Letting 'CCB' abbreviate 'concrete contingent beings,' we may formulate the modified Leibniz question as follows: Why are there any CCBs at all? We have been discussing whether this question is a pseudo-question. To be precise, we have been discussing whether it is a pseudo-question on the assumption that it does not collapse into one or more naturalistically tractable questions: questions that can be answered by natural science.
My thesis is that the modified Leibniz question is a genuine question that does not collapse into one or more naturalistically tractable questions.
I
Consider a universe that consists of a beginningless actually infinite series of contingent beings. Let us assume that each CCB in this universe is (deterministically) caused by a preceding CCB. The beginninglessness of the series insures that every CCB has a cause. Since every CCB has a cause, each has a causal explanation in terms of an earlier one. And since each has a causal explanation, the whole lot of them does. (Some may smell the fallacy of composition in this last sentence, but let's assume arguendo that no fallacy has been committed.) Accordingly, the totality of CCBs, the universe, has an explanation in virtue of each CCB's having an explanation.
Some will say that on this scenario the modified Leibniz question has received a naturalistic answer. Why are there CCBs as as opposed to no CCBs? Because each CCB is causally explained by other CCBs, and because explaining each of them amounts to explaining the whole lot of them. And since the question has this naturalistic or universe-immanent answer, the specifically philiosophical form of the question, the question as Leibniz intended it, is a pseudo-question.
Others, like me, will insist that on the scenario sketched the question has not been answered. We will insist that a legitimate question remains: why is there this whole infinite system of contingent beings? After all, it is contingent, just as its parts are contingent whether taken distributively or collectively. There might not have been any concrete contingent beings at all, in which case there would not have been any CCBs to cause other CCBs. And nothing is changed by the fact that the series of CCBs is actually infinite in the past direction. The fact that the series always existed does not show that it could not have failed to exist. The temporal 'always' does not get the length of the modal 'necessarily.' If time is infinite in both directions, and the universe exists at every time, it does not follow that the universe necessarily exists. But if it contingently exists, then we are entitled to ask why it exists.
It is no answer to be told that each member of the universe, each CCB, is caused by others. I may cheerfully grant that but still sensibly ask: But what accounts for the whole causal system in the first place?
Please note that a possible answer here is: nothing does. The existence of the universe is a brute fact. Nothing I have said entails a theistic answer. My point is simply that the modified Leibniz question is a genuine question that cannot be answered by invoking causal relations within the universe.
II
There another line of attack open to me, one that focuses on the connection between causation and existence. It seems to me that the naturalist or 'immanentist' must assume that if x causes y, then x causes y to exist. The assumption, in other words, is that causation is existentially productive, that the cause brings the effect into existence. But on what theory of causation that the naturalist is likely to accept is causation productive?
This is a huge topic and I can only begin to explore it in this post. Suppose our naturalist, good empiricist that he is, subscribes to a Humean or regularity theory of causation along the following lines:
RT. x (directly) causes y =df (i) x and y are spatiotemporally contiguous; (ii) x occurs earlier than y; (iii) x and y are subsumed under event types X and Y that are related by the de facto empirical generalization that all events of type X are followed by events of type Y.
If this is what causation is, it is is not existentially productive: the cause does not produce, bring about, bring into existence the effect. On the contrary, the holding of the causal relation presupposes the existence of the cause-event and the effect-event. It follows that causation as understood on (RT) merely orders already existent events and cannot account for the very existence of these events.
Of course, the naturalist needn't be a Humean about causation. But then he ought to tell us what theory of causation he accepts and how it can be pressed into service to explain the very existence of CCBs.
For details and a much more rigorous development, see my article "The Hume-Edwards Objection to the Cosmological Argument," Journal of Philosophical Research, vol. XXII, 1997, pp. 425-443.
Hi Bill et al.,
It is somewhat amusing to me that the conservation has gone on this long, when in my mind the matter is so simple and straightforward. The medium of the internet makes communication rather difficult, however, so I cannot but expect that there might be some lapses in the conversation. I'll try to add something of value with this post, because there seems to be an aspect of the whole Leibnizian discussion that until now has gone unmentioned.
In my understanding, as I said in the comments section of the last post, the Leibnizian questioner supposes contingent existence, things existing contingently, to be a phenomenon in need of explaining.
Now it seems to me that the quaestio Leibniziana and Leibnizian reasoning in response to it takes place at the general level -- attempting to offer an explanation for contingent existence in general, whatever the particular explanations of individual concrete contingent beings turn out to be -- whereas Steve M has been trying to deal with the question only at the particular level, by offering what we might call scientific explanations for the existence of particular contingent existents. But this is wrongheaded, because what is primarily puzzling for the Leibnizian is contingent existence in general, not just some particular instance of contingent existence, and it is in general terms that the questioning and reasoning takes place.
Steve M's appeal to scientific explanations is abortive, as Bill has argued a number of times now, because it leaves the phenomenon explanandum entirely untouched: his explanations of some particular contingent existents contain reference to other contingent existents, and so the general problem of contingent existence hasn't been addressed. Considered at the general level, explaining contingent existence in terms of contingent existence explains nothing at all; this is why merely positing a natural explanation for the existence of some contingent beings doesn't address the problem put forth by the Leibnizian questioner.
