« Frege's Regress | Main | Conformity for Diversity's Sake »

Thursday, November 03, 2011

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

Hi Dr. Vallencia,

Long time reader - first time commenter. ;D

You wrote: "Since T exists, and exists contingently, 'T exists' needs a truth-maker."

'T' is a contingent fact and so (as you noted) it has an explanation as to why it came to be. It looks like this explanation would be the efficient cause of 'T' (since it's contingent) and this cause would be the truth-maker of the statement 'T exists'.

In the case of the fact 'Tom exists', if 'T' refers to the truth-maker of this statement then it refers to the cause of Tom's existence. The cause of Tom's existence would be his parents, his conception, birth, or something in that neighborhood. And the truth-maker of 'T exists' would be his grandparents, or their conception, etc.

It seems to me that the suggested infinite regress of truth-makers corresponds to an infinite regress of efficient causes (as far as existence is concerned). Wouldn't a Thomist see this line of thought, smile, and point to Aquinas' Second Way as a reasonable way to hold onto the highly intuitive reality of 'truth-makers' by eliminating the infinite regress?

Archie's objection seems to relate to a question I had a while back on the idea of a truthmaker. I asked if the truthmaker of 'it is day' could be the sun rising (this was not my exact example, which I can't remember now). Bill's reply was that this was not correct.

Thanks for the comment, Mr Dawson.

I get the impression that you are confusing truth-making with causation. You say that 'T' is a contingent fact, but it is not 'T' but T.

The truth-maker of the SENTENCE 'Tom is fat' is not the cause of Tom's being fat. He is fat because he eats too much, doesn't exercise, etc.

Ed,

I'll modify your example slightly to make it more clear. What is the empirical cause of its becoming light? Answer: the rising of the sun (technically: the rotation of the earth).

A truth-maker, however, is not an empirical cause. Truth-making is not a relation that connects one event in space-time with another event in space-time. It is a relation that connects a truth-maker to a representation (a decl. sentence, a judgment-content, a Fregean proposition, etc.) Clearly, the sun's rising is not a representation such as a sentence.

Are you falling into the same confusion as Archie?

While I can't shake off the conviction that if p is true there is something in the world that makes it true, I can't help but feel that the kinds of truthmakers we have been discussing do not advance our understanding of the 'truthmaker relation'. For in going from p to its purported truthmaker t all we seem to be doing is 'gerundising' the verb phrase within p. Examples:

Tom is fat ==> Tom's fatness
Tom is seated ==> Tom's being seated
Tom exists ==> Tom

The noun phrases on the right are readily seen as terms that refer to objects in the world. But we know that not all referring terms succeed in referring. Tom's fatness or seatedness may not exist. Indeed Tom may not exist. It's the existence of Tom's fatness, etc, that implies that Tom is fat, etc. This suggests that 'Tom's fatness' remains a conceptual entity---we have not, as we had supposed, got outside the realm of representations. Do we really expect to achieve this by a trivial syntactic transformation? And is there not the risk that a somewhat convoluted phrase like 'Tom's being seated' gets its meaning by being understood as that which makes 'Tom is seated' true?

>>Are you falling into the same confusion as Archie?

No, I was checking that you reason for disagreement is the one I think it is, and you confirmed that. I.e. we are agreed on a point where we may disagree, namely that a truthmaker is not an 'empirical' cause, but is a relation that connects a truth-maker to a representation.

So there we agree. The central question, therefore, is whether Tom himself is the truthmaker for 'Tom exists'. We agree what that question is. We are not agreed on the answer, but that is because I am not sure how we would answer the question.

OK another try. Here is my reply to the ‘existence not a predicate’ objection.

Truthmakerists say that ‘a girl came top of the class’ has a truthmaker. What makes it true is that at least one thing falls under ‘girl’ that also falls under ‘top of the class’. And such a girl is not herself the truthmaker, because it is contingent that such a girl is top of the class. So what about ‘there is such a person as Tom’? What makes it true is that at least one thing falls under ‘person’ that also falls under ‘Tom’. And it is contingent that there is such a person as Tom, if ‘Tom exists’ is itself contingent. So why is Tom himself the truthmaker? I am seeing symmetry again. I am seeing a similarity between 'someone is top of the class' and 'someone is Tom'.

Of course, a direct reference theorist, who assimilates names to demonstratives, will argue that ‘Tom’ is meaningless unless it refers to an existing Tom. But if direct reference is true, singular existential statements are necessarily true. Which I assume you deny.

