« Spencer Case on Same-Sex Marriage | Main | Saturday Night at the Oldies: Sartorial Songs »

Friday, January 23, 2015

Comments

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

Hi Bill,

Thanks for the invitation to post! Two comments:

The first comment is about one of your key claims in section 3: that if x and y are externally related, then x can exist without y and vice versa. It seems like there are counterexamples to this principle. First, I'm inclined to think that there are necessarily existing externally related things. And van Inwagen certainly does, since he believes that assertibles necessarily exist if at all and that they lack mereological structure. But the principle above entails that there aren't any. Second, I'm inclined to think that I'm externally related to at least one necessarily existing thing, and of course van Inwagen does too. But this this too would be ruled out by the principle above.

The second comment is about the challenge about perception you raise for van Inwagen in section 1. Now, I have no idea what van Inwagen's views about perception are (or even whether he has any). But if I were van Inwagen, I would respond in the following way. (Apologies for the length of what follows.)

I'd begin on insisting one draw a distinction between the sentence "I see that the cup is blue" and "I see the proposition that the cup is blue". I'd insist on this even though "The cup is blue" has as its content the proposition that the cup is blue (outside of attitudinal contexts, at least). I'd insist on this because the second sentence makes a false statement about the object of a particular perceptual state, while the first is a true statement is about the representational content of that state--i.e., which, on my view, is a proposition that describes the conditions under which that state is veridical, and not the object perceived.

Here's a natural view, one that I'd recommend to van Inwagen: take the perceived object to be whatever it is that makes the state's representational content veridical. Now, van Inwagen does seem to believe that there are events, and what he says about them seems to suggest that they're concrete, spatiotemporally located particulars (Material Beings, pp. 82-83). And presumably, he'd deny that events so conceived would have assertibles as constituents, much less the relation of instantiation. So, I'd also recommend that he take these to be objects of perception in the type of case under consideration, since they seem to avoid the worries you raise with taking propositions to be objects of perception.

Finally, then, I recommend that van Inwagen say that one sees the cup's being blue in virtue of some relationship between a particular event involving this cup and a perceptual state--one that that has, as its representational content, the proposition that the cup is blue.

Very curious to hear your reactions...

Thanks for commenting, Alex. You really know your stuff.

Your first point is correct. My formulation was sloppy. Suppose I say this: if x and y are externally related, and if one or both of them is a contingent being, then either one of them can exist without the other, or each of them can exist without the other.

Your second comment is deep and challenging. You distinguish
1. I see that the cup is blue
from
2. I see the proposition that the cup is blue.

I take (2) to be equivalent to
2*. I see the saturated assertible expressed by 'that the cup is blue.'

You say, very plausibly, that (1) is true and (2) false. You say that (1) is true because it is about the representational content of the perceptual state that one is in when one looks at a blue cup. You say that (2) is false because it is about the object (as opposed to the content) of the perceptual state.

So we have the following picture. There is a subject S of a mental act A which has a content C -- this content being a Fregean proposition -- through which a mind-independent object O is presented. A is concrete and intramental while O is concrete and extramental. C is a abstract and extramental.

So your point is that S sees (literally) the cup, a concrete thing, but does not see (literally) the abstract proposition that is the content of S's act of seeing the cup, a content that mediate S's perceiving of the cup.

Then you say that the perceived object is what makes true the representational content. But here is the rub. If O is the T-maker of C, then O cannot be an Armstrongian blob: it cannot be ontologically stuctureless (for the reasons I gave in that long paper I sent you). But PvI's concrete objects are all of them blobs. He has no truck with constituent ontology.

As I understand PvI, there is no place for events in his ontology. Everything is either an abstract object (a relation-in-intension) or a concrete substance. Events have ontological constituents, but no substance for PvI has ontological constituents, and O must be a substance. At most, concreta have mereological parts in the strict sense of classical mereology.

What you are suggesting, I think, is that van Inwagen take perceptual objects to be something like Armstrongian states of affairs (which of course are not abstract and therefore not the same as Fregean propositions). But PvI's ontology does not allow for such critters.

I hope the above is clear, and that I have understood you. Your counterresponse is welcome.

By the way, your comments can serve as a model for what good comments are like.

