It Goes Without Saying: Presuppositions, Entailments, and Implicatures
Language allows us to say lots of things. Anything, actually. Infinite amounts of things. We can talk about now, the next century, and ten thousand years ago; about us, her, that one over there, and people who don’t exist; about what happens here, in Siberia, on Saturn, or in other real, hypothetical, or made-up universes.
No wonder Homo sapiens never shuts up.
But, as interesting as it is to study everything we hairless apes can say with language and the mechanisms that allow it, it is just as interesting to explore everything we could say, but don’t, because language allows it to just be left unsaid.
So, let’s talk about what’s left unsaid, not because it’s taboo, not out of a desire to conceal it, but because we all agree that it goes without saying.
True or False
Most of the things we say are either true or false. There’s a relatively small set of utterances —greetings and goodbyes (good morning, hello, see you later), expressions of emotion or desire (congratulations, thank you, sorry, get well), among others— that are neither true nor false. That is, they’re not propositions.
A proposition is, simply put, a statement that can be either true or false, i.e. that can have a truth value. Thus, I ate a piece of candy is a proposition: it’s true if I actually ate a piece of candy, or false if I didn’t. In contrast, good morning is not a proposition: it doesn’t state anything that can be assigned a truth value.
In this post, then, we’ll be talking about propositional utterances, things we say that have a truth value, that are either true or false.
A proposition’s truth depends on several things. First, it depends on said proposition’s reflecting an obtaining state of affairs in the world. The proposition John is in the dining room is true if and only if a particular state of affairs obtains in the world, namely, that the individual ‘John’ be located in the space called ‘the dining room’. So far, so good.
But a proposition’s truth also depends on the truth of other propositions. For instance, in order for the proposition My sister went to med school to be true, we first need me to have a sister; the proposition I have a sister needs to be true, too.
In the case of John is in the dining room, it also has to be true that John exists and that The house has a dining room.
These are, if you will, statements “concealed” within other statements. These, we call presuppositions.
A presupposition is a statement q that needs to be true in order for a statement p to be true. It answers the question, “What has to be true for this to be true?” In order for My sister went to med school to be true, it has to be true that I have a sister.
In a way, then, presuppositions come “before” the statement: first, John exists, and then, John is in the dining room; first, I have a sister, and then, My sister went to med school. I don’t mean this in a temporal sense, but in a logical one: presuppositions come logically before the statement.
But, careful! While it is true that the statement needs the presupposition to be true, you will have noticed that it doesn’t go both ways. The presupposition can be true while the statement is false. It can be true that I have a sister, but that doesn’t necessarily mean that My sister went to med school. She might have majored in physics. Presupposition is therefore a unilateral relationship; it only works one way: p presupposing q doesn’t mean that q presupposes p.
In short, any statement presupposes the truth of other statements. If those other statements are false, the statement has to be false, too. However, those other statements being true doesn’t mean that my statement is automatically true as well.
Just as the truth of my statement depends on the truth of other propositions, there are yet other propositions the truth of which depends on my statement. For example, if my true statement is that Peter killed John, then it has to be true that John died.
This other kind of “concealed” statements are called entailments.
An entailment is a statement q that is automatically true if another statement p is true. It answers the question, “If this is true, what else has to be true?” If it’s true that Peter killed John, then it is also automatically true that John died.
Since there’s no other way this can play out, since the truth of q necessarily follows from the truth of p, this is a case of what some call a strong implication.
In a way, then, entailments come “after” the statement: first, Peter kills John, and then, John dies. I don’t mean this in a temporal sense, either, but in a logical one: entailments come logically after the statement.
Another warning. Just as presuppositions didn’t guarantee the truth of a statement —me having a sister doesn’t mean that my sister went to med school—, the truth of an entailment doesn’t guarantee the truth of the statement: John dying doesn’t mean that Peter killed him. Same as presuppositions, entailments are a unilateral relationship; they only work one way: p entailing q doesn’t mean that q entails p.
Entailments are not the opposite or the mirror of presuppositions, either, as you may have realized. I feel I’m repeating myself, but I don’t believe it’s pointless:
- My sister went to med school presupposes that I have a sister, but I have a sister doesn’t entail that my sister went to med school.
- Peter killed John entails that John died, but John died doesn’t presuppose that Peter killed him.
In short, a statement can entail, or strongly imply, the truth of other propositions. If the statement is true, then those other propositions have to be true, too. However, those other propositions being, by themselves, true doesn’t mean that my statement is necessarily true.
As I said, these are cases of strong implication, where a proposition’s truth necessarily follows from the truth of my statement.
But there are many cases where my statement merely suggests the truth of a proposition. These weak implications are called implicatures, a term coined by philosopher H. P. Grice, in 1975.
Take a look at this classical example of a micro-conversation:
A: I’m out of gas.
B: There’s a gas station just around the corner.
With her statement, B weakly implies that the gas station is open and that, there, A can buy the gas he needs. This implicature is weak, because it can be negated. B could go on to say But it’s closed now and it would be fine. Note that this would be impossible with a strong implication, i.e. an entailment: I can’t say Peter killed John, but John didn’t die, because that would create a contradiction.
So, we say that, unlike entailments, implicatures are defeasible; they can be negated or cancelled without creating a contradiction.
A: Would you like a slice of cake?
B: I’m a diabetic.
With her statement, B weakly implies that she doesn’t want a slice of cake, because it would be bad for her. This implicature, however, is defeasible. B could go on to say So make it a very small slice or But it’s a special day, and there would be no contradiction.
Implicatures, then, answer the question, “If this is true, what else must be true?” And the operative word here is must: it’s a matter of probability, not necessity as it is with entailments.
You may have noticed that implicatures are vital in conversation, especially because they are so economical. They allow us to provide succinct information, while relying on the other person’s ability to fill in the gaps.
What Do I Do with All This?
So, you know what presuppositions, entailments, and implicatures are. What now? What good is knowing all this?
Well, first, knowledge is its own reward. Knowing is good in and of itself, so you’re welcome.
Second, this lets you look differently at every-day conversations and have fun filling in the gaps or watching people put their foot in it because they can’t infer an implicature. Once again, you’re welcome.
And, lastly, these are reading tools. Knowing that, under what is said, there is much unsaid, and knowing how to get to the unsaid, it’s easier to read or re-read the discourses we come across daily in the press, on TV, or on social media.
The unsaid is a powerful political tool, maybe even more so than what is actually said. The implicatures hiding in apparently trivial words, such as but, can be used to reinforce prejudice and exclusion. Who hasn’t heard or read that little phrase, So-and-so is poor, but honest? The word but implies an opposition between these two things (poor and honest), and presents So-and-so as the remarkable exception to a supposed, otherwise general rule: ‘poor people are dishonest’.
It’s the same implicature we find in descriptions like She’s black, but very articulate (‘black people are ignorant and can’t speak properly’); He’s gay, but decent (‘gay men are indecent’); He’s young, but very capable (‘young people are inept’). All these implicatures reinforce prejudices and exclusions with important political consequences, under the protective cover of the unsaid.
Presuppositions can be used in the same way. A statement like X can’t be done, because we’re defending Y presupposes several things: that someone is attacking Y; that someone wants to do X; that X goes against Y; that Y is more important than X… It further implies that whoever wants to do X is also attacking Y.
So, I want you to take away three things from this post: (1) things aren’t true or false in isolation, but in connection to other things; (2) we need to become literate in the workings of language; and, specifically, (3) we need to learn how to read and listen with our eyes and ears on what is said, and our brain on the unsaid.