Must We Stop Asking Why There's Anything?

1. A Pseudo-Question:  How Many Things are There?

A while back, in response to a reader's question, I argued that one cannot count things as things.  I can count the cats in my house, the tiles on my roof, and 'in principle' the subterranean termites within two feet of my foundation.  What I cannot do is count things (entities, beings, existents, items, objects, etc.) as things. The reason, briefly, is that 'thing,' unlike 'cat,' is not a sortal.  It is a dummy sortal.  It is a term that functions grammatically like a sortal, and can serve as a placeholder for a sortal, but is logically unlike a sortal in that it supplies no criteria of identification and re-identification for the things to which it applies.  I can count the cats in my house because I know what it is to be a cat; I know what 'counts' as a cat and what does not 'count' as a cat: the lifespan of a cat is not a cat; the location of a cat is not a cat; the posture of a cat is not a cat; the parts of a cat are not cats; the properties and relations of a cat are not cats, etc.

But I can't count the things in my house because I don't know what all counts as a thing, and what all doesn't.  Are only middle-sized specimens of dry goods things?  Or are persons also things?  Are my cats things?  Is their blackness a thing? Everything black is colored.  So do I count the cat as one thing, its blackness as a second thing, the being colored of the cat as a third thing?  If I have a cat on my lap, do I have at least three things on my lap, or only one, or perhaps a countable or (heaven forbid) an uncountable infinity of things on my lap?  And what about the parts of the cat, and the parts of the parts, and how far do we go with that?  To the molecular level. the atomic level. the quark level?

I trust the point is clear: one cannot count things (entities, etc.) as things.  It seems to follow that the question 'How many things are there?' is a pseudo-question.  It is a pseudo-question because it is unanswerable in principle.  'Many' and 'more than one' are not answers.  A Parmenidean monist might insist that there is exactly one thing, and a nihilist that there are none.

2. Why is There Anything at All?  A Pseudo-Question?

But now a vexing question arises:  does the fact that 'thing' is a dummy sortal, hence not a sortal, constitute a reason for holding that the question 'Why is there anything (any thing) at all?' is also a pseudo-question?  Stephen Maitzen answers in the affirmative in his paper, Stop Asking Why There's Anything. (Thanks to Vlastimil Vohanka for alerting me to the article.)

Maitzen seems to be reasoning along the following lines.  We can sensibly ask why there are apples, trees, plants, living things, and sensibly expect a natural-scientific answer.  But we cannot sensibly ask why there are things (existents, beings, etc.).  The same goes for the restricted question why there are any contingent beings.  This is because 'contingent being' is just as much a dummy sortal as 'being.'   Dummy sortals are referentially indeterminate unless replaced by a genuine sortal such as 'penguin.' 

Maitzen's point could be put as follows.  There are various sorts of thing, and of each sort we can sensibly ask: why are there things of this sort?  But we cannot sensibly ask: why are there things at all, or contingent things at all?  Things that are are not a sort of thing. And the same goes for things that are contingently.

So perhaps the point is simply this.  'Why is there anything at all?' is a pseudo-question because (a)  things that are are not a sort of thing, and (b) we can sensibly ask the 'why' question only about sorts of things.

3. Tentative Evaluation

Well, I think it is perfectly clear that things that are are not a sort of thing.  Aristotle said essentially that long ago when he said that being is not a genus (Metaph. 998b22, Anal Prior. 92b10).  We could put the point in formal mode by saying that 'being,' ens, das Seiende, are not sortal expressions.  (I am thinking of Heidegger's question, Warum ist das Seiende und nicht vielmehr nichts? )  But who ever said they were?

Maitzen's explanation of why people fall for the pseudo-question 'Why is there anything at all?' is because they confuse dummy sortals with genuine sortals.  But it seems to me that we can avoid the confusion and still sensibly ask the question.