Hi, Bill. Thanks for your substantive summary and critique. Here are a few questions as I begin to think through this post. I'm just starting to reflect on this -- maybe I'm missing something.

I understand how an assertible can be uninstantiated. But can an assertible be necessarily uninstantiated? An assertible is a thing that can be said. I assume this means "meaningfully or intelligibly sayable" rather than merely "utterable."

I'm inclined to believe that if a thing is meaningfully assertible, it is intelligible, and if intelligible, it is about that which is logically possible. (I don't say that everything logically possible is assertible, but that everything assertible is about something that is logically possible.) I'm also inclined to believe that if something is logically possible, that something is possibly instantiated. If these assumptions are right (perhaps they are wrong), then wouldn't all meaningful assertibles, thus all meaningful properties, be possibly instantiated?

And if there are such things as necessarily uninstantiated assertibles then wouldn't this mean that there are necessarily uninstantiated, necessary beings? If so, this seems very strange. It seems to be a view which holds that at least part of reality necessarily exists but is necessarily unrealizable, and thus that reality is in a sense necessarily incomplete.

Thanks for the very interesting post, Bill.

I'm going to comment on both your exposition and critique, but I'll try to focus more on the critique. I'll begin with the exposition.

It does not surprise me at all that van Inwagen is coy about telling us what 'abstract' and 'concrete' mean. I don't know of anyone who clearly states what he means by these terms, but typically something similar is meant. I personally am doubtful that the distinction is a helpful one, and I'm doubtful that these terms "carve nature at its joints." I have always thought that 'concrete' is more or less supposed to track physical objects, that is, objects studied by the natural sciences. But van Inwagen includes God in the category of concrete objects, which strikes me as bizarre. God is certainly not like me. And, if we defined 'concrete' as physical objects, it would be nice to have a plausible account of what it is to be physical, and I've not heard a promising answer. And, what would we make of objects such as my gold truck stripped of its goldness? Is that a concrete object or an abstract object? I don't know. It's an awful lot like many objects that are often called concrete, in that it has a shape, a mass, and a finite extension. But, it lacks a color, which a lot of the concrete objects we use as examples do not. Which is it? So, maybe we can use this distinction, but I don't think it helps much.

Setting aside my dislike of the terms 'abstract' and 'concrete', I don't see why he thinks the difference between abstract things and concrete things is so puzzling. Is it puzzling because he enters his investigation of what there is with an assumption that everything that there is is very similar? Why should we enter the investigation with that assumption? I don't know.

Clarification question: What are the instantiation relations external to? I assume the answer is that they are external to their concrete bearers because relations are in platonic heaven, but not the concrete bearers. I wonder about abstract objects that have multiple properties. Is the intantiation relation external to them as well, despite the fact that they are also in platonic heaven?

I do not think I understand, or could explain to someone else, what you mean in (i) by "...properties for van Inwagen are logical fallout from one-place predicates." Nor do I understand what you mean when you say that properties "outrun" one-place predicates.

I will now move to comments about your critique. In these comments I will try (somewhat) to answer for van Inwagen.

I'll begin by trying to resist your argument that, given PvI's view, coffee cups are colorless. You say, "For if colors are properties (179) and properties are abstract objects, and abstract objects are colorless (as they obviously are), then colors are colorless, and whiteness is not white and blueness is not blue." I'd like you to formalize this argument. Here's one attempt:
1. Coffee cups have a color (premiss).
2. Colors are properties. (premiss)
3. Properties are abstract objects.
4. Therefore, colors are abstract objects. (2,3 transitivity)
5. Abstract objects are colorless. (premiss)
6. Therefore, colors are colorless. (4,5 transitivity)
7.Therefore, coffee cups are colorless.

It doesn't appear to me that 7 validly follows. Is there a way to formulate the argument so that it does? van Inwagen can resist this argument on the grounds that it is invalid. There may be an opening for him to say that, despite the fact that colors are colorless, coffee cups have a color.

I wonder if your criticism here is rooted in thinking that it is "in virtue of" color properties that coffee cups have their coffee. Well, van Inwagen thinks that properties are non-causal, so he might resist your here too. Properties do not cause their objects to have said properties. This sort of response to your perception worry seems to me to lead to another problem. If properties do not cause, then what is the nature of this "in virtue of" relationship? What work are properties doing on this account, anyway, if it is not color properties that make the object that possesses them have color? It is hard for me to understand how properties can account for common features if they are causally inert.

