Awhile back, I did a series of three posts (I, II, and III) in which I was trying to work out the relations between propositions (“props” for short) and states of affairs (“sofas” for short). I’ve since been rethinking things a bit. So here’s my new and (hopefully) improved theory.
Earlier I had written that propositions are assertoric units of meaning, but I now see that this is ambiguous between their being assertions and their being assertible. The second reading is correct, and so “assertoric” should be replaced with “assertible” for clarity’s sake.
One problem with the first reading is that, strictly speaking, it is only persons that make assertions, so defining propositions as assertions makes it analytic that there can be no propositions without persons. But it is doubtful that that should be admitted as an analytic truth. There is no clear contradiction is supposing that some proposition or other being true or false.
A second problem with the first reading is that it makes it difficult to construe the common element in “Is p true?” and “p is true”, where the latter expresses an assertion and the former does not. More generally, if we treat propositions as assertions, then to handle non-assertoric “propositional attitudes” we have to treat them as second-order qualifications of assertions. Thus, doubting that p becomes doubting whether to assert that X; withholding that p becomes withholding whether to assert that X; and so forth. X here, stands for an abstract entity that is just like a proposition except for being attitudinally neutral. Philosopher Roderick Chisholm employs abstract states of affairs to fill that role, and until recently I basically followed him in that. On this view, a proposition p posits an abstract state of affairs S and is true just in case that abstract state of affairs corresponds to a concrete state of affairs that “obtains”.
But all this is awkward and, I now think, needlessly complex. If we simply say that propositions are assertible units of meaning, then we can treat all propositional attitudes in the same way and we don’t need to bloat our ontology by admitting abstract states of affairs in addition to propositions.
My current view may be summed up as follows:
- A state of affairs is a parcel, any parcel, of reality, where the real consists in whatever is and is as it is independently of what any (non-archtypal) intellect thinks about it. States of affairs can overlap and include other states of affairs. The actual world is the totality of reality, the one all-inclusive state of affairs.
- A proposition is an assertible unit of meaning. It is an abstract representation of a state of affairs and, as such, is true if and only if the state of affairs represented obtains (= exists, is actual, is real). Propositions in themselves don’t posit states of affairs; they merely represent them.
- The meaning of a proposition consists in the sum total of its entailments. Two propositions are distinct if and only if they have different entailments. We make explicit the meaning of a proposition by considering what we would be commiting ourselves to were we to assert it.
- We use statements (declarative sentences) to express propositions. A statement is true if and only if the proposition it expresses it true.
- Any state of affairs that suffices to make a given proposition true and that includes no states of affairs distinct from itself that suffices to make that proposition true is a minimal truthmaker for that proposition. There can be multiple minimal truthmakers for a given proposition (e.g., “Some dogs exist” is made true by Fido’s existing; it is also made true by Lassie’s existing; it is also made true by Fido’s and Lassie’s existing, but the latter is not a minimal truthmaker.). Any state of affairs that includes a minimal truthmaker for a proposition is also a truthmaker for that proposition. Thus, the actual world, the all-inclusive state of affairs, is a truthmaker for all true propositions.
Hi Alan,
Could you remind me why on the current version of your model we need to invoke âpropsâ at all if we have sentences and âsofasâ? Arenât sentences bearers of meaning and truth, and donât they represent sofas in various ways?
Your example of a prop seems to me to be a sentence in Englishâor do I understand you? âSome dogs existsâ is presumably a sentence uttered by some speaker on some occasion, though a normal context for such an utterance eludes me.
May I change your example a little? Suppose you and I arguing about the attributes of various breeds of dog, and I assert that some Dalmations are excellent swimmers ( something you quite rightly doubt). In uttering that sentence in that context I posit a certain sofa as actual. What I say is true just in case that sofa accurately represents or describes the world. Etc.
Why do we need or want to invoke props in this model? What necessary or essential role do they have?
Hi Mac (I hope that’s OK for short),
As I conceive them, sentences are merely strings of physical markings, whether sounds, marks on a page, gestures, or whatever. They have no intrinsic meaning of their own, but function as signs to transmit a meaning. Hence, the meaning of a sentence is something distinct from the sentence. This is confirmed by the fact that different tokens of the same sentence-type can have different meanings (e.g., “It is raining today” uttered at different places and times), whereas tokens of different sentence types can have the same meaning (e.g., “Snow is white”, “Der Schnee ist weiss”).
The reason why my props look like sentences is because sentences are the only vehicles we have to communicate props. Since I can’t drop my intended thoughts directly into your mind, I have to express them in a public medium that will I hope invoke in your mind the idea that I have in my mind when I utter the sentence.
