PHIL280: Truth and Reality
Week 12: Deflationism about Truth II

    Strawson’s performative theory of 'true'

  1. Peter Strawson has put forward a theory called the performative theory of 'true'. See:

    To see what it is, we need some background on speech acts.

  2. Austin on Speech Acts

  3. In a series of lectures in the 1950s (later published as a book, How to Do Things with Words), J. L. Austin proposed that we have a special kind of use of declarative sentences, one that he called a performative use (as opposed to what he called their constative use). For example, we typically use the following declarative sentences performatively:
  4. According to Austin, on these uses a speaker is not describing the world, not saying something that is true or false (as in constative uses). Rather, she is doing something, or getting something done: promising, naming, pronouncing, giving notice, sentencing, declaring, apologizing, and voting.

    Here are some other things that we might perform:

    admitting, advising, announcing, assuring, authorizing, censuring, committing, complimenting, conceding, confessing, congratulating, defining, denying, granting, hypothesizing, inquiring, insisting, judging, pardoning, pleading, pledging, predicting, proposing, reporting, reprehending, thanking, urging, vowing, warning.

  5. Some things we can do without using words (making a cake). Some things we can do using words (promising). Are there things that we cannot do using words? Things that we can only do using words? Things that we can do with and without?
  6. Notice that whether or not a sentence is used performatively or constatively seems not to depend on the form of the sentence. The examples above could all be used constatively rather than performatively. And the same acts could be performed with the following sentences instead:
  7. How do we tell one use from the other? One suggestion: a use of a sentence is performative iff the sentence can be put in the form of those in the first list of examples (if it’s not already); and ‘hereby’ can then be added. Question: can we use ‘hereby’ as a test directly?
  8. But wait, aren't some constative uses of declarative sentences also performative uses? Suppose I were to assert ‘Today is Monday’, thereby using the sentence constatively. I have also done something in the process - I have asserted something (that today is Monday). I might even make this clear by using either of the following sentences instead: ‘I assert that today is Monday’, ‘I hereby assert that today is Monday’.
  9. Maybe that’s right. Maybe we should think of every use of a sentence as being performative – as being used to do something. Austin suggests that every utterance has illocutionary force, which is in addition to its locutionary content (i.e. descriptive content, what is said).
  10. Different utterances of the one sentence can have different illocutionary forces (even with the same locutionary content). ‘That dog has been starving for days’ – could be uttered as an observation, or a warning. ‘If you don’t stop I’ll hit you’ – is that a threat or a promise? ‘I will be there’ – is that a prediction, a promise, or a threat?
  11. Austin also proposed that utterances can have perlocutionary effects: frightening, convincing, alarming, amazing, amusing, annoying, boring, embarrassing, encouraging, deceiving, distracting, impressing, informing, inspiring, insulting, irritating, persuading. I might use ‘I will be there’ to promise that I will come to your party (illocutionary force) and thereby inspire you to make a cake (perlocutionary effect).
  12. So when a speaker uses a sentence she performs at least three kinds of act: a locutionary act: saying something (which is more than just uttering the sentence); an illocutionary act: what is done in saying that – asking a question, giving an order, making a promise, stating a fact, etc; and a perlocutionary act: achieving something by means of saying what she did – drawing attention to something, convincing someone of something, getting someone to do something, etc. These are three different kinds of things that we do with words. Example: ‘Shut the door!’ Austin thought that no one of these three kinds of act could be reduced to the other two.
  13. Strawson on 'true'