Continuing with the perception critique, you say, "How can he say that we don't see the property but we do see the proposition?" I didn't take him to be saying that we can see the proposition, 'that the cup is blue'. I thought he was merely saying that we can see that the cup is blue. That is not to say that we see the proposition. I think he would say we see neither the property nor the proposition. You ask then, "If van Inwagen says that we don't see the proposition, then what do we see when they cup is blue? I think PvI's answer would be, "You merely see the object--the blue cup." Remember that properties are not parts or constituents, so it isn't as though the properties are parts for you to see as the grill of my truck is. Objects are not bundles of properties on his view (perhaps characterizing objects are bundles would allow for you to think of properties as being like grills on trucks, since a bundle is made up of properties that constitute the bundle). (Honestly, I don't like what I'm saying here, but I'm trying hard to defend PvI, ha)

I think I'm suggesting a response, according to which, you do not see the property when you see the blue cup, but you do somehow see the affect it has. But how? This strikes me as a total mystery, and it might be unsolvable if properties are truly causally inert.

I think your first paragraph of the 'But is This Ontology?' section makes a nice point that relates up to your perception critique. Since properties are not objects of sensation (because they are in Plato's heaven), the connection between abstracta and concreta is unintelligible. This also relates to my points above about causation, and talking about causation might be another way of making your point about the connection being unintelligible. How is it that the coffee cup is blue "in virtue of" the property 'blueness' if there isn't some causal relationship? We know the connection cannot be causal.

There is one place where I might disagree with you in this section. In the third paragraph you are making the point that if we read off our ontology from the sentences we accept, then it is logically possible that there are no properties, since it is logically possible that the sentences we accept are false. Right. Then you say, "The following is not a contradiction: The sentences we accept as true entail that there are properties & There are no properties." How could THE SENTENCES, in and of themselves, ENTAIL both of these things (sorry, I want to use italics, but can't, so I'm using caps). Given that it is possible that the sentences that entail that there are properties could be false, it is logically possible that there are no properties. But this possibility isn't entailed by the sentences, is it? It's entailed by the possibility that the sentences are false.

My last point regarding this section is on whether what van Inwagen is doing should be regarded as ontology. I take your points about van Inwagen's method being 'subjective' and 'modern'. Maybe you are right that this doesn't deserve to be called ontology, but it seems to me that it is the kind of think that contemporary analytic philosophers call ontology. Perhaps that is because contemporary analytic philosophy isn't always mindful of the history of philosophy.

Moving on to criticism 3, I agree that van Inwagen is a bare particularist. Here is a question for you about van Inwagen's kind of bare particularism: If God severed the relations between my bare particular and my external properties, and if he did the same thing to the tree in my backyard (assuming for a moment that trees are things), wouldn't it be the case that there would be no difference between my bare particular and my tree's bare particular? That seems to me to be what should follow from van Inwagen's kind of bare particularism, since, according to the view, bare particulars do not have natures or essences.

Here is a clarification question for criticism 4. You say, "For given the externality of the instantiation relation, both Socrates and the putative property must 'already' exist for said relation to hold between them." Is the "'already' exist" a logical priority, a temporal priority, or both? I suppose if Socrates and the putative property do not both exist, the trouble is that there isn't enough there for the instantiation relation to relate, right? If what I said is right, then I'm tracking the problem and I agree with you.

Lastly for this section: "...metaphysics without explanation is not metaphysics at all in any serious sense." YES!

I hope these comments are helpful.

Good comments, Elliot.

>>I'm inclined to believe that if a thing is meaningfully assertible, it is intelligible, and if intelligible, it is about that which is logically possible.<<

That's reasonable, but think of it this way. The open sentence or predicate 'x is both round and not round' is not gibberish in the manner of 'x is qroolublish and not qroolublish.' So the open sentence or predicate is meaningful as opposed to meaningless. If I put 'Obama' in for 'x' I get a necessarily false (closed) sentence. If I assert the sentence, I say something intelligible (understandable). If it weren't intelligible you would not be able to assign it a truth value, the value false. Now it is logically impossible for anything, including Obama, to be both round and not round. So the property corresponding to the predicate is necessarily uninstantiated.