Consider the question, 'Does anything exist?'  The question makes sense and has an obvious answer: 'Yes, things exist.'  Both the question and the answer make sense despite the presence in them of the dummy sortal 'thing.'  So why shouldn't it also make sense to ask why things exist?

Maitzen mistakenly assimilates the question 'Why does anything exist?' to the obviously senseless question 'How many things exist?'  This is the central weakness of his paper.  He never adequately explains the connection between the 'how many?' question and the 'why?' question.  The former is senseless and precisely for the reason that 'thing' is not a sortal.  But from the fact that 'thing' is not a sortal, how is it supposed to follow that the 'Why?' question is also senseless? 

Comments

[Vlastimil]

Bill,

Thanks for this post!

You agree with Maitzen that things can't be counted as things. If so, aren't general claims about the nature of reality typically made in different senses that are dependent on the usage of 'thing'?

Take physicalism. Physicalism boils down to the claim that any thing is numerically identical with a purely physical thing. (Whatever 'purely physical' means.) Which, by definition, boils down to the claim that for any thing X there's a natural thing Y such that X and Y are to be counted as 1 thing. But what 'thing' means here?

People speak of 'things' in a wild variety of senses. But general claims about the nature of reality (claims such as physicalism) are made with a rather specific sense of 'thing' in the speaker's mind -- an objective sense, a sense in which it is not up to anyone to determine what is and what isn't a thing, or how many things exist. Of course, even when two philosophers agree there is such an objective sense of 'thing', they often explicate it differently. For instance, an Aristotelian might mean by 'thing': a substance, an accident, an aggregate from these, or a substance's objectively carved part (extended or not). But others would balk at this scholastic baggage, and they would prefer some other explication of 'thing' as used in their sweeping metaphysical claims.

[Steve Maitzen]

Dr. Vallicella:

Thanks for blogging my article, and thanks for your comments. You're unconvinced that the senselessness of the "How many?" question implies the senselessness of the "Why any?" question. I tried to make the case on pp. 9-10 of the article, but clearly I didn't entirely succeed. Fortunately, I get to try again in the chapter I'm contributing to this anthology due out next year.

For now, though, recall the context in which I challenged the "Why any?" question. It's a context in which natural-scientific explanations have been given for the existence of pens, plums, penguins, and so on, and yet the supernaturalist objects that there's still an unanswered question left: "Why are there any concrete, contingent things at all?" Here the objector is making the same mistake made by someone who asks "Exactly how many concrete, contingent things are there?" -- namely, treating "concrete, contingent thing" as if it denoted a kind of thing distinct from pens, plums, penguins, and so on, a kind of thing whose "instances" must be explained. It doesn't, or else (as you say) we'd be able to count by it.

Absent the relevant context, I agree that it can look excessive for me to demand that we "stop asking why there's anything." In the relevant context, however, that's exactly what I mean to demand.

Cheers.
Steve Maitzen

[Robert Sloan Lee]

Dr. Vallicella,

Thanks for this post. I've long enjoyed your writing and work on this blog, and it's high time I said so. I think your focus on questions of intentionality is especially valuable.

Best Regards,

Dr. Lee

[Bill Vallicella]

Vlastimil,

So you think that there is an objective, definite answer to the question as to how many things there are? What's your answer? 27?

Suppose it is true that one cannot count things in the way one can count cats. It is still possible, I think, to refer in an unrestricted way to absolutely everything (of whatever category) and maintain that it is physical. Or do you think that existents must be countable as existents if one is to be able to maintain that everything is physical?

[Bill Vallicella]

Dr. Lee,

Thanks for the kind words. I'll be coming back to intentionality eventually. I have plenty of bones left to pick with London Ed . . .

It is supposed to get up to 112 F today. I don't reckon Dallas is much cooler.

[Bill Vallicella]

Dr. Maitzen,

Thanks very much for responding. I found your article extremely interesting in that it addresses an issue that I hadn't really thought about, namely, the connection between the 'How many?' and the 'Why any?' questions. As far as I know, no one else has addressed this issue in print. Or does Thomasson engage it?

