Saturday, September 27, 2008

Concern Post #2

Maybe I can state my concern over AEs and propositions to a general concern over Types and Tokens.

I seems that Mr. Realist would want to hold that:

red rot rouge

all are instances of the same word-type. The 'redness' type. I'm not totally sure how they would go about expressing this, probably just with saying: these words all express redness. Likewise:

△ ▽ ▷ ► ▼ ◬ ▿ ◿ ▲

all are instances of ▲ity, or triangularity.

Now it seems that Mr. Realist is going to have some problems. Maybe they are just small and bred by my personal confusion. But 'redness' seems to be an English word. If there is an English and a German red-type, this seems to be really the talk of a English and German red-token-class. (Same for the triangle case, with the variety of shapes and designs.) Thus, the Sellarsean move to introduce honest-to-God types: 'red' in English, 'rot' in German, 'rouge' in French all play the same role. Each are •red•s.

Mr. Realist cannot be happy with this, I don't think. Triangularity is supposed to be an abstract entity, not some sort of functional class! Likewise, the sign-designs *△*, *▽*, *▷*, etc. are all of the triangular-kind. Mr. Nominalist wants to say that each shape is called triangular, where Mr. Realist wants to say that each shape is an instance of a three-sided closed-plain figure, i.e. instances of triangularity.

To claim that each shape plays the triangular role, that each is a •triangle• cannot make Mr. Realist very happy at all. We might say that the German 'dreieck' and English 'triangle' each play the same linguistic role: each are •triangle•s. Likewise,

△ ▽ ▷ ► ▼ ◬ ▿ ◿ ▲

are each distinct token-classes of the triangle type. Each shape stands for triangularity, in that: "This is a triangle" is true of each shape.

Mr. Realist cannot seem to be happy about this at all.

It seems that Mr. Realist wants the type to be an entity in the full-blooded sense, distinct from its tokens. The type should be able to be real (or on some accounts, exist) even if its tokens do not exist (or are not real, on those same accounts). To claim that types are only functions seems be incompatible with Mr. Realist's general philosophy. Mr. Realist wants us to have to compare things in the world against things in Platonic heaven to see what they really are. We have to compare instances and exemplifications with the damn Universal! Not with a function!

Certainly we can claim that types or numbers are real, even if they are just functions or structures, but I don't think we are being full-blooded, honest-to-God realists any more. What need have we of Universals when we outsource their explanatory role to functions?

Also, I see a problem in that presumably "the universal blah" and "an instance of the universal blah" seem to be rigid designators, where "the structure that plays the role blah" does not seem to be. Consider:

The killer of Jones is Smith.

It seems that there could be possible worlds where Smith didn't kill Jones. So the killer of Jones isn't a single entity which we can necessarily identify with Smith. Other people could play the killer role. Or no one could. Doesn't the same problem crop up if we want Universals and their instances? It seems that if we appeal to a structuralist account, then in any world, we can literally identify a different entity with the structure! This cannot make Mr. Realist happy at all. That



is an instance of ◣ity seems to be necessary, not just a matter of the shape fitting a role! But this seems to be false, doesn't it.

13 comments:

Dan said...

Let me try and parse down what I think you're getting at.

1) 'red' is just a word
2) 'rot' means 'red'
3) •red• = •rot•
And that's all there is to it. Since the dot quotes denote a functional role, and all universals can be explained in terms of them, there is no need for universals. Is this something like what you're getting at?

