PHL280: Truth and Reality
Week 2: The Correspondence Theory of Truth

    The basic idea

  1. Here's the basic idea: To be true is to correspond with the facts.
  2. This is actually a specific instance of a more general idea: To be true is to stand in a certain relation to reality. Specific instances of this general idea are obtained by specifying (a) which relation is involved, and (b) what is meant by 'reality'. We will focus on the version above (it has become a fairly standard version).
  3. A long history

  4. Many people are thought to have endorsed a theory of truth along these lines:
  5. The correspondence theory is often said to be a very natural theory, perhaps even an obvious one. And the burden of proof is often thought to be on oppenents of the theory, to come up with arguments against it, rather on proponents of the theory, to come up with arguments for it. That is, it is a default position.
  6. Whether or not this is right, anyone who argues against the theory would be well-advised to explain why so many people are attracted to it.
  7. Clarifying the basic idea

  8. We need to make the basic idea more precise and conspicuous before we can subject it to careful scrutiny. (Williamson: "To be precise is to make it as easy as possible for others to prove one wrong. That is what requires courage.")
  9. First, we are looking for a theory of propositional truth, so we need to understand the correspondence theory as being such: For a proposition, p, to be true is for p to correspond with the facts.
  10. Next, we need to clarify what is meant by 'the facts'. Is this referring to a single thing, the totality of facts? If so, then according to the correspondence theory, there is a thing, the totality of facts, such that for any proposition, p, for p to be true is for p to correspond with this thing, the totality of facts.

    But this is not how the theory is usually understood. Usually it is understood to just require that for each proposition p, there is some fact with which p corresponds - possibly a different fact for different propositions.

  11. So here is how we might better formulate it:

    CT: For a proposition, p, to be true is for there to be a fact, f, such that p corresponds with f

  12. The proposition that snow is white is true. According to CT this is because there is some fact with which it corresponds. CT does not say which fact, but presumably it is the fact that snow is white. Note that CT allows that there might be other facts with which it also corresponds.

    The proposition that snow is black is false. According to CT this is because there is no fact with which it corresponds (note: there is no fact that snow is black).

  13. Some things to note about CT:
  14. What about not being true?

  15. According to CT, for a proposition, p, to not be true is for there to be no fact, f, such that p corresponds with f. In such a case, does the proposition still correspond with something other than a fact? Or does it not correspond with anything at all?
  16. Proponents of CT need not take a stand on this. But if they do, here are two things they might say:
  17. A concern about the direction of explanation

  18. It is natural to think that there are such things as facts, such as the fact that grass is green, and the fact that snow is white. So it is not controversial for the theory to appeal to such things - it doesn't really need to give any independent reason for thinking that there are such things.

    Suppose instead that CT appealed to 'klinks' rather than facts. Then it would have extra work to do, getting us to accept that there are such things.

  19. Suppose that there are indeed such things as facts. It may nevertheless be wrong to give an account of propositional truth in terms of them. It may be, for example, that we should do things the other way around - give an account of what facts are in terms of propositional truth.

    To illustrate: It seems natural (and correct) to give an account of being a bachelor in terms of being male (e.g.: to be a bachelor is to be an unmarried adult male). It seems very unnatural (and incorrect) to give an account of being male in terms of being a bachelor (e.g.: ... ).

  20. There is another way of putting this concern: CT is circular, because it accounts for propositional truth in terms of facts, and facts are to be accounted for in terms of propositional truth.
  21. A concern about what facts there are

  22. Some people think that proponents of CT need to tell us more about what facts there are.
  23. For example, in addition to the fact that I have one brother, is there also a fact that I have fewer than 3 brothers? In addition to the fact that grass is green and the fact that snow is white, do we have the following more complex facts as well?
  24. But it is open to proponents of CT to argue that this is everyone's problem - similar questions could be asked about just about everything, and so it is no reason to think that facts are especially suspicious.
  25. A concern about the identity of facts

  26. Some people think that proponents of CT need to tell us more about when two facts are and are not the same fact (i.e. identical). That is, more about how to individuate facts.
  27. For example, is the fact that Brutus stabbed Caesar the same fact as the fact that Caesar was stabbed by Brutus? Is the fact that Socrates died the same fact as the fact that Xanthippe became a widow?
  28. Again, it is open to proponents of CT to argue that this is everyone's problem.
  29. A concern about the number of facts

  30. Donald Davidson has famously argued (following an idea of Frege's) that there is just one fact. More about this in one of the exercises.
  31. Concerns about correspondence

  32. Why does the proposition that grass is green correspond with the fact that grass is green, rather than some other fact, such as the fact that snow is white?
  33. One possible explanation is this: propositions and facts both have structure; a proposition, p, corresponds with a fact, f, just in case p and f have (a) the same structure, and (b) corresponding constituents at each structural position (i.e. p and f are isomorphic). More about this in one of the exercises.
  34. An epistemological concern

  35. Here is an epistemological concern about CT. If CT is right, then we can never know p, for any p. Why? Because to know p we have to know that p is true, and to know that p is true we have to (if CT is right) know that there is a fact, f, such that p corresponds with f, and to know that there is a fact, f, such that p corresponds with f we have to get outside our heads to observe such correspondence, something which we cannot do. So if CT is right then we cannot know anything.
  36. This is a common form of objection to a theory: argue that the theory has a consequence which is either (a) false, or at least (b) undesirable. The consequence in this case being that we cannot know anything.
  37. How might a proponent of CT react?

Exercise

Do one of the following (not both):

  1. On pages 752-3 of his 1969 paper, ‘True to the facts’, Donald Davidson argues that there is just one fact. How does the argument go? What are the weaknesses of the argument?
  2. In Chapter 12 ('Truth and Falsehood') of his book Problems of Philosophy, Bertrand Russell presents a correspondence theory of truth. At the time, Russell did not believe in the existence of propositions, and his theory is a theory of truth for beliefs. But can you think of a way of converting the theory into a theory of truth for propositions?