Category: Existence

At Tractatus Logico-Philosophicus 4.1271 we read: "So one cannot say, for example, 'There are objects', as one might say, 'There are books'."

In endnote 9, p. 194,  of "The Number of Things," Peter van Inwagen (Phil. Issues 12, 2002) writes:

Wittgenstein says that one cannot say " 'There are objects', as one might say, 'There are books'."  I have no idea what the words 'as one might say' ['wie man etwa sagt'] could mean so I will ignore them.

Is van Inwagen simply feigning incomprehension here?  How could he fail to understand what those words mean?  Wittgenstein's point is that object is a formal concept, unlike book.  One can say, meaningfully, that there are books.  One cannot say, meaningfully, that there are objects.  Whether Wittgenstein is right is a further question.  But what he is saying strikes me as clear enough, clear enough so that one ought to  have some idea of what he is saying rather than no idea.  By the way, van Inwagen is here engaging in a ploy of too many analytic philosophers.  In a situation in which it is tolerably, but not totally, clear what is being said, they say, 'I have no idea what you mean' when, to avoid churlishness, they ought to say, 'Would you please clarify exactly what you mean?'

Be this as it may.  Philosophers are a strange, in-bred breed of cat, and they acquire some strange tics.  My present topic is not the tics of philosophers, nor formal concepts either.

According to Wittgenstein, one cannot say (meaningfully) that there are objects.  Van Inwagen responds:

Why can one not say that there are objects?  Why not say it this way: '(Ex)(x = x)'? (p. 180)

Without endorsing Wittgenstein's claim, or trying to determine what exactly it means, my thesis is that van Inwagen's translation of 'There are objects' as 'Something is self-identical' is hopeless.

I do not deny the logical equivalence of the two sentences.   I do not claim that there are self-identical items that do not exist.  Everything exists.  My claim is that to exist is not to be self-identical.  They are not the very same 'property.'    If they were, then van Inwagen's translation would be unexceptionable.  But they are not.  Here is a reductio ad absurdum argument to show that existence and self-identity are distinct, that existence cannot be reduced to self-identity.

0. Existence and self-identity are the very same property. (Assumption for reductio)

1. If existence and self-identity are the very same property, then nonexistence and self-diversity are the very same property, and conversely. (Self-evident logical equivalence.)

2. Possibly, I do not exist.  (Self-evident premise: I am a contingent being.)

3. Possibly, I am not self-identical. (From 1, 2)

4. What is not self-identical is self-diverse.  (True by definition)

5. Possibly, I am self-diverse. (From 3, 4)

6. (5) is necessarily false.

7. (0) is false. Q.E.D.

The thin theory of existence is the theory that existence is exhaustively explicable in terms of the purely logical concepts of standard first-order predicate logic with identity.  Identity and quantification are such concepts.  Now the only way within this logic to translate 'There are objects' or 'Something exists' is the way van Inwagen suggests.  But what I have just shown is that 'Something is self-identical' does not say what 'Something exists' says.

If things exist, then of course they are self-identical.  What else would they be? Self-diverse?  But their existence is not their self-identity.  Their existence is their being there, their not being nothing, their  reality — however you want to put it.  If something is self-identical, it cannot be such unless it first exists.  It astonishes me that there are people, very intelligent people, who cannot see that.  What should we call this fallacy?  The essentialist fallacy?  The fallacy of thinking that being = what-being?  Or maybe it is not a fallacy of thinking, but a kind of blindness.  Some people are color-blind, some morally blind, some modally blind.  And others existence-blind. 

(Note to Alfredo and Peter L:  I need your help in understanding this particularly opaque portion of PvI's paper.)

Here are some notes on section 2 of Peter van Inwagen's "Being, Existence, and Ontological Commitment" (pp. 476-479 of the Metametaphysics volume).

The first of the Quinean theses that van Inwagen maintains is that "Being is not an activity."  Here is the opening sentence of the section:

Many philosophers distinguish between a thing's being and its nature.  These philosophers seem to think of, e.g., Socrates' being as the most general activity Socrates engages in.

These two sentences have me flummoxed.  Let me explain why.

First, the two sentences taken together imply that philosophers who distinguish between a thing's being (existence) and its nature think of a thing's being as the most general activity it engages in.  That's just false.  There are philosophers who distinguish between being and nature (Aquinas for example) without holding that a thing's being is the most general activity it engages in.  Since I hesitate  to impute something plainly false to the dean of the thin theorists, I must question what he's driving at.

