Source: p 268 Middle, Introducing Philosophy for Canadians: A Text with Integrated Readings (2011 1 ed).

  Second, the theory doesn't satisfactorily explain what true beliefs correspond to. With what kind of thing or entity are they supposed to 'agree' or 'correspond'? To many philosophers, the answer has seemed obvious: true beliefs correspond to 'the facts'. But what, exactly, are facts? Here we need to be careful. Sometimes when we speak of 'facts', We are simply using that word as a synonym for truths, meaning 'true propositions or beliefs'. However. a defender of the correspondence theory had better not be using the word fact in this sense; for then all she would be saying is that true beliefs correspond to true belief s— and that certainly isn't very helpful. Perhaps what is meant by fact is something like an actual state of affairs in the world, such as the cat being on the mat in the kitchen. This suggestion sounds more promising; but it, too, faces problems. For what are we to say about beliefs such as 'There is no cat on the mat', or 'There is no milk in the refrigerator', or 'I do not have a brother'? If beliefs of this sort can be true—and they obviously can— doesn't this mean that there must be lots of shadowy 'negative' facts out there in the world, as well as positive ones? And here is another problem: How are we to identify facts? How do we count them or distinguish them? For instance, is the fact that makes the man's belief [1.] 'My cat is on the mat' [End of 1.] true the same fact as the fact that makes his belief [2.]' There is an animal in my house' true? [End of 2.] Questions like these have kept the defenders of facts busy.

I do not understand the difficulty in deducing 2 from 1; so what have I neglected? 2 is really a Conclusion from 1, if we add the following Hidden Premises:

  1. A cat is an animal.

  2. The mat is in my house.

2 Answers 2


I read the quoted text so that the author is not questioning whether one fact can be deduced from the other. The author is asking :

How are we to identify facts? How do we count them or distinguish them?

Note that he is using the word identify here in the sense of equality - to identify one expressed fact with another expressed fact as being one and the same fact. He is asking "how do we know when we are expressing the same fact in two different ways" so that we may correctly count them and distinguish them.

With this in mind, it should be clear how the example given highlights this problem. (Note that "there is an animal in my house" does not necessarily entail "the cat is on the mat".)


The correspondece theory of truth addresses the sort of relation that must hold between beliefs or propositions and the "actual states of affairs in the world," to use the author's phrase. (It helps to keep in mind that we can only know those states as we experience them, i.e. phenomenally.) The author presents two possible problems with maintaining this position:

  • Shadowy negative facts
  • The identification of facts

The idea of shadowy negative facts is only a problem if we make unfounded assumptions about how our propositional representations must relate to reality. It suggests that the form of our sentences must have some ontological realization in the real world. The author proposes that speaking in negative terms could imply that shadowy negatives must exist in reality. Of course, there's no reason to make such an assumption. The same state of affairs can be represented by a countless number of sentences formed in different ways.

I believe the second problem which he describes is motivated by a similar assumption. "My cat is on the mat" and "There is an animal in my house" may both be true and thus be verifiable with respect to the same state of affairs. As you pointed out, if the mat is in your house, one proposition follows from the other. There's no reason to assume that a form of description has to be identical with another form describing the same state of affairs. Both descriptions may be perfectly true even though they are expressed differently. Therefore, identifying such facts presents no real problem, at least not in the way he described.

You must log in to answer this question.