You did a fine rigorous job with the 'How many?' question and I agree. It is the transition to the other question that is problematic, at least for me.

The pages are not numbered in the online version, but well into the article you said, to my surprise, "It is not that I regard counting as especially relevant to the answering, or even dissolving, the 'fundamnetal question.'" Then why did you go on for so long and in such detail about counting?

You should also not assume that interest in the Leibniz question (as a question unanswerable by natural science) is confined to theists and 'supernaturalists.' There are atheists who maintain that the question is genuine and answerable but without reference to God. You might want to take a look at my catalog of possible responses to the question: http://maverickphilosopher.typepad.com/maverick_philosopher/2012/05/catalog-of-possible-responses-to-why-is-there-anything-at-all.html

In my sketchy evaluation I did not show that the Leibniz question can be formulated in a way that avoids your criticism; but on the other hand, it seems to me that you did not show that one who poses the question must be confusing 'concrete contingent being' with a sortal. I of course agree that it is not a sortal.

Very stimulating paper! I'll be coming back to this in future posts.

[Vlastimil]

Bill,

I think there's a precise correct answer to the question, How many substances and accidents are there? But I don't know the answer.

If someone asks, How many things are there?, I need to know what he means by a 'thing'. Only then I can think about the question further.

I think that: (i) Everything is physical iff (ii) For any thing X, there's a purely physical thing Y such that X and Y are 1 thing. Now you concede that (i) is meaningful. Then, by the said equivalence, you're committed to the possibility of counting things. Granted, you are not committed to the possibility of counting instances of a dummy variable. You are committed to the possibility of counting instances of some specific ontological category (substance, accident, particular, universal, event, ...). It's up to you to explicate the category, and this way to specify what (ii) and (i) mean to you.

[Steve Maitzen]

Dr. Vallicella:

Thanks for your kind words. I'm not aware of anyone else who's offered a dummy-sortal dissolution of the "Why any?" question, so I thought it worth pursuing. I emailed Thomasson, who also hadn't seen it before. So I went with it.

In the paragraph from which you quoted, I may underestimate the relevance of the "How many?" question to the "Why any?" question, since the former is senseless for the same reason, I think, that the latter is (when read in the way usually intended).

I'll certainly study your catalog of possible responses to the question. True, my opponent isn't only the theist or the supernaturalist; it's anyone who thinks "Why are there any concrete, contingent things?" is a question over and above what natural science can in principle answer. I list both theistic and non-theistic exemplars in footnote 2. I'll make this point clearer in the text of the next version. Thanks again for your timely comments!

[Bill Vallicella]

Vlastimil writes, >>If someone asks, How many things are there?, I need to know what he means by a 'thing'.<<

That's exactly the point. One cannot count how many things there are until one gives content to 'things.' To use Maitzen's example, there is no definite answer to the question of how many things I have in my hand if I have in my hand only a capped pen. Do you count the cap and the pen separately? But if the question is how many pens, then the answer is clear: one pen.

This is related to Frege's doctrine that numerals are predicates of concepts, not objects.

>>(i) Everything is physical iff (ii) For any thing X, there's a purely physical thing Y such that X and Y are 1 thing.<<

There seems to be a confusion of the number 1 with the relation of identity. I'm afraid I don't follow your argument here.

[Bill Vallicella]

Steve (if I may),

The anti-theism of your article gives it current interest, but I think it blunts its real point, which is to show that the 'Why any?' question is ill-formed and for this reason unanswerable. Your response to the question is that of the rejectionist (by my typology). You reject the question as pseudo-question.

>>In the paragraph from which you quoted, I may underestimate the relevance of the "How many?" question to the "Why any?" question, since the former is senseless for the same reason, I think, that the latter is (when read in the way usually intended).<<

I think this puts it the wrong way around. If anything, you OVERestimate the relevance of the 'How many?' question to the 'Why any?' question.

