Thus, if one says, “Nabīdh (a drink prepared by steeping ingredients in water) is forbidden because it is khamr (intoxicating liquor),” its being khamr is the evidence, and this is a necessary attribute of the nabīdh, while prohibition necessarily applies to khamr. In a syllogism consisting of the two premises, “All the nabīdh under dispute is intoxicating or is khamr,” and “All khamr is forbidden,” you have not used the universal—“intoxicant” or “khamr”—to establish the very subject under dispute, which is narrower in scope than either khamr or nabīdh. This is therefore not an inference from that universal to the particular. Rather, you have used it to establish that this nabīdh is forbidden. Since the prohibition of this nabīdh falls within the prohibition of every intoxicant, some have described this as an inference from the universal to the particular.

The precise account, however, is that whatever holds for a universal holds for each of its particular instances. Prohibition is broader in scope than khamr, and since it applies to khamr, it applies to every particular instance of it. This is therefore an inference from one universal establishing that another universal applies to the particular instances of the former. The evidence is like a particular in relation to that other universal, while itself being a universal in relation to those particular instances. This is something they do not dispute: the evidence is the middle term, which is broader in scope than the minor term or coextensive with it, while the major term is broader in scope than the middle term or coextensive with it. The major term is the judgment, the attribute, and the predicate; it is the predicate of the conclusion. The minor term is that about which the judgment is made, the thing described, and the grammatical subject; it is the subject of the conclusion.

As for their statement that analogy is an inference from one particular to another: if this is stated without qualification and taken to mean an inference from one particular to another merely on the basis of their being particulars, it is mistaken. Analogical reasoning establishes its conclusion only through a middle term: the two cases share the cause of the ruling, or an indication of the ruling together with the cause. It is thus either an analogy based on a cause or an analogy based on an indication.