>>And if there are such things as necessarily uninstantiated assertibles then wouldn't this mean that there are necessarily uninstantiated, necessary beings? If so, this seems very strange. It seems to be a view which holds that at least part of reality necessarily exists but is necessarily unrealizable, and thus that reality is in a sense necessarily incomplete.<<

Yes it would entail that. And I agree that it is somewhat strange. But strangeness is no objection. And I can grant that a bit of reality necessarily exists but is necessarily unrealizable, i.e., uninstantiable in concreto. But it doesn't follow that reality is incomplete.

Analogy: would you say that God is not completely omniscient because he does not know false propositions? (If you don't immediately see the relevance of this analogy, don't object to it; it would only lead us off topic.)

Thanks for the comments, Mr Shields.

>>I do not think I understand, or could explain to someone else, what you mean in (i) by "...properties for van Inwagen are logical fallout from one-place predicates." Nor do I understand what you mean when you say that properties "outrun" one-place predicates.<<

When I say that properties outrun predicates I mean that there are far more properties than there are actual predicates. I take it that predicates are essentially linguistic in that they are tied to natural languages. And I take it that in English (and in every natural language) there is a finite number of actual predicates, where an actual predicate is one that has been used by some person at some time either in overt speech or it thought. For example, up until just now there was no predicate 'x is a roodle.' But I just introduced it. It is true of every red noodle.

If properties are necessary beings, then there are as many properties as it is possible that there be. For every real number r, there is the property of being r units long. If so, then there are 2-to-the-aleph-nought properties. But the number of predicates of all natural languages taken together is finite.

The 'logical fallout' remark simply means that for every predicate with a definite sense, there is a corresponding property. So, corresponding to 'x is a violinist if a card shark' there is a property.

Further responses tomorrow. It's time for Saturday Night at the Oldies!

Thanks, Bill! I see that your open sentence is meaningful/intelligible. And your analogy is very good!

Omniscience is normally defined as the property of knowing all true propositions and believing no false ones. So I wouldn't say God lacks complete omniscience because He doesn't know false propositions, like I wouldn't say God lacks complete omnipotence because He doesn't create unmarried married men. Neither case shows a meaningful limit to the omni-attribute. I see how this relates to reality. It's a helpful analogy!

I agree that strangeness is no objection. After all, it may be that reality is quite strange to us finite minds. Shakespeare may be correct when he has Hamlet say that there is more to reality than we dream up in philosophy!

>> I agree that van Inwagen is a bare particularist. Here is a question for you about van Inwagen's kind of bare particularism: If God severed the relations between my bare particular and my external properties, and if he did the same thing to the tree in my backyard (assuming for a moment that trees are things), wouldn't it be the case that there would be no difference between my bare particular and my tree's bare particular? That seems to me to be what should follow from van Inwagen's kind of bare particularism, since, according to the view, bare particulars do not have natures or essences.<<

I'm glad we agree. Of course, he denies that he is committed to bare particulars because he gives 'bare particular' an absurd interpretation. He thinks it means a particular with no properties. Nobody, not Bergmann, not Armstrong, nobody, thinks that there are particulars that have no properties.

Could God disconnect a particular from its properties? I do not think so. For even if it is contingent which properties a particular has -- since BPs lack natures or essences -- it is necessary that a particular have some set of properties or other.

As for you and you tree, the BP in you and the BP in your tree are barely different, they differ *solo numero,* i.e., they are just numerically different. They do not differ in respect of some property internal to the BP (there are no such internal properties). So the BP in you and the BP in your tree do differ, but they differ only numerically, not qualitatively.

Thanks for the clarification of "outrun" and "logical fallout". I understand what you mean now.

What you say about God being able to disconnect a particular from its properties makes sense. I think I was aiming with that question to show that van Inwagen's bare particulars are merely numerically different. Of course, it is possible that that is the way bare particulars are. It seems it follows from this view that if my bare particular were somehow switched out for another one, there would be no change in me, except a numerical change in my bare particular. Again, that is possible, but it strikes me as odd.