I think you succeeded in showing that the former question is senseless. That is a fairly solid result. The claim that the latter is senseless is much harder to swallow. So I think you need to make it clearer how the senselessness of the former transfers to the latter.

That 'thing' is a dummy sortal makes it useless for counting; but why must we be able to count things if we are to be able to sensibly ask why there are any? After all, 'Do things exist?' make clear sense even though we cannot count how many things there are.

[Alfredo]

Dr. Vallicella,

How would we formulate the question "Do things exist?" formally? I am thinking about your objection to the thin conception of existence. You argue, compellingly in my opinion, that 'Something exists' is not equivalent to '(Ex)(x=x).' But then how can the question 'Do things exist?' be any easier to formulate? Maybe this would drive someone to think your question is meaningless, insofar as the answer to it cannot be formulated in predicate logic.

[Vlastimil]

Bill,

Not confusion but explication.

I understand that 'X is numerically identical with Y' as 'X is the same Z as Y.' Which I understand as 'X and Y is 1 Z.' In other words, no _numerical_ identity without counting, and no counting without a unit. I do not know what numerical identity period is.

[David Brightly]

Bill, Steve,

Perhaps we can connect the Why are? and How many? questions. The Why question isn't Is there anything and if so why? which gives us the chance to choose something and explain how it came about. Rather, it asks us to give a single explanation for everything. And if we can't answer the How many? question we can't say what everything amounts to.

[Steve Maitzen]

Bill,

On your taxonomy, I count as a "qualified" rejectionist. If someone asks, "Why is there anything at all?" I think an appropriate answer is "There had to be something, namely, every necessarily existing object, such as the empty set or the integers." In that case, the original question can't be a pseudo-question, contrary to (pure) rejectionism.

But in my experience such a questioner isn't asking about noncontingent or abstract things and hence isn't satisfied by an answer invoking them. Instead, the questioner wants to know why any of the concrete, contingent things endorsed by common sense exist: pens, penguins, plums, and so on. To that question, I'd give answers from natural science for each kind of thing in question. At this point, however, the questioner typically objects

(O) "You haven't thereby explained -- and couldn't -- why any concrete, contingent things exist at all: i.e., concrete, contingent things without further specification."

Claim O, I'm convinced, treats "concrete, contingent thing" as if it named a kind of thing, which you and I agree is a mistake.

Hence my qualified rejectionism.

[Bill Vallicella]

Alfredo,

Thanks for stopping by. Actually, it was you who some time ago got me thinking about the 'How many?' question and then in any e-mail you alluded to its being connected to the 'Why any?' question -- although you didn't explain the connection.

There is indeed a problem with formulating 'Things exist' in standard logic (first-order predicate logic with identity). We can't write 'For some x, x is a thing.' Equivalently, there is no concept whose instantiation is the existence of things in the way that there is a concept whose instantiation is the existence of cats.

If 'Do things exist?' and 'Things exist' cannot be expressed in standard logic, then I conclude: so much the worse for standard logic. I do not conclude that the question and its answer are senseless -- since they obviously are not.

Which is better known: that things exist, or that the Frege-Russell treatment of 'exist(s)' is correct? The former. Plus there is my battery of arguments against the Frege-Russell view.

What's not clear to me is whether Maitzen's position requires the Frege-Russell view. If it does, then his view is untenable.

[Bill Vallicella]

David,

The 'Why any?' question could be set up like this:

There are concrete beings. There might not have been any, i.e., they are contingent. Why then are there any such beings?

On this approach the demand is not for an explanation of everything.

>>And if we can't answer the How many? question we can't say what everything amounts to.<<

This is not evident to me. How many? demands an answer in terms of a definite cardinal, finite or transfinite. How many nat'l numbers? Aleph-nought. Now suppose there is no definite number that is the number of existents. Why can't we still sensibly use 'everything' to refer to them?

[Bill Vallicella]

Vlastimil,

So you subscribe to the sortal-relativity of identity?

