If you're seeing this message, it means we're having trouble loading external resources on our website.

If you're behind a web filter, please make sure that the domains *.kastatic.org and *.kasandbox.org are unblocked.

Main content

Language: Conventional Implicature

William introduces us to different aspects of meaning, as studied by linguistics and philosophers. He tells us about the difference between the literal meaning of a sentence someone says, and what they intend to convey by using that sentence at that particular time.

Speaker: Dr. William Salmon, Assistant Professor of Linguistics, University of Minnesota Duluth.
Created by Gaurav Vazirani.

Want to join the conversation?

Video transcript

(intro music) My name is Will Salmon, and I'm an assistant[br]professor of linguistics at the University of Minnesota Duluth. Today I'm going to talk to you about different kinds of linguistic meanings, specifically about[br]conventional implicature and conversational implicature, one way of dividing up linguistic meaning that's been very influential[br]over the last 40 years or so, which grew out of the work of the Oxford philosopher Paul Grice and his notions of[br]conversational implicature and conventional implicature. So a good starting place for us is to notice that there's[br]often a difference between what a speaker says[br]and what a speaker means. Take the sentence "I'm tired." Someone asks you to go out on a date and you say "I'm tired"[br]as a response to them. The linguistic meaning[br]of what you had said is just that you were tired, that is, the meanings of[br]the words "I," "am," and "tired" are taken literally,[br]combined into a sentence, and give you the straightforward meaning simply that you were tired. And that's all that you[br]have said, taken literally. However, you've likely[br]conveyed much more than this to the person who asked you out on a date. You most likely conveyed to the person that you don't want to go[br]out on a date with them. So the linguistic meaning here would be that you were tired while the speaker meaning would be what you really intended[br]to convey to the person, namely, that you don't[br]want to go out on the date. So you've turned down the invitation in a kind of an indirect way. This is what Grice would have called a "conversational implicature." So conversational[br]implicatures are messages that speakers convey that[br]are often above and beyond the literal meaning of[br]the words that they speak. The trick to it all is that[br]speakers hope to get hearers to recognize their intention in speaking, even if this intention[br]doesn't line up directly with the words that a speaker speaks. So in uttering a sentence, I hope to get my hearer to recognize what my purpose is in uttering it. If I say "I'm tired" in[br]response to an invitation, I hope that my hearer will ask herself why I'm telling her that I'm tired. And as a result, I hope that she'll infer that I don't really want[br]to go out on the date. So we now have two[br]potential messages here. The first message is the literal meaning of the sentence "I am tired." The second message is the implied meaning, that I don't want to go on the date. Now, the first of these[br]meanings is what we call "conventional meaning"[br]or "linguistic meaning," that is, it's attached directly to the meanings of the words in question. The second meaning, which[br]is the implied meaning, is non-conventional. Now, by this I mean[br]that the second meaning isn't attached to the meanings[br]of the words "I am tired." So if I say "I'm tired"[br]in another context, it won't count as turning down a date. For example, I could say[br]"I'm tired" late at night in order to imply that[br]I'm ready to go to bed. Or I could say "I am tired"[br]first thing in the morning in order to imply that I[br]don't want to get out of bed. Or maybe I'm in an argument with someone and I can say "I am[br]tired" in order to imply that I don't want to argue it anymore. In all the cases, I will just have said literally that I'm tired,[br]but I might have implied different things in all of the cases. So you can use the same sentence to imply many different things and[br]many different contexts. And the implied meaning[br]is non-conventional. Now, philosophers and linguists often talk about the truth conditional[br]meaning of sentences. Now, this is a kind of[br]meaning that's directly part of the literal meaning of a sentence, and it's what's used to determine whether a sentence is true or not. So for the sentence "I am tired,"[br]this sentence will be true if and only if I am actually tired. The truth conditional[br]meaning of the sentence is not necessarily a part then of the implied meanings[br]we've discussed so far. So if I say "I'm tired" to imply that I don't want to go on a date, the truth conditions still just depend on whether or not I'm actually tired. The part about not wanting[br]to go on the date doesn't factor into the truth[br]conditions of "I'm tired." So the conversational implicature[br]meaning, the part about not wanting to go on the date isn't[br]truth conditional. So let's sum up what we've got so far. We have truth conditional meaning, which is literal and conventional. And we have non-truth conditional meaning, which is implied and non-conventional. This basic distinction takes us a long way toward understanding the kinds of meaning that we find in everyday language. But there's one other dimension of meaning which is also very important. This is the conventional[br]implicature meaning. And this kind of meaning[br]seems to sit midway between the two kinds of[br]meaning we've discussed so far, that is, conventional[br]implicature is conventional, is part of the direct, actual, literal meaning of the words in question. But at the same time it's[br]not truth conditional. So let's look at an example. Think about a sentence like "John is short, but he is strong." This sentence is equal in[br]terms of truth conditions, to John is short and he is strong. As long as John is both short and strong, both sentences are true. If he is not both short and strong, then both sentences are false. There's no in between here. The question is, what is[br]the "but" sentence have that the "and" sentence is lacking? And the difference seems to be in there being some kind of[br]contrast in the "but" sentence that's not present in the "and" sentence. And this contrast is between[br]the expectations we have about being short and about being strong. So maybe we don't normally[br]think of someone who is short as also being strong, or[br]something along these lines. In any case, there's a contrast there between these two parts of the sentence. Now, this contrast is clearly part of the conventional[br]literal meaning of "but." If we substitute "and" for "but," we can see that the truth[br]conditions stay the same. But now the contrast is gone. So the contrast must[br]be a conventional part of the meaning of "but." At the same time, though, the contrast is not part of truth conditional meaning. We know this because the[br]"and"-"but" sentences we just mentioned have[br]the same truth conditions. But only one of the sentences has the actual contrasting feature. So Grice called this kind of conventional but non-truth conditional meaning a conventional implicature. They're found all over everyday language. For example, sentences like "Frankly, the Vikings didn't win" or "Those darn Vikings lost the game" are used to express truth-evaluable[br]statements as well as a speaker's[br]attitudes toward the statements. Consider the second sentence, "Those darn Vikings lost the game." Here a speaker's made a simple claim about the Vikings losing a game. The speaker has also conveyed[br]an emotion toward that claim, namely, that the speaker[br]is frustrated or angry over the Vikings having lost. This communicated emotion is important in helping us understand[br]the speaker's feelings, but it's not important to the[br]truth conditions of the claim. All that matters to the truth conditions is whether or not the[br]Vikings lost the game. So this kind of emotive meaning is directly associated[br]with the word "darn," that is, its conventional meaning. But it doesn't enter[br]into the truth conditions of the sentence. As a result, we can talk[br]about the meaning of this word and many, many related words in terms of conventional implicature. To finish here, we've now discussed three dimensions of meaning. There's conventional[br]truth conditional meaning which is very close to the[br]literal meaning of what is said. There's also the[br]non-conventional implied meaning, which would be conversational implicature, and which is potentially quite far from the literal meaning of what is said. And finally, there's conventional non-truth conditional meaning, which would be the[br]conventional implicature that we just discussed. So there we have, in short, an[br]introduction to three different dimensions[br]of linguistic meaning, a set of distinctions which is very useful in fields like philosophy and linguistics, or really in any field[br]in which it's necessary to understand the relation between language and how we[br]use language to communicate. Thank you. Subtitles by the Amara.org community