It is odd that he denies he is committed to bare particulars of some kind. Since he is not a bundle theorist, he needs something for the instantiation relations to relate the properties to!

And that something is the concrete particular itself, a tree say. Suppose the tree is green like my palo verde. The greenness is an abstract object in Plato's heaven. But then the tree here below is bare in respect of that property.

His entire conception is absurd. The properties of a thing cannot be stripped off the thing and installed in some realm disjoint from the realm in which the concrete thing exists.

Yes, that's what I think too. I'm curious about what you would say about my entailment comment. Am I misunderstanding something?

Hi Bill,

Many thanks for your very kind words -- the feeling is most definitely mutual! Two quick reactions to your responses (forgive me if this has already been covered in the discussion above).

Regarding your reformulation of the link between external relatedness and modal connectedness: my worry now is that the reformulated principle allows for van Inwagen to slip through. Suppose that Socrates and the property being human are externally related. Then the reformulated principle only requires that it be possible for at least one of the two to exist without the other. To meet that requirement, van Inwagen only needs to allow for it to be possible that the property being human can exist even when Socrates doesn't (which of course he would). But that's consistent with it being impossible for Socrates to exist unless the property being human exists too, and (presumably) as well for it to be impossible for Socrates to exist unless he is human. So, even if the reformulated principle is true, van Inwagen could still maintain that Socrates is externally related to properties he has non-accidentally.

Regarding your comments on van Inwagen and perception: suppose I grant that van Inwagen won't allow events into is ontology (or at least, events of the sort that would serve his purposes here). I wonder whether I could push my response without them. Couldn't van Inwagen maintain that it's in virtue of some (at least partially) causal relationship between the cup and a particular perceptual state of mine -- one whose representational content content is that the cup is blue -- that I see that the cup is blue? That is: why couldn't one take the perceived object to be just the cup -- and not, as I was suggesting, an event involving the cup -- but take my perceiving that it's blue to be in virtue of some (at least partially) causal relationship between this perceptual state and the cup?

One response might be to say that without allowing for the existence of events (or states-of-affairs or etc.), van Inwagen cannot even make sense of apparent truths like "there is a causal relationship between a perceptual state of mine and this cup". If that's so, the van Inwagen is in even worse trouble than you allege! But I suspect van Inwagen would (claim to) be able to make sense of them in the same way he (claims to) make sense of other relational facts.

I could just be misunderstanding the force of the challenge you're raising for van Inwagen here, so please correct me if I've gone off the rails anywhere here!

All the best,
Alex.

Thanks for the further comments, Alex.

I am assuming that when one looks at a blue cup and sees that it is blue, one sees not only the cup but also something that is blue, whether it be a blueness trope, or an immanent constituent universal, or a blue noema, or something else that is blue, right at the cup. I suspect that van Inwagen would deny my assumption and, following Chisholm and a few others, adopt an adverbial theory of sensing. I have quoted him above as saying that properties are not objects of sensation. And of course they cannot be if they are abstract objects. And yet things have properties. So the cup is blue despite the fact that the blueness of the cup is not an object of sensation. So what's going on?

Perhaps this: the concrete thing causes me to sense blue-ly. If this makes sense, then there needn't be anything that is blue that I sense. This may be a version of what you are getting at above.

There is the property blueness in the realm of abstracta, but it is not blue. And there is no constituent of the cup that is blue such as a blueness trope. (PvI is not a constituent ontologist.) So, despite the phenomenology of the situation, blueness does not appear when I look at the cup. I am being appeared to bluely or bluely appeared to.

Is this coherent? I am running slowly iff my running is slow. Note that we cannot say: I am sensing blue-ly iff my sensing is blue. For it is nonsense to say of a state of consciousness that it is blue. If my sensing were blue it would be in space, but it is not. So what is it for me to sense blue-ly if my sensing is not blue?

Furthermore, if I am being appeared to, then I am being appeared to by something. If am being appeared to bluely by something, then the something is either blue or not. Can't be blue, else we are back at square one. So I am being apeared to blue-ly by something that is colorless. But then what makes it the case that my sensing is veridical?

The comments to this entry are closed.

My Photo
Blog powered by Typepad
Member since 10/2008

Categories

Categories

September 2024

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          
Blog powered by Typepad