But you don't think that 'thing' ('existent,' etc) are sortals do you?

[Bill Vallicella]

Steve,

I agree that the question is best put in terms of concrete contingent beings (CCBs). I think we also agree that there had to have been something: some necessarily existent abstracta. We also agree that there is no need for a separate explanation of a contingent abstractum such as the set consisting of the Moon.

You think that every CCB belongs to a kind or sort, and that the natural sciences can adequately explain why these sorts are instantiated. (I assume that you are some sort of naturalist about minds.) And having explained why all these sorts are instantiated, one has eo ipso explained why there are CCBs as opposed to no CCBs.

Question: what about the sorts themselves? Don't they exist? What explains their existence? Or are they necessary beings? But of course you don't believe in immutable animal species do you?

[Steve Maitzen]

Bill,

Good questions. Here are tentative answers. Yes, sorts exist, as abstract objects. What explains their existence? Their instances: the sort penguin exists because penguins exist, and likewise for pens, plums, and so on. But why are there any sorts at all? Here I'd respond as I did to "Why are there any CCBs at all?": by explaining individual sorts, such as penguin. For a sortalist, to explain the existence of penguins is to explain the existence of the sort penguin, assuming of course that "penguin" is a sortal. Someone who replies that any such explanation must in principle fall short seems to treat "sort" as a sortal, which I don't think it is.

[Account Deleted]

Dr. Maitzen,

I am unsure whether your tentative answer to Bill's question about the existence of sorts is viable. You maintain both that sorts are abstract objects and that their existence depends upon the existence of their instances which (in the cases under consideration) are typically contingent. The later thesis entails that if, for instance, penguins failed to exist at all, then the sortal *penguin* would not exist. But that contradicts the abstract character of this sortal which requires that it exist necessarily. Something must give, don't you think?

[James Chastek]

Steve and Bill,

What do you think of this: if things aren't sortal, then knowns aren't either. But we can meaningfully ask "Why are there any known things at all?" (and we know the answer, sc. because there are knowers) Therefore, we can ask absolute why questions about things that aren't sortal.

[Bill Vallicella]

Peter,

Maitzen could respond by saying that sorts are like Armstrong's immanent ('Aristotelian') universals: they exist but only if instantiated. An abstract object need not be a necessarily existent object: the singleton set consisting of you is abstract but as contingent as you are.

Still, there is something puzzling about Steve's view. If the sorts are as contingent as their instances, then there might not have been any sorts at all and the Why Any? question is back up and running: why are there any sorts at all?

[Account Deleted]

Bill,

Yes, you are right about my original point; I realized it shortly after posting and your example of sets came to mind. But in the case of sorts I find this Aristotalian view puzzling.

A penguin is a penguin because it instantiates the sortal *penguin*. Moreover, if the sortal *penguin* exists because penguins exist, then we have something like a vicious circle. For before penguins came into being, the sortal *penguin* did not yet exist. The later came into being just when at least one penguin did. But how can the first penguin which came into being be a penguin, rather than say a horse, if no sortal *penguin* even exists: i.e., how can something instantiate the sortal *penguin* if the later does not yet exist? If this *thing* that just came into being is not a penguin, since no sortal *penguin yet exists, then it is not a penguin. But then its existence cannot bring about the existence of the sortal *penguin*.

I am sure Dr. Maitzen has an answer to this problem.

[Steve Maitzen]

Peter: I think anyone who accepts that contingent things exist and who accepts standard set theory is committed to the existence of contingently existing abstracta, such as the set whose only member is you (to use Bill's example).

Bill: The two questions are parallel, and I handle them the same way. "There might not have been any sorts [of contingent beings] at all" gets the "Why any sorts?" question up and running only if "There might not have been any CCBs at all" gets the "Why any CCBs?" question up and running. As I've argued, someone who claims that natural science can't in principle answer each of those question is treating a dummy-sortal as if it were a genuine sortal.