If this is the case, I'd say you'd have to take on the platonist on a case to case basis. One could easily be a nominalist about some things, and a platonist about others. By the same token, it may be easier to identify some entities with structures of functions than others. For instance, this argument would be useless as a reduction of platonist functions and structures, but one could be a realist about structures and functions and use these arguments to reduce properties and propostions to functions and structures.
A realist about properties and propositions my have independant reasons for resisting identification of these entities with structures or functions(as I think they do in the case of properties), but that's not really the concern here. Platonism about propositions is just the thesis that propositions are platonistic entities. If propositions are identified with platonistic entities we already accept, all the better. That's not a blow against realism, that's just being theoretically economical.
The argument would have a solid impact if it were shown that functions and structures are not platonic entities, however this seems difficult. Both seem to be multiply instantiable, not located (especially functions), and not particularly dependant on their instances. This argument leads down different roads (in the past, I've found it leads to debates in modality, and modal epistemology).
The point about rigid designators I think it an important one. However since most of these views boast being necessarily true, if true at all, it's not an immediate concern.
This has been mostly a meta-theoretic comment, but a note about propositions in particular might be worthwhile to keep in mind. So far NONE of the views we've seen have posited propositions as unique entities of a kind we've never seen before. We've seen views of which they are sets of circumstances, fusions of properties, relations and individuals, mathematical constructs, structures etc. The closest thing we've seen to a genuine unique entity being a proposition is Frege's view of propositions being sets of senses, but we haven't considered it as a live option in particular as of yet. ALL of these views could be correctly seen as realist views.

Wes McPherson said...

Thanks Dan, your comments have been elucidating.

I suppose I'm hoping that one could be a realist about things, from cabbages to numbers, without being a Platonist about them as well.

I personally find the various strategies against Platonism to be successful, which leads me to hope that propositions are not abstract entities of a Platonic sort. (Especially if we can explain abstract entities in terms of functional classification, I would hope propositions are not abstract entities.)

Certainly they are multiply instantiable, but I think they are located and dependent on their instances. If nothing else, our concepts of them are located in our theories and models, and if our theories and models are realist and not merely instrumentalist they have to be representing something.

Adam said...

Nice post and response guys.
I think Dan’s point that dot quotation does not succeed in reducing all Platonic entities to functional or linguistic roles is a good one. I thought I’d try and bolster it, since it seems like the chief problem facing that reductive view. The problem facing the Sellarsian in this case seems to be similar to a problem faced by someone who thinks that all qualitative properties and relations can be reduced to tropes. These philosophers think that no object ever shares a property with any other object, but similarity and difference can be explained in terms of perfectly similar tropes (which are basically just necessarily uniquely instantiated properties). Trope theorists have a really hard time explaining similarity. Similarity relations are either Platonic, or are themselves tropes, or they are primitives. The first option seems to deflate the trope theorist’s project. On either of the two remaining options, we get an ontological explosion- in the form of an infinite regress of higher-order tropes if we go the first way, or by the need to posit a unique (first order) trope for every distinct occurrence of ‘similarity.’
This is I think where the strength of Dan’s point comes in. Sellars is arguing against the view that Platonic universals are the semantic values of predicates. He thinks that we can eliminate commitment to universals by reducing the linguistic expressions used to denote them (predicates) to functional roles within a language. But what explains the fact that .female fox. and .vixen. play the same functional role in English, or that .red. and .rot. play the same role with respect to English and German? The linguistic device ‘plays the same functional role as’ seems to denote a Platonic relation that the members of each pair bear to each-other. The Sellarsian has three options: analyze this linguistic item as denoting a Platonic universal holding between dot quoted pairs; as itself denoting some distinct functional role; or as denoting a theoretical primitive. The first option seems to entirely deflate the Sellarsian program. The second will generate a regress problem. And the third generates commitment to potentially infinitely many distinct functional roles (and that’s just with respect to one language).

Adam said...

Another quick question each for Dan and Wes. Wes is looking for a way to be a realist about abstract objects without being a Platonist about them. Dan’s worry is that entities like functions seem like Platonic entities, in virtue of being multiply instantiable and unlocated, etc.
But I’m wondering if we could be realists about universals, without being Platonists about them, and get functions as well. If a function is supposed to represent some kind of ordered mapping of entities from a domain D of arguments onto a set S of outputs, or extensions, then it seems like for any function F, we can identify a structured, dyadic universal U such that D stands in U to S (in virtue of each unique element of the argument domain of F standing in some distinct relation R to some unique output of F). We could then drop Platonism and just say that universals- including the universals we are identifying with functions- are wholly present in their instances. Would this work?

Wes McPherson said...

