Is Everything an Object Among Objects?

My opponent says Yes; I return a negative answer.  This entry continues the discussion in earlier theological posts, but leaves the simple God out of it, the better to dig down to the bare logical bones of the matter.  Theologians do not have proprietary rights in the Inexpressible and the Ineffable.

Argument For

The opponent offers a reductio ad absurdum:

a. It is not the case that everything is an object. (Assumption for reductio)
Therefore
b. Something is not an object. (From (a) by Quantifier Negation.)
c. 'Something' means some thing, some object.
Therefore
d. Some object is not an object.  Contradiction!
Therefore
e. Everything is an object.  (By reductio ad absurdum)

The argument could also be put as follows.  An object is anything that comes within the range of a logical quantifier.  So someone who denies that everything is an object must be affirming that something is not an object, which is tantamount to saying that some item that comes within the range of a quantifier -- 'some' in this instance -- does not come with the range of a quantifier. Contradiction. Therefore, everything is an object!

Argument Against

First, two subarguments for premises in my main argument against.

Subargument I

Every declarative sentence contains at least one predicate.
No predicate functioning as a predicate is a name.
Therefore
No declarative sentence consists of names only.

For example, 'Hillary is crooked' cannot be parsed as a concatenation of three names.  A sentence is not a list of names.  And the unity of a proposition expressed by a sentence is not the unity of a collection of objects.   A proposition attracts a truth-value, but no collection of objects attracts a truth-value.  The mereological sum Hillary + instantiation + crookedness is neither true nor false. But Hillary is crooked is true.  

Adding a further object will not transform the sum into a proposition for well-known Bradleyan reasons.

So what makes the difference between a mereological sum of sub-propositional (but proposition-appropriate) items and a proposition?  A noncompound proposition is clearly more than its sub-propositional constituents.  The proposition a is F is more than the sum a + F-ness.  The former is either true or false; the latter is neither.  (Bivalence is assumed.) What does this 'more' consist in? The 'more' is not nothing since it grounds the difference between sum and proposition.  The 'more' is evidently not objectifiable or reifiable.  

The ancient problem of the unity of the sentence/proposition was already sighted by the 'divine' Plato near the beginning of our tradition.  The problem points us beyond the realm of objects.

The paradox, of course, is that I cannot say what I mean, or am 'pointing to.'  For if I say: 'Something lies beyond the realm of objects,' then I say in effect: 'Some object is not an object.'  But I am getting ahead of myself.

Subargument II

Names refer to objects and predicate expressions refer to concepts.
Anything that can be quantified over can in principle be named.
Concepts cannot be named.
Therefore
Concepts cannot be quantified over.

In support of the second premise:   'Some horse is hungry' cannot be true unless there is a particular horse in the domain over which the existential/particular quantifier ranges, and this horse must in principle be nameable as, say, 'Harry' or 'Secretariat.'  There needn't be a name for the critter in question; but it must be possible that there be a name.

Now for the main argument contra:

A. There are declarative sentences.
B. No declarative sentence consists of names only; predicative expressions are also required.  (Conclusion of subargument I)
C. Predicates refer to concepts, not objects.
D. Concepts cannot be quantified over. (Conclusion of Subargment II)
Therefore
E. Concepts are real ingredients of propositions but they are not objects.
Therefore
F. Not everything real is an object among objects.

Summary

The unity of the sentence/proposition is one of several problems that point us beyond what I have been calling the Discursive Framework (DF).  These problems, properly understood, show the inadequacy of this framework and refute its claim to unrestricted applicability.  The unity of the sentence/proposition  needs accounting.  (There is also the unity of concrete truth-making facts or states of affairs that cries out for explanation.)  

Now we should try to account for sentential/propositional unity as parsimoniously as possible.  We shouldn't bring in any queer posits if we can avoid them, a point on which my opponent will insist, and in those very terms.  Unfortunately, we cannot eke by with objects alone.  To repeat:  a sentence is not a list; a proposition is not a collection of objects.  So we need to bring in some queer entities,whether Fregean unsaturated concepts, or Strawsonian nonrelational ties, or relational tropes, or some odd-ball Bergmannian nexus, even my very own Unifier. (See A Paradigm Theory of Existence, Kluwer, 2002.)