[Steve Maitzen]

Peter: I reject the explanation that "A penguin is a penguin because it instantiates the sort[al] penguin." By "A penguin is a penguin," I take it you don't mean a tautology but instead the non-tautology "Some penguin is a penguin," i.e., "At least one penguin exists." I think the explanation has to go the other way: the sort penguin exists because at least one penguin exists. Why does at least one penguin exist? Here the explanation is biological, surely, rather than philosophical. The existence of penguins didn't await the emergence of the concept of penguin or the sortal term "penguin."

[Account Deleted]

Dr. Maitzen,

You say that "...the sort penguin exists because at least one penguin exists."

I am unsure how to interpret your 'because'? Is this a causal explanation? If so, then the coming into existence of at least one penguin causes the sortal *penguin* to come into existence. But this would require that the relevant sortal’s coming into existence is an event in time. And if it is an event in time, then there is a time when the first penguin came into existence, say t1, and at some time later, t2, the sortal *penguin* came into being. But, now, we face the problem of why is the entity that came into being at t1 (i.e., the first penguin) is a penguin and not something else; e.g., a horse, or just a mereological sum of some parts (a CCB, in Bill’s newly coined term)? This problem arises for you because according to your view prior to t1 there is no sortal *penguin*. Therefore, no entity that comes into being at t1 can be a penguin because it instantiates a sortal *penguin*, since no such sortal exists.

You might want to say that the first penguin and the sortal *penguin* came into being simultaneously. But then the *because* above cannot be causal. What is it then?

[Steve Maitzen]

Peter: My "because" was meant to be non-causal. Compare: "Why does the set {Mars} exist? Because Mars exists. But not conversely." I'm uncomfortable saying that the coming into existence of Mars causes {Mars} to come into existence, since the abstract object {Mars} isn't spatiotemporal. I'd be more comfortable saying that the existence of Mars is logically, or explanatorily, prior to the existence of {Mars} but bears no temporal relation to the existence of {Mars}. I concede, though, that this is tricky territory!

[Account Deleted]

Dr. Maitzen,

The example of Mars and {Mars} is somewhat different since 'Mars' is a name not a sortal. However, I understand the point illustrated by the example. The trouble is that I find it more difficult to see how explanatory or logical priority will work here.

First, explanatory priority is either intentional or extensional. If the later, then we are talking about causal explanation and we are back to my original objection. If it is the former, then we are talking about explanation within a theory, in which case it is more like an epistemic, or pragmatic, notion. But, then, I do not see how such epistemic or pragmatic matters can settle the question of the metaphysical priority of sortals versus their instances.

Second, I am not aware of any familiar logical system (e.g., first-order, second-order, etc.,) that would decide whether penguins or the sortal *penguin* has priority. So 'logical priority' cannot mean some ranking determined by a set of logical principles or rules that belong to a specific system of logic.

Alternatively, you might mean something broader by 'logical priority': i.e., either conceptual or metaphysical. While the question of metaphysical priority is at the heart of these questions, I thought that ultimately you would prefer to derive the conclusion of the metaphysical priority of instances (penguins) over their corresponding sortals from some more fundamental considerations. What could such considerations be? The only alternative I can think of is some conceptual priority. I wonder, then, how would you state such conceptual priority without being forced to use in the process some of the dummy sortals you deem meaningless?

[Steve Maitzen]

Peter: It's true that 'Mars' is a name rather than a sortal, such as 'penguin'. I'm not sure what difference that makes, if any, to the relation (logical, explanatory, whatever) between concrete particulars such as penguins and the sorts to which they belong, since the sort penguin and the set {Mars} are alike in being abstract objects. It seems to me that {Mars} depends (non-causally) on Mars but not conversely, since sets depend for their identity on their members but (I think) not conversely. Maybe I'm wrong: maybe there's no dependence, and no explanatory relation, at all holding between Mars and {Mars}. In that case, however, why accept the presupposition of the "problem" you identified several comments ago, i.e., that some such relation must hold between penguins and the sort penguin?