My suspicion is that van Inwagen (447) gets the notion that being is an activity entirely from J. L. Austin's jocose footnote to p. 68 of Sense and Sensibilia (Oxford, 1962):  "The word ['exist'] is a verb, but it does not describe something that things do all the time, like breathing, only quieter — ticking over, as it were, in a metaphysical sort of way."  That's clever all right, but too frail a reed to support a global imputation to all thick theorists of the view that being is a peculiarly quiet activity.

Second, since van Inwagen goes on to deny that being is an activity, are we to conclude that he rejects the distinction between being and nature?  Is PvI denying that there is a distinction between Socrates' nature and his existence?  Is he suggesting the following argument:

a. If there is a distinction between a thing's being and its nature, then being is the most general activity the thing engages in.

b. Being is not the most general activity a thing engages in.

Therefore

c. There is no distinction between a thing's being and its nature.

I hope van Inwagen is not suggesting any such argument.  For that would not cohere with his commitment to a Quinean translation of 'Socrates exists' into 'It is not the case that everything is identical to Socrates.'  This implies that the existence of Socrates is his identity-to-something — in which case there is a distinction between Socrates' nature and his existence.  After all, Socrates' nature and the property of being identical to something are distinct.

Third,  PvI speaks of "many philosophers' but gives no examples.  He needs a footnote right at the end of the second sentence above.  He needs to quote philosophers who explicitly say that being is a most general activity.  Farther down the page he mentions Heidegger and Sartre, but no page references are given and no quotations.  So my third point is that PvI seems to be committing a Straw Man fallacy.  Which philosopher ever said that being is the most general activity a thing engages in?

The view van Inwagen ascribes to thick theorists such as Heidegger and Sartre  involves the following propositions:

1. Being is an activity.

2. Being is the most general activity that a thing engages in, one that is implied by every other activity the thing in question engages in.  Thus if Socrates is running, then he is moving on his feet, and if so, then he is moving through space, etc.  until we come to some one terminal activity that is implied by all the other activities  the thing is engaged in at the time.

3.  This most general terminal activity — being — is the same activity at every time the thing in question is engaging in any activity.

4.  This most general activity is the same for each member of a given category, Thus it is the same for Socrates and Plato, but presumably not the same for a bridge or an ass.

5.  This most general terminal activity of being (existing) is different (or can be different) for different categories of entity. Thus the most general activity of a table is not the same as the most activity of a human being.  And so there are different kinds of being, different kinds of this most general terminal activity. 

Van Inwagen imputes the above five theses to Heidegger and Sartre and, it appears, to all thick theorists.

There are  several topics to discuss.  One, which I will leave until later, is whether Heidegger and Sartre are committed to the five theses listed.  A second is whether thick theorists in general are committed to them.

Well, I'm a thick theorist and I don't see that I am committed to them.  As a thick theorist I am committed to the intelligibility of the idea that there are modes of existence (ways or modes of being).  The thin theory, however, entails the unintelligibility of this idea.  For van Inwagen, the idea springs from a clear-cut mistake, namely, the mistake of transforming a difference in nature into a difference in mode of existence.  For van Inwagen, the vast difference between a human being and a rock is simply a vast difference in their natures, and does not imply any difference in the mode of being of that which has these natures.  The idea is not that a rock and a human being have the same mode of being, but that one cannot intelligibly speak of one or more modes of being at all.  The rock exists, the man exists, and to say that is just to say that each is identical to something or other.

I will now given an example which to my mind shows that it is intelligible that there be modes of existence.  We will have to see if I am committing the mistake of transforming a difference in nature into a difference in mode of existence.

Pains and Brains

Phenomenal pains exist and brain states exist.  More generally, there are non-intentional mental states and there are physical states.  But felt pains and felt pleasures and such have a “first-person ontology” as John Searle puts it.  The being of a pain is (identically) its being perceived.  But nothing physical is such that its being is (identically) its being perceived.  This certainly looks like a difference in mode of existence.  Pains exist in a first-person way while brains exist in a third-person way.  

What can the thin theorist  say in rebuttal?  The thins think that we thick-heads illicitly transfer what belongs to the nature of an item to its existence.  So a thin theorist must say that it belongs to the nature of a particular pain that it belong to some particular person.  But this cannot be right.  It cannot belong to the nature of this pain I am now enduring that it be felt by me.  For natures are multiply realizable.  We can of course say that it is the nature of pains in general to be perceived by someone or other. If a pain exists, however, it is a particular pain and it cannot be part of the nature of that particular pain to be perceived by some particular person sich as me.  The dependence of a particular pain on its being perceived is therefore due to its dependent mode of existence and not due to its nature.