The problem, of course, is that these queer items entangle us in contradictions when we try to state the theories in which they figure.  The contradictions give aid and comfort to the Opponent who takes them as justifying his claim that the DF is unrestricted in its applicability.

Frege's paradox of the horse illustrates this very well.  Frege notoriously asserted, "The concept horse is not a concept."  Why not? Because 'the concept horse' names an object, and no object is a concept.  An application of existential/particular generalizattion to Frege's paradoxical sentence yields:  Some concepts are not concepts.  But that's a contradiction, as is the original sentence.

But Frege was no 'stoner' to use an expression of the Opponent.  His contradiction is, shall we say, motivated.  Indeed, it is rationally motivated by the noble attempt to understand the nature of the proposition and the nature of logic itself.

Why can't concepts be named?  Suppose we try to name the concept involved in 'Hillary is crooked.'  The name would have to be something like 'crookedness.'  The transformation of the predicate into an abstract substantive loses the verbal chararacter, the characterizing character of the predicate '___ is crooked' functioning as a predicate.  If 'crookedness' has a referent, then that referent is an object.  But as I said, the proposition Hillary is crooked is not the mereological sum Hillary + crookedness.  The former attracts a truth-value; the latter doesn't.

The unity of a proposition, without which it cannot be either true or false, is not the unity of an object or a collection of objects, which is just a higher-order object.  This peculiar truth-value attractive unity cannot be accounted for in terms of any object or collection of objects.  And yet it is real.  So not everything real is an object.  

Impasse?

We seem to be in an aporetic bind.  We need to bring in some queer elements to solve various problems that are plainly genuine and not pseudo.  But the queer items generate paradoxes which, from within the DF, are indistinguishable from bare-faced contradictions.  The paradoxes/contradictions arise when we attempt to state the theories in which the queer entities figure.  They arise when we attempt to talk about and theorize about the pre-objective or non-objectifiable.  I cannot state that no concept is an object, for example, without treating concepts as objects.  But doing so drains the concept of its predicative nature.  I cannot say what I mean.  I can't eff the ineffable.

One move the Opponent can make is to flatly deny that there is the Inexpressible, thereby defying the author of Tractatus 6.522. Das Mystische does not exist, and, not existing, it cannot show itself (sich zeigen).

If the Opponent is a theist, then his god must be a being among beings, a highest being, a most distinguished denizen of the Discursive Framework, but not ipsum esse subsistens.

How might the Opponent deal with the problem of the unity of the sentence/proposition?  Perhaps he will say that a noncompound proposition is a partially but not wholly analyzable unity of sense, but that the 'more' that makes the proposition more than the sum of its constituents has no Deep Meaning, it does not 'point' us anywhere, and certainly not into Cloud Cuckoo Land but is  merely a curious factum brutum for which there is no accounting, no philosophical explanation.

I don't think this would be a good answer, but this entry is already too long.

At the moment I would happy if I could get the Opponent to make a minimal concession, namely, that I have mounted  a strong, though not compelling, rational case for the thesis that reality is not exhausted by objects, and that I have not "destroyed all of logic" in so doing.

But I am undermining the claim of the DF to have universal applicability.  This undermining takes place within the DF by reflection of something essential to the DF, namely, propositions.  As long as I refrain from making positive assertions about the Transdiscursive, I avoid contradiction. 

Related articles

Is it Coherently Conceivable that One Person Have Two Natures?

Does the Potentiality Argument Prove Too Much?

Thinker, Thought, Sentence

The Discursive Framework, Logic, and Whether the Via Negativa is the Path to Nowhere

More on One Person-Two Natures: Response to Timothy Pawl

Comments

[Astute]

On subarguments I and II, it seems to me you are taking my doctrine of (I) the ineffability of the assertion and (II) the ineffability of the predicate against me. Yes?