In summary: if singular existential statements are contingent, the referent cannot be the truthmaker of the statement. Ergo etc.

Ed,

That truth-making is not causation if of course not a controversial point. There are controversial points, but this is not one of them. Truth-makers are ontological grounds, not empirical causes.

If causation is 'horizontal,' then truth-making is 'vertical.'

Ed,

Yes, 'A girl came top of the class,' if true, is contingently true, so has a truth-maker. I assume that this sentence is logically equivalent to 'Some girl came top of the class.' Note that the latter sentence follows by Existential Generalization from 'Sally came top of the class' assuming that 'Sally' names the girl who came top of the class.

Truth-maker theorists do not hold that every sentence having a truth-maker has its OWN truth-maker; they hold merely that every contingent sentence -- more generally, every contingent truth-bearer -- has a truth-maker. So in the case being considered, the fact of Sally's being at the top of the class is the truth-maker of both 'Sally came top of the class' and 'A girl came top of the class.'

There is exactly one truth-maker for both the singular sentence and its existential generalization, and that one truth-maker is Sally's being at the top of the class.

Do you agree with this? We are still merely at the level of clarifying what the truth-maker theorist's view is.

Ed,

You see a similarity between

1. There exists an x such that x is at the top of the class
and
2. There exists an x such x = Tom

where (1) translates your 'Someone is top of the class,' and (2) translates 'Someone is Tom.'

I say: Tom himself is the T-maker of (2) but that Tom himself is not the T-maker of (1): the T-maker of (1) is the FACT of Tom's being at the top of the class. So I claim an asymmetry: in the one case Tom is the T-maker; in the other a complex having Tom as constituent is the T-maker.

You say: There is symmetry: either an individual (a nonfact) is T-maker in both cases, or a fact in both cases.

You may be making the following mistake: you may be confusing the 'is' of predication with the 'is' of identity. My translations above make clear what you fudge in you uses of 'is.'

Hi David,

I hope you are well.

Your comments are excellent and get at some of the difficulties involved in the positing of facts as truth-makers.

As you say in other words, the move from 'Tom is seated' to 'Tom's being seated' is the move from a declarative sentence which is either true or false and expresses a complete thought to a noun phrase that is neither true nor false nor expresses a complete thought.

Since there is apparently no way to get at these truth-making facts except via the nominalization of sentences, one can suspect, with P. F. Strawson, that facts are but shadows cast by sentences and not genuine extralinguistic entities. And so, as you say, it may be that we are not in the end breaking out of the circle of representations and making contact with extralinguistic reality.

And yet surely we cannot rest content with saying that 'Tom is seated' is just true. Surely there is more to a true sentence than the sentence that is true. It can't be language all the way down. Or all the way out.

I get the sense that nominalists like Ed are flirting with linguistic idealism.

In Butchvarov Against Facts, http://maverickphilosopher.typepad.com/maverick_philosopher/2011/08/butchvarov-against-facts-.html, I consider one form the critique of facts takes and try to respond.

>>You may be making the following mistake: you may be confusing the 'is' of predication with the 'is' of identity.

Actually we had this discussion before. My objection is valid if it is valid to predicate a proper name, i.e. if the proper name stands in fundamentally the same relation to its referent as a common name stands to the multiple things that fall under it. This is the position of traditional logic, but not of modern predicate logic.

The problem with modern predicate logic is that it seems to imply 'direct reference'. Indeed, the principle that the truthmaker of 'Tom exists' is Tom himself already looks like a restatement of the direct reference position.

If you don't believe in direct reference (which I define as the theory that a proper name loses its semantics, or part of its semantics when it loses its referent), can you still justify your position on truthmakers for singular sentences.

I have a further objection based on the non-atomicity of truthmakers, which I will put on the blog today.

I followed up David's point http://ocham.blogspot.com/2011/11/flirting-with-linguistic-idealism.html .

The nominalism here is not linguistic idealism. Rather, there is something out there, but it's not a 'something', i.e. not the potential referent of a noun phrase.

More like the ineffability of extralinguistic reality, rather than idealism.

Well that's interesting! You think extralinguistic reality is ineffable? Something like Kant's Ding an sich?

Ed,

You're right: our present dispute is related to (and perhaps reduces to) the dispute about the predicability of proper names.

And this is connected to our haecceity dispute.