Sure, sentences can be truth-bearers. We can meaningfully predicate ‘truth’ or ‘falsity’ of some sentence (specifically declarative ones). But they are only derivate truth-bearers. A true sentence is true by virtue of the truth of the proposition it expresses. But a true proposition is not true in virtue of the truth of anything else. It is true in virtue of the way the world is.
Alan, help me understand props a little more.
Suppose we are thinking about a day of water sports and we drive out to the lake. I take one look at the waves and say, âitâs way too choppy to water ski today.â
Thatâs a sentence, not a prop, right? But the sentence is meant to express a thought or idea I have ( the thought being a private mental event ). So the chronology is: I looked at the lake. An idea ( prop) popped into my mind. Then I searched for and settled on a sentence in English to express that idea. ( In principle I could have chosen Latin or ASL or some other language, but since I intended to communicate with you in a normal fashion, I selected a sentence in English.)
Props– let me be sure I understand– are not sentences or clauses, but the private inner thing that ends up getting expressed in language or symbolically? Does that mean that props arenât linguistic? So is expressing a prop taking something that is fundamentally non-linguistic and structuring it in language? Or is expressing a prop something like translating from a private inner âlanguageâ to a natural language like English? Do you see my puzzlement here in trying to understand how we get from props to sentences that are said to “express” them?
If I introspect, is it possible for me to experience the prop as something different from the sentence I intend to utter?
If I look for neural correlates in the speech process, is there brief event to identify with entertaining the prop as distinct from framing the sentence?
Mac,
As you seem to realize, propositions (just like every other philosophically interesting concept) quickly lead us into a host of difficult philosophical issues having to do with the nature of thought, representation, language, and so forth. For that reason, I’m probably not equipped to answer all of your questions to your satisfaction.
For what it’s worth, here’s my take. Bear in mind that whereas the points I made in my last comment are widely (though not universally) agreed upon by philosophers, much of what I’m about to say is very controversial. Anyway, my working theory is that propositions are prelinguistic mental entities (“ideas” if you will) that we have to formulate in linguistic or quasi-linguistic terms to communicate with others and to critically reflect on our own thoughts. I say prelinguistic because I think some animals and very young children think and even reason, although I don’t suppose they are capable of carrying on anything like a conscious interior dialogue about it. And it’s hard to see how anyone could even begin to learn a language like English if they weren’t already capable of reasoning. Now, there is good reason to think that ourselves and some animals are born “hardwired” with various mental categories in place. Perhaps this amounts to a kind of natural “language”. If so, then propositions need not be prelinguistic vis-a-vis that sort of “language” while still being prelinguistic vis-a-vis acquired languages like English.
Well, I hope that helps. Gotta run.
“We make explicit the meaning of a proposition by considering what we would be commiting ourselves to were we to assert it.”
I am not sure what you mean by this. Could you elaborate or give an example?
One other interesting problem I see here. What about the relation between propositions and statements?
“A statement is true if and only if the proposition it expresses it true.”
A prop can represent reality or not, but you also have the fact that a statement can accurately or inaccurately express the prop.
A simple example is this. My toddler calls all meat chicken. Now if we are having beef for dinner he may say, ‘Chicken, please’ to indicate that he wants some beef. Because I am familiar with how he uses language, he has adequately expressed the proposition, although according to normal usage he has inaccurately used the words.
The question this raises in my mind is – In what sense must a statement ‘express’ a proposition in order to be true? ie It seems that the statement must not only express a true proposition in order to be true, but expess it in such a way as to truly represent (cause a representation of?)the proposition either to ourselves or the person hearing or reading it.
Alan: A proposition is an assertible unit of meaning. It is an abstract representation of a state of affairsâŚ
Tom: If a prop is only âassertibleâ but not necessarily an âassertionâ thenâŚ
(A) How does a unit of meaningâs being an âassertible abstract representation of a sofaâ differentiate it from an âassertionâ which is also an abstract unit of meaning that represents a sofa? Are assertions no longer propositions at all? Are assertions proposition-bearers which in turn are truth-bearers so that assertions arenât either true or false but rather their propositions are?
(B) Can you describe for me the makeup of an âabstract representation of a sofaâ which is not an assertion (but only assertible). Is this abstract representation non-linguistic? If so, how is it âmeaningfulâ? In what does its meaning inhere abstractly if not in language & grammar? In other words, if a prop is not a âlinguisticâ unit of meaning, what sort of unit is it? But that a prop is an abstract unit of âmeaningâ suggests to me itâs linguistic. I canât think of a ânon-linguistic abstract unit of meaningâ that might represent sofas; and I canât think of a ‘linguistic unit of meaning’ that is not an assertion to some mind. So if props are only assertible but not assertions, Iâm left wondering just what sort of abstract representation they are and in what precisely their meaning consists.