[Astute]

I think I will go for it not 'pointing' us anywhere, and certainly not into Cloud Cuckoo Land but is merely a curious factum brutum for which there is no accounting, no philosophical explanation.

[BV]

Your second comment is helpful because it confirms my guess at what your view is, and challenges me to say what is wrong with your view.

As for your first comment, I didn't think I was using doctrines of yours against you.

Have you posted some statements of (I) and (II)?

If you think predicates are ineffable, doesn't that commit you to some objects being ineffable?

[Astute]

>> As for your first comment, I didn't think I was using doctrines of yours against you. Have you posted some statements of (I) and (II)?

I have long urged the ineffability of both the assertoric component and the predicative component. Once quite recently, when I distinguished strawman nominalism of the Armstrong knockdown kind, which has predicates arbitrarily imposed, with enlightened nominalism, which holds that not everthing that looks like a name, is a name. We can only express what a verb expresses by using a verb, therefore we can’t quantify over what a verb expresses. Indeed the noun phrase ‘what a verb expresses’ is itself misleading, because it is a noun phrase and not a verb. We discussed this very recently.

>>Your second comment is helpful because it confirms my guess at what your view is, and challenges me to say what is wrong with your view. If you think predicates are ineffable, doesn't that commit you to some objects being ineffable?

Actually ‘ineffable’ means too great or extreme to be expressed or described in words, implying our vocabulary is insufficient. Verbs and predicates by contrast express exactly what they are intended to express. There is nothing unclear about ‘Socrates is arguing’. The proposition expresses exactly what it is meant to express.

The problem is the noun phrase ‘what it is meant to express’. This cannot have a referent, otherwise the proposition ‘Socrates is arguing’ would have a referent, and the referent would be the truth of the proposition. Similarly if the verb phrase ‘is arguing’ had a referent, the proposition ‘Socrates is arguing’ would be a list of nouns, which it isn’t.

For that reason, it does not commit the Ockhamist to ‘some objects being ineffable’. Quite the opposite.

This suggests a curious factum brutum for which there is no philosophical explanation, but perhaps a psychological one.

[BV]

>>We can only express what a verb expresses by using a verb, therefore we can’t quantify over what a verb expresses.<<

You have just conceded defeat nolens volens. You have just told us that a verb expresses something but that this thing cannot be quantified over. So this thing is not some thing. Contradiction.

[BV]

More clearly perhaps:

The question is whether everything can be quantified over. You admit that what a verb expresses cannot be quantified over. So you must admit that not everything can be quantified over

Reality is not exhausted by objects.

[Astute]

Not so fast. A that-clause and a sentence stand in two fundamentally different relations to a fact. The clause ‘that grass is green’ refers to the fact that grass is green. The sentence ‘grass is green’ states that grass is green. (Or if you want to be picky, the person using the noun phrase refers, the person using the sentence states etc).

In both cases the object of the verb phrase (refers to/states) can be quantified over. So for some x, the that-clause refers to x, the sentence states x. But the relation is different.

[BV]

Are you telling me now that you accept facts?

But you are right that there is an important difference between 'That grass is green' and 'Grass is green.' Only the second expresses a complete thought, a proposition. The former is a nominal expression.

Is referring the same as naming?

Your point is that referring to a fact is different from stating a fact. Let us say that picking out is the genus of which referring and stating are species. Then your further point is that in both cases the object picked out can be quantified over.

I'll grant you that the object of the nominal phrase can be quantified over, but not the object of the sentence.

Moreover, you contradict yourself. How does what you say comport with your earlier >> we can’t quantify over what a verb expresses<< ??

[Astute]

>>Moreover, you contradict yourself. How does what you say comport with your earlier >> we can’t quantify over what a verb expresses<< ??

I thought about it an realised this was wrong. Clearly we can quantify over what a verb expresses. The verb 'run' expresses running. The noun 'running' refers to running.