Note also that nothing I said implies that the being of the particular pain I am in is a most general activity the pain is engaging in.  My pain is not an agent engaged in an exceedingly quiet activity; it is not an agent at all but a subjective state.

We must also note that the being of my felt pain and the being of your felt pain are numerically different contra van Inwagen's #4 above.

As far as I can see, little or nothing van Inwagen says in  section 2 of his paper  touches the thick theory.  What he has given us is a straw man argument.

(Theme music: Ballad of a Thin Man)

Phoenicians and Londoners agree that 'Some F is a G' and 'An FG exists' are logically equivalent.  Thus, 'Some man is white' is logically equivalent to 'A white man exists.'  But I take a further step: some man is white because a white man exists, where 'because' expresses the asymmetrical relation of metaphysical grounding.   London Ed refuses to take this step and finds my position unintelligible.  He objects:

Now I agree that if 'a white man exists' has a different meaning from 'some man is white', then the question of whether some F is a G because some FG exists, is an intelligible one. But it is not intelligible if they have the same meaning, as London 'thin' theorists claim. After all, the statement

(1) Some man is white because some man is white

is not intelligible. Nor is

(2) Some man is white because the sentence 'quidam homo est albus' is true

For the Latin sentence 'quidam homo est albus' means the same as 'some man is white'. The one sentence translates into the other. So there is no meaningful 'because' here. So why does Maverick think that

(3) Some man is white because some white man exists.

is intelligible? He says as much in his comment #6. So does he think that 'some man is white' has the same meaning as 'a white man exists'? Surely not, for the reasons stated here. But if the meaning is different, what is that difference?

I agree with Ed that if 'Some man is white' and 'A white man exists' have exactly the same meaning, then 'Some man is white because a white man exists' is unintelligible.  That's entirely clear.  So I have to show that the two sentences — call them the some-sentence and the existence-sentence — do not have the same meaning.

We agree that the two sentences have the same truth-conditions.  But sameness of truth-conditions does not entail sameness of meaning.  'X is triangular' and 'x is trilateral' have the same truth-conditions but not the same meaning.  So I hope that Ed is not inferring sameness of meaning from sameness of truth-conditions.

Or maybe Ed thinks that the thin theorist stipulatively defines 'exist(s)' in terms of the particular quantifier.  If that is what the thin theorist is doing, then I grant that the existence-sentence and the some-sentence have the same meaning.  For then they are arbitrarily stipulated to have the same meaning. But then the thin theory is wholly without interest.  Substantive philosophical questions cannot be answered by framing stipulative definitions.  The substantive question is: What is the nature of existence?  If the thin theory is worth discussing it is the theory that "existence is what existential quantification expresses" (Quine), that existence is wholly understandable in  terms of such purely logical notions as the particular quantifier and identity.  Thus Quine explicates 'a exists' in terms of '(Ex)(x = a).'  That existence is a purely logical notion is what I most strenuously deny.

What positive reason is there for thinking that the two sentences have different meanings? Well, 'A white man exists' says all that 'Some man is white' says, but it says more: it makes explicit that there are one or more existing items that are such that they are both human and white.  The existence-sentence is richer in meaning than the some-sentence.  It makes explicit that the item that is both human and white exists, is not nothing, is mind-independently real — however you want to put it.

Or perhaps we could put it this way.  The some-sentence abstracts from the existential aspect of the existence-sentence and so does not capture the whole of its meaning.

Do the gods love the pious because it is pious, or is the pious pious because the gods love it?   That, Ed agrees, is an intelligible question.  And so I take it he agrees that there is a relation we can call metaphysical grounding which is neither logical not causaI.   I say it's the same with the existence problem.  One can intelligibly ask whether some F is G because an FG exists, or vice versa. 

My answer is that existence is metaphysically prior to somemess, and metaphysically grounds it. 

London Ed must have known by some paranormal means  that I was talking about him over Sunday breakfast with Peter Lupu.  For his post upon return from sunny Greece is about the alleged circularity of the thin conception of existence.  Peter and I were discussing Peter van Inwagen's  "Being, Existence, and Ontological Commitment" (in Metametaphysics: New Essays on the Foundations of Ontology, eds. Chalmers et al., Oxford 2009, pp. 472-506.)  Van Inwagen is a Quinean about existence and perhaps the most prominent and formidable of the contemporary thin theorists.  Me, I'm a thick-head: existence is not (identical to) what so-called 'existential' quantification  expresses, and existence comes in modes.  The negations of these convictions I reject as two dogmas of analysis (from the title of a forthcoming paper).