Hi Adam,

Good posts as well. I'll just quickly add that my main discomfort with Platonism is how the heck we would ever know about them without some grasping of some sort going on.

And ask: If we were to hold that Universals are wholly present in their instances, would this move by way of analogy to holding that types are wholly present in their tokenings? This does seem so crazy.

Dan said...

Hi All,
Sorry for my brief absence, I've been high-holidaying.
I think what Adam and Wes are eluding to is the first step down a road to modal arguments and intuition pumping. Basically, if you want to identify a universal with some bunch of physical stuff, there are modal arguments that can be brought against it (the universal could've instantiated different stuff).
The other strategy, granting a universal existence but saying it's co-located with its instances also has some counter-intuitive results. For instance, on that view, since I have a mostly blue background on my computer, I made the universal of blueness bigger by turning it on. Also, it would actually be possible to wage a war on terror, just by making sure people aren't terrified. This isn't knock-down of course, but I think it's a pretty odd way of looking at things.
To bolster the point a bit. Even if you had a universals-are-located-just-where-their-instances-are view, it would still be the individuals themselves playing the causation game. It would be even weirder to say that a blue individual's blueness caused something, but that individual didn't. I find it much more natural to say that such an individual caused something, perhaps in virtue of it instantiating blueness. I think granting something location without granting it causation is just grasping at straws, trying to make something less mysterious.
In all honesty, I don't think there's anything mysterious about many platonic entities. Even our knowledge of them can be very strait forward. I wrote a post elaborating on this earlier this term.
http://36507160novelproposition.blogspot.com/2008/09/objection-to-swoyer.html
Wes and I had some discussion on it, but it went off-topic.

Wes McPherson said...

Hi Dan,

What if one were to subscribe to objectless sensations (akin to objectless processes)? Then we would run Lockean primary and secondary qualities together and into the minds of perceivers. Then I would have a colour or shape, but strictly speaking no object...

Adam said...

Hi Dan.
I think that the objections you raise are directed at only one of the two views you bring up. On the first view, a universal is just the fusion of its instances. If one held the view that a universal was the fusion of its instances, then I think there would be a worry in that it seems clear that a universal could have had more instances than it in fact has (although I think there are probably things one could say in this case to defend the view).
The view that universals are fusions of their instances is also subject to the other worry, that creating an instance of a universal (by turning on your computer monitor) makes a universal “bigger” (by adding a part to it). This does seem like a weird way of looking at things.
But the view suggested doesn’t look at things this way. On the suggested view, universals are instantiated in individuals (just like on the Platonistic account). Nothing about this account requires us to say that a cape’s redness causes the bull to charge, and that the cape doesn’t; instead, we ought to say just what the Platonist says, that the cape causes the bull to charge, in virtue of instantiating redness. The only difference is that, on this account, redness doesn’t exist anywhere else but the cape (and in other things of the same colour).
So it seems like the real worry is that if we grant that (i) universals are wholly present in their instances, but that (ii) universals are only derivatively causal, then we are saying that there are some things that are located but are causally inert. This does seem weird. But it also seems to be the exact thing the Platonist says. Suppose universals are as the Platonist says they are, and suppose the universal redness is instantiated in some matador’s cape. This means that a particular instance of redness is located in the matador’s cape. Does it cause the bull to charge? Probably not.

Dan said...

I'll respond to Adam's comment first, since it's more my area.
You're exactly right in your characterization of the view I was suggesting. The objection was that there is no point in positing a location of a universal unless you're also positing a causal role of that universal. It seems less weird to posit a strait up platonic entity than to posit an entity that is concrete in the sense that it has location, is co-located with normal concreta, and has the causal role that a platonic entity has. It seems like you're throwing a bunch of conditions on this thing (that are unnecessary) to make it look more concrete, when (I think) a better approach would be to investigate what the nature of this is assuming of it only what is required for your theory.