>>I'll grant you that the object of the nominal phrase can be quantified over, but not the object of the sentence.
I don’t understand. If by ‘object’ you mean referent, then only a noun phrase has a referent. A sentence cannot refer, it can only express or state.

That is, the sentence ‘Socrates runs’ expresses the object/referent of ‘that Socrates runs’.

[BV]

>> Clearly we can quantify over what a verb expresses. The verb 'run' expresses running. The noun 'running' refers to running.<<

'Run' can be either a verb or a noun depending on how it is used. 'I run slowly' versus 'I went for a run.'

Running is a noun -- a gerund to be precise -- in 'Running is better for weight-loss than walking.' But 'running' in 'Socrates is running' is part of the predicate. Same in 'Running Socrates sweats.' In both of the last two examples, 'running' is a participle

We can quantify over what the noun 'run' and the gerund 'running' refers to. Not so sure about the verb and the participle.

[Astute]

Right. But my point is that there must be at least two different relations between language and reality if we are to have a science of meaning at all. Proof:

1. Every science has a subject matter.
2. The propositions of that science refer to its subject matter. E.g. in the proposition ‘water is H2O’, the word ‘water’ refers to water, ‘H’ to hydrogen etc.
3. Assume there is a science of meaning.
4. Then the propositions of that science must refer to meanings.
5. But to do this, those propositions of the science must themselves have meanings.
6. The meanings that the propositions have cannot be the same as the meanings they refer to. Otherwise the propositions of the science would be the same as the propositions that they are about. It would be like the science of water consisting of bowls of water.

Example: we can refer to the meaning of the word ‘and’, by using the noun phrase ‘the meaning of the word ‘and’’. But that noun phrase cannot mean the same as the word ‘and’, since it is a noun phrase and not a conjunction. Thus:

The meaning of the word ‘and’ = the meaning that the phrase ‘the meaning of the word ‘and’’ refers to.

Hence there must be at least these two different semantic relations, namely ‘referring to’ and ‘having’.

Frege’s error was to suppose it can all be explained by a single relation of referring. Thus ‘Socrates’ refers to an Object, ‘runs’ refers to a Concept, which quickly leads to a contradiction. He should have said that the word ‘runs’ has a meaning, and that the noun phrase ‘the meaning of the verb ‘runs’’ refers to that same meaning.

>> We can quantify over what the noun 'run' and the gerund 'running' refers to. Not so sure about the verb and the participle.

That is because the verb and the participle are not noun phrases, and so cannot refer. However, they have a meaning, even though they can’t refer to the meaning. Moreover we can construct a noun phrase that refers to the meaning they have. Thus everything whatsoever can be quantified over.

[Astute]

That said, I am still not certain which of the following is correct:

1. There is no difficulty.
2. There is a verbal, not a real difficulty.
3. There is a real difficulty.

[BV]

I'm not buying it. I say that in reality there must be something that corresponds to a predicate functioning as a predicate and that this something is not something over which we can quantify.

I assert: The coffee is hot!

Your view is that the meaning that 'is hot' HAS when the sentence is assertively uttered = the meaning that 'the meaning of "is hot" ' REFERS TO.

This item, moreover, is an object and nothing like a Fregean unsaturated concept, and it can be quantified over.

I'd say you are ignoring the datum that a proposition is not a collection of objects. There is a missing ingredient: the unity that attracts a truth-value.

[Londiniensis]

As I say, I am still uncertain. There must be more to the sentence than the noun phrase, I.e more meaning, as it were. So something more. But something more is something. Yet not quantifiable.

You see I am not disagreeing.

[Astute]

>>I'd say you are ignoring the datum that a proposition is not a collection of objects. There is a missing ingredient: the unity that attracts a truth-value.

But there is the puzzle. 'a proposition is not a collection of objects'. Agreed. there is something else: the unity.

But if the unity is 'something else', then there is some *thing* else. And then the proposition is a collection of objects, if you include the unity.