Meaning vs. presupposition

Sentences in natural language are used for asking questions, giving commands, making assertions, expressing feelings, etc. I shall deal with a distinction between the presuppositional aspect of the semantic structure of a predicate on the one hand and the ‘meaning’ proper of the predicate on the other hand. We may identify the presuppositions of a sentence as those conditions which must be satisfied before the sentence can be used in any of the functions just mentioned. Thus the sentence identified as (54):

(54) Please open the door

can be used as a command only if the LT is in a position to know what door has been mentioned and only if that door is not, at TLA, open.¹ The test that the existence and specificity of a door and its being in a closed state make up the presuppositions of (54) rather than part of its meaning is that under negation the sentence is used to give quite different instructions, yet the presuppositional conditions are unaffected.

(54') Please don’t open the door.

The presuppositions about the existence and specificity of the door relate to the use of the definite article and have been much discussed in the philosophical literature on referring.² The presupposition about the closed state of the door is a property of the verb open.

Presuppositions of sentences may be associated with grammatical constructions independent of specific predicate words (such as those associated with the word even or with the counterfactual-conditional construction),³ but I shall mention here only those that must be identified with the semantic structure of predicate words. If we limit our considerations to sentences which can be used for making assertions, we can separate the basic meaning of a predicate from its presuppositions by describing the former as being relevant to determining whether as an assertion it is true or false, the latter as being relevant to determining whether the sentence is capable of being an assertion in the first place. If the presuppositional conditions are not satisfied, the sentence is simply not apt; only if these conditions are satisfied can a sentence be appropriately used for asking a question, issuing a command, making an apology, pronouncing a moral or aesthetic judgment, or, in the cases we shall consider, making an assertion.

Let us illustrate the distinction we are after with the verb prove in construction with two that-clauses. Consider sentence (55):

(55) That Harry is still living with his mother proves that he is a bad marriage risk.

It is apparent that if I were to say (55) about somebody who is an orphan, nobody would say that I was speaking falsely, only that I was speaking inappropriately. If prove has a that-clause subject and a that-clause object, we say that the truth of the first that-clause is presupposed, and that the verb is used to assert a causal or logical connection between the two clauses and thus (when used affirmatively) to imply the truth of the second clause. That this separation is correct may be seen by replacing prove in (55) by doesn't prove and noting that the presuppositional aspects of (55), concerning the truth of the first that-clause, are unaffected by the change.

It is difficult to find pairs of words in which the presuppositional content of one is the meaning content of the other, but a fairly close approximation to this situation is found in the pair of verbs accuse and criticize. The words differ from each other on other grounds, in that accuse is capable of being a ‘performative’, while criticize is not; and criticize, unlike accuse, is capable of being used in senses where no negative evaluation is intended. In sentences (56) and (57) we are using accuse in a non-performative sense and we are using criticize as a three-argument predicate in a ‘negative-evaluative’ sense:

(56) Harry accused Mary of writing the editorial.

(57) Harry criticized Mary for writing the editorial.

I would say that a speaker who utters (56) presupposes that Harry regarded the editorial-writing activity as ‘bad’ and asserts that Harry claimed that Mary was the one who did it; while a speaker who utters (57) presupposes that Harry regarded Mary as the writer of the editorial and asserts that Harry claimed the editorial-writing behavior or its result as being ‘bad’. The content of the presupposition in each one of these verbs shows up in the assertive meaning of the other.

Certain apparent counter-examples to the claims I have been making about presuppositions can be interpreted as ‘semi-quotations’, I believe. Some utterances are to be thought of as comments on the appropriate use of words. Uses of the verb chase presuppose that the entity identified as the direct object is moving fast. Uses of the verb escape presuppose that the entity identified by the subject noun-phrase was contained somewhere ‘by force’ previous to the time of focus. These presuppositions, as expected, are unaffected by sentence negation:

It seems to me that sentences like (60) and (61) are partly comments on the appropriateness of the words chase and escape for the situations being described. These are sentences that would most naturally be used in contexts in which the word chase or escape had just been uttered:

(60) I didn’t ‘chase’ the thief; as it happened, he couldn’t get his car started.

(61) I didn’t ‘escape’ from the prison; they released me.

It is important to realize that the difference between assertion and presupposition is a difference that is not merely to be found in the typical predicate words known as verbs and adjectives. The difference is found in predicatively used nouns as well. In the best-known meaning of bachelor, for example, the negation-test reveals that only the property of ‘having never been married’ is part of the meaning proper. Uses of this word (as predicate) presuppose that the entities being described are human, male and adult. We know that this is so because sentence (62):

(62) That person is not a bachelor

is only used as a claim that the person is or has been married, never as a claim that the person is a female or a child. That is, it is simply not appropriate to use (62), or its non-negative counterpart, when speaking of anyone who is not a human, male adult.

¹ I am only dealing here with those presuppositions which are relatable to the content of the utterance. It is also true, of course, that (54) can be used appropriately as a command only if the LT understands English, is believed by the LS to be awake, is not totally paralyzed, etc. These matters have more to do with questions of ‘ good faith ’ in speech communication than with information that is to be understood as knowledge about individual lexical items.

² P. F. Strawson, ‘On referring’, Mind (1950), LIX, 320-44.

³ For some examples see my mistitled paper, ‘Entailment rules in a semantic theory’, Project on Linguistic Analysis Report no. 10, 1965.