  14. Strawson agrees with the redundancy theory that 'true' does not add any content to sentences in which it is used (‘true’ is not used to describe anything), and that there is no such property as truth. But he disagrees with the redundancy theory's claim that all we do by asserting 'It is true that grass is green' is assert the proposition that grass is green. We do that, he argues, but we also do more - we perform the speech act of confirming (or endorsing, conceding, praising, approving, admitting, condoning) this proposition. This is the central function of the predicate 'true'.
  15. Instead of saying ‘It is true that snow is white’, I could say ‘I embrace the claim that snow is white’, or ‘I commend to you the claim that snow is white’ - this would achieve the same thing.
  16. What about when I say something like, 'What the policeman said is true'? Strawson proposes that we can break this into a descriptive part and a performative part: 'The policeman made a statement. I confirm it.' In uttering the words ‘I confirm it’ I am not describing something I do, but doing something. My utterance is true iff the policeman made a statement.
  17. In the case of an utterance of the liar sentence, the speaker is attempting to praise something that it isn’t there. Like saying ‘ditto’ when no one has spoken. The liar sentence is grammatical, but isn’t being used to express a proposition, and so is not something from which a contradiction can be derived.
  18. Some problems for Strawson

  19. First problem. Suppose that John, not knowing what the policeman said, but knowing that he is reliable, assertively utters the following sentence:

    Many things the policeman said are true.

    According to Strawson’s theory, John is thereby endorsing or confirming several things that the policeman said. But John has actually done no such thing. There is not even one thing that the policeman said that John has confirmed.

  20. Second problem. Suppose that Person A says several things, including something false. Person B says: “Every assertion made by A was true.” According to Stawson's theory, person B's utterance is true iff person A made some assertions. Since person A did make some assertions, person B's utterance is true. But this is the wrong result - person B's utterance is not true.
  21. Third problem. According to Strawson's theory, the following have the same meaning:

    If that is right, then the following pairs of sentences ought to have the same meaning, but they don’t (and they might even differ in truth value):

  22. Fourth problem. What does it take for an argument to be valid? The normal account says this: an argument is valid iff it is not possible for its premises to be true and its conclusion not true. If Strawson's theory is correct, then this amounts to saying: an argument is valid iff it is not possible for me to confirm its premises and yet not confirm its conclusion. But that is wrong.
  23. The prosentential theory of 'true'

  24. We next consider a theory of 'true' called the prosentential theory of 'true'. See:
  25. The basic idea is that we use ‘true’ to form prosentences. To see what this means, we need some background on proforms.
  26. Proforms

  27. We are very familiar with pronouns. Here are some examples:

    Each pronoun here has an antecedent expression, which is a noun or noun phrase (hence the name 'pronoun'). These are all so-called lazy uses of pronouns - we could equally well have used the antecedent expression again:

  28. There are also quantificational uses of pronouns:

    In these cases we cannot just replace the pronoun with its antecedent expression - doing so would change the meaning of the sentence:

  29. We also have proverbs:
  30. We also have proadjectives:
  31. We also have proadverbs:
  32. Prosentences

  33. Given that we have pronouns, proverbs, proadjectives, and proadverbs, it would not be surprising to find that we have prosentences as well. According to the prosentential theory of 'true' we do.
  34. Suppose you want to avoid the repetition of using the sentence 'snow is white' twice below - you want to avoid using it the second time. What could you write instead?

    John believes that snow is white and Mary believes that snow is white too.

  35. Here are a couple of ways that you could do it:
  36. But none of these employs a prosentence - they each employ some other kind of proform. We can make this clear by marking the antecedent in each case - in no case is it the sentence 'snow is white':
  37. But here is another way we might do it:

    According to prosententialists, 'that is true' here is a prosentence, whose antecedent is the sentence 'snow is white':

  38. So the idea is that 'that is true' is a prosentence, and it is because it is handy to have prosentences (as it is all other proforms) that we have the word 'true' in our language.
  39. Like other proforms, 'that is true' has the same content as its antecedent expression. In this case its antecedent is a sentence whose content is a proposition - 'that is true' expresses the same proposition.
  40. Note that 'that is true' has the same content as its antecedent sentence, but it might have different pragmatic effects:
  41. Falsity

  42. A prosententialist might also give a prosentential theory of falsity: ‘that is false’ is a prosentence, which has the same content as the denial of its antecedent sentence.


No exercise this week - time to catch up/tidy up.