Conjunction is most often marked by the word and, but there are English sentences without this word that also may be analyzed as conjunctions. First of all, there are quite a number of expressions—such as also, in addition, and moreover—that serve as stylistic variants of and. But conjunctions also may employ another group of words that are not simple stylistic variants of and. The principal example is the word but.
This may be a surprise. Although a sharp ear might detect a slight difference in meaning between and and moreover, the difference between and and but is unmistakable. Consider, for example, the following two sentences, which differ only in the use of these two words:
Adams spoke forcefully to the committee, and they agreed to the expenditure
Adams spoke forcefully to the committee, but they agreed to the expenditure.
These sentences would be used under different circumstances, and it may seem odd to count them as logically equivalent, which is what we must do if we are to analyze both as conjunctions of the same two components.
This is the first of several points at which we must recall the distinctions between truth and appropriateness and between implication and implicature. As was noted in 1.3.3, our concern is with only the first concept in each pair and thus with only certain aspects of meaning. Specifically, we count two sentences as equivalent if they have the same truth conditions. Any differences between their meanings that have no effect on their truth and falsity are irrelevant for our purposes.
So we must look more closely at the nature of the difference in meaning between and and but. It is clear that the second sentence above carries a suggestion of contrast between the two components—perhaps Adams spoke against the expenditure or the committee usually rejected Adams’s advice—and it is also clear that the suggestion of contrast is absent in the first. Now, suppose that the second sentence was used in a context where the suggested contrast is not present—perhaps the expenditure was approved because Adams spoke for it. The assertion of the second sentence would then be inappropriate, but would it be false?
Let us use the test of a yes-no question. Imagine that you attended a meeting were Adams persuaded a committee to agree to a certain expenditure and that later someone who had heard rumors of the proceedings asked you the question Is it true that Adams spoke forcefully to the committee, but they agreed to the expenditure?. How would you reply? This is something you must decide for yourself; but, for my own part, I would say something like, Yes, but he spoke for the expenditure, not against it.
That is, I would give a yes-but answer, reacting to the sentence whose truth was asked about as one whose assertion would be true but inappropriate. And it is for this reason that I will suggest we analyze sentences formed using but and other similar words—such as however, though, and nonetheless—as conjunctions. These words are not just signs of conjunction; but their differences from and lie outside their effect on truth conditions.
There are cases of other sorts where analysis by conjunction is legitimate though not obvious. Sometimes, for example, there is no word at all marking the conjunction. The operation of conjunction produces a compound sentence that commits us to the truth of both its components, and there are linguistic devices other than the use of particular words that enable us to roll two claims up into one in this way. For example, the sentence It was a hot, windy day is equivalent to It was a hot and windy day and can be analyzed as the conjunction
It was a hot day ∧ it was a windy day.
An analysis of a sentence might even separate a modifier from the expression it modifies. One common case of this is provided by adjectives used attributively—i.e., applied directly to the noun they modify. For example, we may treat Sam’s car is a green Chevy as if it were Sam’s car is a Chevy, and it’s green. It is important to note that, for reasons discussed in the next section, these analyses work only because the adjectives appear in a predicate nominative employing the indefinite article—i.e., in the form represented by
X is a ... Y
or by a similar form with a different tense. However, this is a very common pattern so there will be many occasions to apply this sort of analysis.
Another rather specific but important case of separating modifiers concerns relative clauses. There are really two cases here. The first is non-restrictive relative clauses—that is, ones marked off by commas. These can usually be analyzed as conjunctions. For example, Ann, who you met yesterday, called this morning can be understood as a conjunction of You met Ann yesterday and Ann called this morning.
The second sort of case is a restrictive relative clause—one not marked off by commas—appearing as part of a predicate nominative using the indefinite article. The grammatical pattern in this case is
X is a Y that ...
or a similar pattern using a different tense or another relative pronoun (such as who or which). A sentence like this can be analyzed as a conjunction of X is a Y and the result of putting X in the expression marked by ... at the point governed by the relative pronoun. For example, Sam's car is a Chevy that's green could be analyzed as the conjunction of Sam's car is a Chevy and Sam's car is green—i.e., analyzed in the same way as Sam’s car is a green Chevy. But relative clauses of this sort can be used to express many sorts of modification other than the simple application of adjectives. One example is The speaker was a writer who Sam admired, which can be analyzed as the conjunction of The speaker was a writer and Sam admired the speaker; here the second conjunct has the subject of the original sentence as its direct object rather than its subject.