Locke famously defines judgment, knowledge, etc., in terms of the joining or separating of ideas. It is quite probable that Locke’s source for this is the Port-Royal Logic. There are two well-known problems with this view. First, according to this view in order to think that Peter is not living I must mentally separate the idea of Peter from the idea of living, but if I do that then its not clear how this judgment, that Peter is not living, can be a unit which can be, for instance, embedded in complex sentences. Locke makes matters worse by talking about the joining and separating of verbal signs. Taken literally, this ought to have the consequence that ‘Peter ………………………………….. living’ means ‘Peter is not living,’ since the signs are so far apart. Obviously he can’t mean anything so ridiculous as that, but he doesn’t really tell us what he does mean.
The second problem is that it seems, on this view, that the proposition is the same as the act of judging (affirming or denying) and so one cannot entertain a proposition without either affirming or denying it. This is especially a problem given the theory of language in Locke and the Port-Royalists: a sentence is conceived as a sort of recipe for constructing a certain complex mental state. In successful communication, the speaker translates her mental state into words according to the linguistic conventions, and the hearer then ‘decodes’ the words and reconstructs the mental state. But it seems to follow that you can’t understand someone without believing him, and that’s surely wrong.
I want to suggest, briefly, that the Port-Royalists can answer these objections. The first problem is actually quite simple, and here Locke may simply be less than explicit because most of his audience had read the Logic. In their treatment of judgment, the Port-Royalists sound most like Locke when they write, “After conceiving things by our ideas, we compare these ideas and, finding that some belong together and others do not, we unite or separate them. This is called affirming or denying, and in general judging” (Buroker, p. 82). However, when they first introduce judging as one of “four principal operations of the mind,” they define it as “the action in which the mind, bringing together different ideas, affirms of one that it is the other, or denies of one that it is the other. This occurs when, for example, having the idea of the earth and the idea of round, I affirm or deny that the earth is round’ (Buroker, p. 23). Here it is said that in both affirmation and denial the subject idea and the predicate idea are ‘brought together.’ Also, affirmation and denial are clearly not being explained or analyzed in terms of uniting or separating. In fact, the Port-Royalists take these ‘principal operations’ as primitive and known by introspection. What we should really understand them as claiming is that affirming involves thinking of the ideas as united (or: as going together), and denying involves thinking of the ideas as separated (or: as not going together). Thus the mental act, thinking of the idea of Peter and the idea of living as not going together does form a unit that can be embedded in larger complexes, although the idea of Peter and the idea of living are in some sense ‘separated in thought’ by the act. They are separated in thought only in the sense of being thought of as separated.
In the case of this first problem, I think this is probably what Locke intended as well. However, it’s not clear to me that Locke is even aware of the second problem or has any inkling of an answer. The answer is found mostly in the Port-Royal Grammar, which Locke may never have read, rather than the Logic which he seems to have studied carefully. The long and short of it is that affirmation and denial are not the only mental operations for putting ideas together. In particular, according to the Grammar, there are different mental operations which can be signified by different moods of the verb, or by specialized particles, depending on the language. So, for instance, the Grammar mentions the Latin particle ‘utinam’, the French idiom ‘plut a Dieu’ and the Greek optative mood as signifying wishing, and imperatives and so forth as signifying various stronger dispositions of the will toward the connection of the ideas (Rieux and Rollins, pp. 137-138). The subjunctive is treated as signifying ‘modified affirmations,’ such as those that occur in conditionals (Rieux and Rollins, p. 136). Now the Port-Royalists treat the antecedent of a conditional as a sort of supposition from which we reason (Buroker, pp. 99-101), and the fact that the antecedent is merely supposed is what is (sometimes, in some languages) being explicitly flagged by the subjunctive mood of the verb. Sometimes the subjunctive can play the same role standing alone, as in the Greek deliberative subjunctive (the Port-Royalists don’t mention this example).
In other words, the long and short of it is, I think, that for the Port-Royalists thinking of two ideas as perhaps going together, is a different mental operation upon them than either affirmation or denial, and this is what goes on when we merely entertain a proposition. Indeed, just such a thing also happens, according to the Port-Royalists, in certain subordinate clauses. For instance, they say that in the sentence ‘people who are pious are charitable,’ “the mind judges that the idea ‘pious’ is not incompatible with the idea ‘people,’ and so we can consider them as joined together and then examine what belongs to them as unified” (Buroker, p. 90). In other words, we first have a modified affirmation, something like, ‘perhaps some people are pious,’ and then we affirm (simply) of those (hypothetical) people that they are (would be) charitable.
(Cross-posted at blog.kennypearce.net)