I've asked before as to what propositions count as meaningful, and, as some commentators and responders helpfully pointed out, 'meaning' and 'propositions' appear to be identical entities in the philosophical jargon. So I've decided to ask a new question and it is this: What are the propositions (or what is the meaning of sentences)? Is there a formal way of defining propositions? Or are they defined ostensibly only?

  • 4
    Please see plato.stanford.edu/entries/propositions
    – virmaior
    May 13, 2015 at 22:53
  • Put simply, propositions can be evaluated for truth and falsity. Meaning is what assigns truth values to propositions. One influential view (due to D. Kaplan) defines meaning as a function from contexts to extensions. It all comes down to whether you believe Frege's axioms (=the extensions of propositions are truth values).
    – Atamiri
    May 14, 2015 at 9:02
  • what do you mean "defining propositions" - how to define the term propositions, or how to do something called "defining" for any proposition? the former is what the questions answer
    – user6917
    May 14, 2015 at 13:06
  • Ways to get some... gratification.
    – Neil Meyer
    May 14, 2015 at 18:41
  • I would tend to think that meaning (or not) depends strongly on context. But that the general idea of meaningful is that a communication affects a recipient's model of the world. The pure information content aspect was tackled by Shannon in 1946, and is known as Shannon/Weaver information theory, IIRC. Simply put, a symbol with N equally probable possibilities requires log2(N) bits, and can be said to carry that much info; probability p = 1/N then carries -log2(p). But that was very primitive, not taking account of knowledge. I have not heard of any theory that does include knowledge. May 16, 2015 at 2:31

2 Answers 2


I'm not aware of a formal definition of proposition, and the issue is highly contentious, but one way to think about them is as follows. Take a sentence like "Clarissa is a fish". You can believe this, as in the sentence, "I believe Clarissa is a fish". You can hope it: "I hope Clarissa is a fish." You can doubt it: "I doubt Clarissa is a fish". These are various different propositional attitudes you take toward the proposition, but through them all the proposition is one and the same.

You can also say the sentence today and/or tomorrow. You can say it to your friend and/or to your teacher. You can write it down and/or say it. You can say it in English or translate it into German. The sentence is what changes in all these permutations but not the proposition.

In sum, the proposition is in effect the idealized content of the sentence, which is constant through time, place, translation, and propositional attitude. In this sense you can understand it as an abstraction from the sentences in which it occurs.


Well excuse the pseudo-philosophy BUT

A proposition is anything which I can express with a that in front, right?

I believe that, it is true that, that...

introducing a subordinate clause expressing a statement

It seems to me that this suggests a meta-language. e.g., if I think that A<->B then I am asserting not just a material equivalence, but a semantic (not syntactic) identity: such that it is always true in my language that A<->B.

  1. It is snowing iff there are snowflakes in the sky
  2. That it is snowing iff there are snowflakes in the sky

In 2 I am claiming the equivalence can be said in a metalanguage, that the the equivalence is "subordinate" to something else which can be expressed.

The supraordinate clause in 1 is augmented with THAT, and whatever it is which the that is about (e.g. I believe that) cannot be expressed without 1 [or something like it).

So in conclusion: a proposition is an expression which depends on some metalanguage.

Nagarjuna claims

I do not make a proposition


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .