Philosophy Of Language in 9 Sentences


Let’s introduce some ideas that will stick with us as the foundations of our theory of language, and before presenting 9 sentences which challenge those foundations, and the changes that have been made in response to them. At the heart of the philosophy of language is the truism that we use language to describe objects. Objects are a certain way, have certain properties, and that’s what we talk about. To do so, we have words which talk about objects (such as names and pronouns like ‘this’ or ‘she’) and words which describe them (such as verbs and adjectives and adjectival phrases like ‘runs’ and ‘tall’ and ‘is a cat’).

9 Tricky Sentences

1. Everybody loves somebody (Frege, Begriffsschrift)

Surprisingly, this sentence held up logic for a long time. Here’s why. We introduced the idea of referential semantics: language describes objects, and some parts of language describe, while other parts some stand for, objects.

Frege’s 1879 Begriffsschrift introduced many of the themes and problems that remain important today in the philosophy of logic and language. Thankfully the notation has improved since then.

2. The King of Ireland doesn’t exist (Russell, ‘On Denoting’)

Russell was very taken with the idea of referential semantics: with the thought that words mean things in the world. Indeed, he thought that if a word didn’t have such a worldly meaning — if there wasn’t something out there in the world that it pointed to, the sentence was meaningless.

3. This sentence is false (from antiquity)

You’ve probably heard this one before, but to see the problem with this sentence, consider: for a sentence to be true, what it says must be true. So assume sentence 3 is true, then what it says must be true. But it says sentence 3 is false, so assuming sentence 3 is true, sentence 3 is false. And likewise if you assume it’s false. What we get is that if the sentence is true, it’s false, and if it’s false, it’s true. And this is a contradiction according to most systems of logic.

4. ∃xGolfs(x) (Tarski, ‘The Concept of Truth In Formalized Languages’)

So far, we have introduced quantified sentences of a formal language, but we haven’t said much about how they work. But there are deep problems here whose resolution is our fourth high point of the philosophy of language.

5. Everybody golfs (Montague, ‘Proper Treatment of Quantification In Ordinary English’)

Still on quantification! Things get subtle here, but the key point is quite simple and can be simply made. Temporarily ignore everything above about quantifiers and variables. Looking at our sentence 5 it certainly seems similar to ‘John golfs’, doesn’t it? Traditional grammarians, for example, would say that both consist of a subject and a predicate.


If you’re still on-board, we’ve got pretty far. We’re now able to provide an accurate semantic analysis of sentences containing quantifiers using just a few rules. The key ideas are referential semantics, an account of meaning according to which truth is the central concept, and an attempt to provide rules for how the meaning of more complicated expressions gets built up by the meanings of the parts of those expressions and how they are combined. In the next two sentences, I want to consider some important ways in which this account of meaning must be supplemented by something extra: a theory of how language is used.

6. I thee wed (J.L. Austin, How To Do Things With Words)

Remember I said that we use language to describe the world. But that’s not all we do with language. We also use language to change the word. Some actions are such that the only way you can perform them is by saying something. When you say things like that, you don’t describe some existing bit of reality, but rather make it the case that a new bit of reality exists. When I utter 6 (in the appropriate circumstance) I am not reporting on a preexisting bit of reality — rather, I am making it the case that I am married to the person I am addressing. These are known as ‘performative utterances’ and were made famous by J.L. Austin, and the concept has proved very influential as showing the power of language, and has been used, for example, in feminist analyses of gender and pornography.

7. She was rich but nice. (Grice ‘Logic and Conversation’)

Compare this sentence: ‘She was rich and nice’. Both these sentences make the same demand on the world: that the person has both the properties of being rich and of being nice. They are truth-conditionally equivalent. That means that, as far as our theory of meaning is concerned, they are equivalent as to meaning. But there is clearly something extra to 7 — it convey something over and above its meaning. In particular, it conveys the attitude or prejudice on the part of the speaker that rich people typically aren’t nice. Such conveyings, which are over and above the meaning assigned by formal semantic theories, are called ‘implicatures’ by Grice, and again have been the object of much study. Together 6&7 form part of the domain of pragmatics, which is concerned, roughly, with how we use language, particularly when that use enables us to do things that go beyond what an expression’s literal meaning permits.

8. It’s necessarily the case that 2+2=4 (Kripke, Naming and Necessity)

Remember we introduced ‘it is not the case that’ and treated it as if it were a description that described sentences, and pointed out that treating the object sentences described as truth values worked pretty well?

9. I am here now (Kaplan, ‘Demonstratives’)

So thanks to Kripke, we now think that sentences are associated not with a single truth value, but rather with a truth value for every possible world, and we can use that to account for necessary truth (there’s a similar generalization to be made for the meaning of subsentential expressions, but I won’t do so here).


We’ve covered a lot: about 100 years of sophisticated work and you’ve seen how we’ve complicated our basic model to account for quantifiers, pronouns, modal (like ‘it is necessary that’), performative and context-sensitive expressions. That’s a lot! And work continues to this day in these and other areas, as we build better and better theories to accommodate more and more data.



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store