I was lamenting to Peter that I couldn't get London Ed to see my point about circularity.  I now think I understand why Ed doesn't accept it.  It has to do with his not accepting a different notion, that of metaphysical grounding.

Let's start with a Quinean explication of a sentence such as 'Peter exists.'  It goes like this:

1. Peter exists =_df for some x, x = Peter.

What does (1) accomplish?  Well, it shows how one can get rid of 'exists' as a first-level predicate, and with it a reason for thinking that existence is a property of individuals.  For it is clear (assuming that there are no nonexistent objects) that the sentences flanking '=df' are equivalent, indeed logically equivalent: there is no possible situation in which  one is true (false) and the other false (true).

Now in one sense of 'circular' I want to concede to Ed that (1) is not circular:  the definiens — the RHS of (1) — does not contain 'exists.'  In other words, (1) is not circular in the way the following are circular:

X is a human being =_df x has human parents.
Knowledge is the state one is in when one knows something.
Knowledge is cognition.
A book of pornography is one that contains pornographic material.

The following, whether correct or incorrect,  are not circular definitions in the above sense:

Knowledge is justified true belief.
Justice is whatever is advantageous to the stronger.
A circle is a locus a points in the same plane equidistant to some common point.

(1) is clearly not circular in the manner of the above examples: the definiendum is not repeated in the definiens.  So in what sense is (1) circular? (1) is true iff the following is true

1a. Peter exists =_df for some existing x, x = Peter.

(1a), however, is plainly circular.  After all, (1) is  not equivalent to

1b. Peter exists =_df for some x, whether existent or nonexistent, x = Peter.

For if (1) were equivalent to (1b), then (1) would be false.

One response I anticipate Ed making is to say that there is no difference between 'x' and 'existing x': whatever is a value of the one is a value of the other, and vice versa.  If so, then perhaps (1a) collapses into (1) and there is no circularity in the sense in which the examples above are circular.

I would insist, however, that (1) is circular in a different and deeper  sense.  A presupposition of (1)'s truth is that the domain of quantification — the domain over which the variable 'x' ranges — is a domain of existents.  Therefore, if I want  to know what it is for x to exist, you have not given me any insight by telling me that for x to exist is for x to be identical to something that exists.  For of course x is identical to something that exists, namely x!

Suppose we distinguish between semantic and metaphysical circularity.  I am willing to concede that (1) is not semantically circular.  But I do maintain that (1) is metaphysically circular: its truth presupposes that the domain of quantification is a domain of existing items.  To put it another way, the truth of (1) has an ontological or metaphysical ground, namely the existence of the items over which we quantify.

Consider a domain consisting of just three items: Peter, Paul, and Mary. Peter exists iff one of these items is identical to Peter.  Paul exists iff one of these items is identical to Paul.  Mary exists iff one of these items is identical to Mary.  Perfectly true and perfectly trivial.  Although we learn something necessarily true about Peter, about Paul, and about Mary, we do not learn what it is for Peter or Paul or Mary to exist in the first place. 

I want to know that is is for Peter (who stands in here for any individual) to exist. You tell me that for Peter to exist is for Peter to be identical to something.  But in giving this true but trivial answer you have helped yourself to the existence of the thing to which Peter is identical.  You have evaded my question by assuming that we are just given existing individuals.

What form could an answer take?  One answer is that the existence of the items in the domain of quantification is a brute fact and thus inexplicable.  To exist is just to be there inexplicably.  That would at least be an honest answer as opposed to the silly triviality that to exist is to be identical to something.  A radically different answer is to say that for a concrete contingent ndividual to exist is for it to be a divine creation.  Both the brute fact answer and the theistic answer are consistent with Quine's triviality.

Getting back to London Ed, why doesn't he accept my circularity objection to the thin theory?  He doesn't accept it because he is operating with an exclusively semantic notion of circularity which remains at the level of sentences and does not descend to the level of the truth-makers (ontological grounds) of sentences.  (In earlier discussions it became clear that Ed has no clue as to what a truth-maker is supposed to be.) The thin theory, as expressed in (1), however, is not obviously semantically circular: 'exists' is not found on the RHS. All one finds there is a quantifier, a variable bound by the quantifier, the identity sign, and a name that functions in this context as an arbitrary constant.  My claim, however, is that (1) is metaphysically or ontologically circular.  This notion is one that Ed does not understand.

Metaphysical grounding, one of whose forms is truth-making, is for Ed a wholly unintelligible notion.  For Peter and me, however, it is an intelligible notion .  Here I think we can locate the ultimate root of our disagreement.

What say you, gentlemen?