Wes, let me know if I get your proposal right. We treat sensations as properties that have no object associated with them? Well, one way to go would be to treat them as objects (sense data theory). In that case I think you still have the object-property dynamic, since some sense-data have the property of being blue, and some don't. Or you could treat them as objectless bundles of properties with no "bare particular". I'm not sure this is any different either. You can call the bundles "objects" and say that an object instantiates a property iff the property is in the bundle.
Remember that the platonist has a very loose criterion for what it takes to be an object. Pretty much anything gets in, so it's hard to argue against a platonist by positing things that aren't objects. One platonist view is that anything that instantiates a property is an object. Using one small premise that everything has the property of being identical to itself makes everything an object (including properties).
I'm not sure I've addressed the question, was that ok?

Wes McPherson said...

Dan,

I was thinking of an adverbial theory of sensation. But a general case of pure process will do. Like if I blare my music loud, it seems that the loudness can cause my mom to have a headache. Doesn't the loudness play a causal role? If I flash red and blue at a epileptic child and they seizure, didn't the red and blue cause the seizure?

It seems that in some cases it isn't so odd to say that a property of an individual (like the volume of a stereo or the colours of a screen) causes something to happen.

For interest, an adverbial theory thinks that if Jones sees blue, we represent this not as aRb, but Fx. The analogy would be that if Jones goes running, we write Fx and not aRb.

Thus, 'Jones sees blue' becomes 'Jones sees bluely' or 'Jones sees (blue)ly'. It is the seeing which is blue on this account. (Or, regions of Jones' visual space. The same region of visual space may be red now, blue now, etc.)

Also, I'm curious about 'making a universal bigger.' If I have a set and I add something to it, doesn't the set get bigger in a sense? Even if, of course, in a sense it stays the same size.

Dan said...
This comment has been removed by the author.
Dan said...

Hi Wess,
I think we're leaving the realm of argument and entering the realm of intuition bumping, but that doesn't mean there's nothing left to say.
As for properties causing things, I generally leave it up to the scientists to say what causes what. Physical theories would say that a wave of high amplitude with a frequency within human hearing range caused your mothers headache. There's no need (or even warrant) to attribute the cause to the properties. The common sense view is that the properties gave the individual certain causal capacities, but themselves did not cause anything. I see no reason as of yet to deviate from that.
Another reason to think it was the individual causing stuff: Having causal powers is itself a property. If it were the property of the properties of the individual who had causal powers you'd need a second order property to accound for that. As things stand in the common sense view, a certain causal capacity can simply be identified with the property in question (much of the time), avoiding needless multiplication of entities.
The adverbial theory is interesting. I'd say it has a nice added simplicity, assigning perceptual properties to an individual. I'd be all for that.
As for making sets bigger, that's controversial, yet plausible. My only response to that would be to say that sets aren't universals (they're not instantiated, if located they're probably not multiply located). Sets are weird because impure sets straddle the line between concrete and abstract. However I do take your point, making an abstract object bigger isn't incoherent. That was a personal intuition.

Wes McPherson said...

Dan,

I'm not sure what intuition pumping is, but I do know that I am bad at argumentation, so I wouldn't be surprised if I were intuition pumping. Though I'm not entirely sure what that is.

I guess this is more intuition pumping: when I hear a C# I might get a headache. It seems that it is the C# that causes the headache.

Maybe a physicalist says that really the C# is blah blah vibrations and sound-wave patterns. Those are identical to C#s. Those blah blahs cause c-fiber stimulation (or whatever) which are identical to pain. So the pain and the C# are really physical stuff.

But what if we resist this physicalism? Maybe the blah blahs cause the C#. Maybe the whatevers are caused by them, and cause headache pain. I guess on this view the pure C# and the pure headache pain are epiphenomenal.

But what if we resist that physicalism too. What if we have an event ontology / process metaphysics. Then we have a happenings of the C# kind. It is objectless in an important sense, though we would derivatively add objects I suppose. So then the happenings of the C# kind cause a happening of the headache-pain kind. It is also objectless in an important sense, since we are operating within an event ontology.

But I suppose then that here we just treat 'events' as 'individuals' and you original point remains?

I guess I'm just intuition pumping away. Pump pump pump...