a dot. Thus, among jīm, hā, and khā, jīm is marked by a dot below, khā by a dot above, and hā by the absence of a dot. The same applies to rā and zāy, sīn and shīn, sād and dād, and tā and zā.

Likewise, grammatical particles are said to be distinguished by the absence of the identifying features of nouns and verbs. The same applies to expressions. If someone says to another, “I owe a thousand dirhams,” and then falls silent, this indicates that he means a thousand dirhams of full weight. But if he says, “A thousand debased or underweight dirhams,” or, “Except fifty,” his addition of the qualification or exception provides an indication contrary to the first.

Here, “a thousand” is connected to another expression; there, “a thousand” is left unconnected to any continuation. Its being unconnected is distinct from its signification. Thus, what signifies is not the expression alone, but the expression together with