If 'Tom' is predicable of an individual, then 'Tom' must express an haecceity-property. But I have argued ad nauseam that there are no such properties.

Your definition of 'direct reference' sounds vacuous. Surely a proper name loses PART of its semantics when it loses it referent! Even a descriptivist will say that. Suppose my use of 'Ed' is routed through the sense 'proprietor of Beyond Necessity' and then you are annihilated by divine fiat. Wouldn't you say that 'Ed' as used by me in this context after the annihilation loses PART of its semantics?

One thing is certain: we are mucking around in the deepest shit it is possible to muck around in! And that's what makes us true philosophers.

The direct reference dude says: meaning is exhausted by reference. But the descriptivist does NOT say that meaning is exhausted by sense; he says that meaning involves both sense and reference.

Do you agree?

>>Suppose my use of 'Ed' is routed through the sense 'proprietor of Beyond Necessity' and then you are annihilated by divine fiat. Wouldn't you say that 'Ed' as used by me in this context after the annihilation loses PART of its semantics?
<<

No I wouldn't say that, not in the proper sense of 'semantic'. If the meaning of a word has changed, even partly, and I know its meaning, then I must know that its meaning has changed. Suppose e.g. a German word has changed its meaning slightly, but I do not know this, and I mistranslate from German into English. Then I have not grasped the proper, i.e. current meaning of the German word.

>>The direct reference dude says: meaning is exhausted by reference. But the descriptivist does NOT say that meaning is exhausted by sense; he says that meaning involves both sense and reference. Do you agree?
<<

It depends whether 'reference' involves the meaning, i.e. semantics of the term, or whether it involves a mere external or accidental relation to the term. E.g. I understand the meaning of 'black hole', as something possessing certain physical properties. They were postulated as possible entities before a black hole was sighted. Thus, whether or not a black hole exists, or falls under the concept 'black hole', is independent of its semantics.

>>Well that's interesting! You think extralinguistic reality is ineffable? Something like Kant's Ding an sich?

Far too Germanic and deep. The phenomenon is a mere triviality of language, a trifle, a mere 'muddle'. That's the English way of putting it. Time to bring out the sherry, and a comfy chair, and a warm fire.

Part of the problem now is terminological. I am not using 'meaning' as equivalent to 'sense.' You may be.

Well we got here via my claim that " But if direct reference is true, singular existential statements are necessarily true. Which I assume you deny."

It's clear you do deny DR, and you seem to do so via the Fregean distinction between sense and reference. Here, reference is an accidental relation between term and referent, so that the term can lose its referent without losing its meaning or sense or semantics.

Unlike with classic DR.

So, whether the proper name has a referent or not is contingent, and so 'Tom exists' is contingent. But nevertheless you will insist that the truthmaker of 'Tom exists' is Tom himself. That is your position, right? Steering a narrow course between Direct Reference and the Infinite Regress. Yes?

Ed Ockham has put up a further post on the regression argument, placing it in a wider context. He suggests that any attempt to express what we could call 'accuracy' in a representational system will lead to regress. This feels right to me, but what then of the argument here that Tom is the truthmaker of 'Tom exists', which threatens to truncate the regress? Originally I thought it obvious that Tom satisfied the truthmaking criterion. In looking for a truthmaker for 'Tom exists' we seek an x such that 'x exists' entails 'Tom exists'. Substituting x=Tom seems to satisfy this 'equation', so Tom seems to be the required truthmaker. There is something fishy about this though. We appear to be trying to define a predicate phi by

(*) phi(x) iff E(x) entails E(Tom),
where E() denotes the existence predicate, and then asking if there exists an x such that phi(x). This is starting to get murky to say the least. First, how does existential quantification interact with the E() predicate, assuming we allow this? Second, how does it interact with the modal machinery implicit in entailment? Can anyone explain what (*) might mean?

One possibility might be to interpret entailment as provability: P entails Q iff there exists a proof of Q from P. This avoids having to explain (*) in modal terms. But the existence of a proof is a property of the representational system---the truthbearers. The predicate phi appears to be projecting this back onto the represented entities. Isn't this the kind of thing that gets us into really hot water (or deep shit)?

The comments to this entry are closed.

Google Search Engine

My Photo
Blog powered by Typepad
Member since 10/2008

Categories

Categories

March 2017

Sun Mon Tue Wed Thu Fri Sat
      1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 31  
Blog powered by Typepad