Tom
Alan-
I see your comment to Mac about “prelinguistic mental entities.” So that answers my question I suppose. I don’t see what evidence there is for such entities apart from their needing to exist to support the view that props are only assertible units of meaning, but who knows? Could be.
Tom
C Grace,
In saying that “we make explicit the meaning of a proposition by considering what we would be commiting ourselves to were we to assert it” what I have in mind is that we come to better understand a proposition the more we can trace out its entailments. I may believe (as I do) that time is continuous, but do I really understand what it means to say that something is ‘continuous’? Could I give a precise account of that concept? Oftentimes we only grasp concepts in a vague and half-articulate sort of way. Considering what our ideas commit us to is one way of refining them so as to come to a deeper and more precise understanding.
Regarding statements and propositions, the example you give is a good one. Miscommunication is a disconnect between what proposition a speaker has in mind when they utter a statement (i.e., what they intend to express by the statement), and the proposition that is evoked in a listener’s mind by their hearing of that statement (i.e., what the listener takes the statement to express). In short, there can be a difference between ‘speaker-meaning’ and ‘listener-meaning’. I take it that what proposition a statement expresses is determined by speaker-meaning. But unless the speaker expresses that meaning in a way that the listener can grasp, that’s not the meaning that will come across. And because all of us have somewhat different experience and background beliefs, there is always a possibility for miscommunication.
Hi Tom,
I take it that what differentiates the ‘assertible’ from the ‘asserted’ is that the latter necessarily involves both something ‘assertible’ and a mind or person that posits or asserts it. Here’s an example: God presumably could have created centaurs, and he knows this. The proposition “Centaurs exist” is, therefore, not something God would assert, but it is something that he could assert, provided that he creates centaurs. So if propositions in themselves are merely assertible, then an assertion is something more, namely, a proposition intended by some mind to accurately represent reality. As assertion is true iff its propositional content is true, which obtains when the world is as represented in the proposition.
As far as abstract representations of sofas go, I don’t have any elaborate theory to offer, but it seems to me that they could be linguistic and/or pictorical.
On the linguistic/pre-linguistic issue, haven’t you ever had an idea in your mind that you couldn’t articulate because you lacked the right words? Doesn’t that suggest that we have a level of understanding that is more basic than language? I suspect that good metaphors work by tapping into that.
Alan: The proposition “Centaurs exist” is, therefore, not something God would assert, but it is something that he could assert, provided that he creates centaurs.
Tom: But I understand you to being saying that, supposing God created centaurs and they existed, the proposition âCentaurs existâ would still only be assertible and not an assertion, that its being true doesnât entail any sentient mindâs asserting it (even Godâs). That sounds strange to me. Itâs one thing to say a false proposition (like âCentaurs existâ) is assertible in the sense that it would be true (and therefore assertible) in worlds where centaurs exist. Itâs another thing to say âCentaurs existâ remains merely assertible, and not asserted by any sentient mind, (even Godâs) in those worlds where centaurs exist.
Iâm interested in the whole dialectic between language and meaning, and the better part of it all is well beyond me. Iâll probably in the end get with you on your arguments over on Prosblogion about the possibility of an omniscient knower (and Grim). Iâm just not clear on a lot of things.
Tom
Alan, thanks that clears things up.
Alan said:
1)A proposition is an assertible unit of meaning. It is an abstract representation of a state of affairs
2)assertion is something more, namely, a proposition intended by some mind to accurately represent reality
Let me combine these
an assertion is an abstract representation of a state of affairs intended by some mind to accurately represent reality.
This is the kind of circular intuition I was trying to explore in my post on intention. Your definition of assertion here includes both a given (is a rep of a sofa) and an agreement (intended to rep a sofa)
There seems no way to get around the fact that two minds must be accounted for in a theory dealing with our knowledge.
I think Tom has a point, God’s assertions must be prepropositional and not abstract. In fact I think that in this argument (logical metaphor?) creation of centaurs should be described as identical with God’s assertion of centaurs. (He speaks things into existence :))
Alan said: As far as abstract representations of sofas go, I don’t have any elaborate theory to offer, but it seems to me that they could be linguistic and/or pictorical.
This makes no sense if Props are AR’s (see (1) above) I concur with your tenative assertion that props are prelinguistic. Also, representations are not pictorial (I certainly don’t have a picture of anything in my head for the prop God is patient) I would tend to say that props/reps are metaphorical or symbolic.