Steve M cannot complain that there is no sense in addressing the problem from the general level because beings are only ever particular and ultimately explanation will have to make reference to particulars. This does not entail that the problem cannot be legitimately and safely addressed from the general level, even if this is true. It is true that, if beings are only ever particular and not general, then the actual explanation of the existence of all contingent beings will have to make reference to a particular existent, but that doesn't mean that the question and its possible solutions cannot be considered from a general level.
Imagine an analogous hypothetical example from the physical sciences, which I think should prove illuminating. Suppose we are all primitive scientists attempting to explain the phenomenon of burning. Bill and I, considering the issue from the general level, suggest that ultimately, what causes burning must not itself be on fire; otherwise burning will have gone unexplained. Steve M proposes, on the other hand, that one burning thing was caused to burn by some other burning thing, and that other burning thing was caused to burn by a third burning thing, and so on. Bill and I respond that this does not explain why things burn at all, and is not really an answer to the question -- it was burning in general that was to be explained (just as the Leibnizian says contingent existence in general is to be explained), not just the burning of some particular thing.
Can Steve M rightly complain that our objection to his proposal is incoherent or unintelligible? Or that somehow we've posited "burning things" as a kind of being, a genus to be found in the tree of Porphyry, which requires explanation? I can't see how these responses are anything else but confused. We have not, in our objection, proposed burning things as a kind which needs explanation, but simply used a general term to refer a universal phenomenon (whether existing qua universal only in our minds or whatever) that covers the particular instances we may or may not take into consideration.
So also, for the same reasons, when Steve M proposes contingent existents which explain the existence of other contingent existents, he cannot complain when Bill and I object that this doesn't explain contingent existence at all, which was the phenomenon to be explained from the beginning.
We can know by considering the problem at the general what will not work as an ultimate explanation at the particular level. Just as we can know that, if burning ultimately is to have an explanation, it cannot be in terms of something that is itself burning, so also, if contingent existence is to have an explanation, it cannot be in terms of something that itself exists contingently. Particular instances of burning may have been brought about by other instances of burning, just as particular contingent beings may have been brought about by other contingent beings, but simply positing more and more cases such as these, even an infinite number of them, will never explain the matter of burning/contingent existence.
This is something that seems rather obvious to me, so I don't know how to express myself in a significantly different or clearer manner than I already have. I think Bill and Peter should agree with what I've said.
Posted by: Steven | Saturday, June 09, 2012 at 12:34 PM
Steven,
Yes, I agree. If the question is why there are any concrete contingent beings at all, that question cannot be answered by invoking as explanantia contingent beings. Could Steve M really be denying this?
I sense that he is committing a straw man fallacy. Some things he says suggest that he thinks that we take 'CCB' to refer to instances of some special sort distinct from every ordinary sort, like cat. But nobody, as far as I know, ever made that mistake, and we certainly aren't.
'CCB' is a term that in the actual world refers to cats, dogs, penguins, etc. etc. and in other possible worlds to other sorts of contingent beings. In a world containing philosophizing cats and talking donkeys and flying horses it refers to, among other things, those critters.
Pointing to a cat, I say 'This is a cat.' Saying this, I specify WHAT the referent of 'this' is. Pointing to the same cat, I can also say 'This is a CCB.' Saying this, I say nothing about WHAT the referent of 'this' is; I say something about the mode of existence of the referent: it is concrete (causally active/passive) and contingent(possibly such as not to exist).
As Steven suggests, one cannot arrive at 'CCB' by climbing the tree of Porphyry.
'CCB'is not a mere covering term.
Posted by: Bill Vallicella | Saturday, June 09, 2012 at 06:04 PM
Trust me: it's frustrating for both sides. Part of the problem is that I'm accusing my opponents of an error that's subtle enough to have gone unnoticed until now. It shouldn't surprise me if they're not easily convinced!
Again, it puzzles me that Bill endorsed Steven's reasoning (a mistake Bill admits he's made before). The burning analogy is inapt, but it would take me too long explain why, so I'll focus on Steven's conclusion. He wrote, "when Steve M proposes contingent existents which explain the existence of other contingent existents, he cannot complain when Bill and I object that this doesn't explain contingent existence at all, which was the phenomenon to be explained from the beginning" (emphasis added).
What is it to "explain contingent existence"? (Notice the "ce" rather than "ts" at the end of that sentence.) Bill has said it's neither of these:
(Q1) Explain why CCBs are contingent, i.e., explain why CCBs are contingent rather than necessary.
(Q2) Explain why anything is contingent, i.e., explain why not everything is necessary.
Instead, Bill confirmed, it's answering this question:
(LQ) Why are there any CCBs?
I've said that LQ can be answered by empirically explaining why there are any pens, plums, penguins, and so on through every sort of CCB; admittedly this explanation will (I think it must) go on indefinitely. The issue between us is this: You guys say that this explanation, no matter how long it goes on, must in principle fail to answer LQ. Indeed, its alleged failure is the basis of the Leibnizian cosmological argument.
Now (to borrow from my article) suppose I mention pens, plums, and penguins, and you guys ask me why there are any of those things at all. I then give what you accept as an adequate empirical explanation for the existence of each of those three sorts. We all agree (don't we?) that you'd be perverse if you then demanded, "But why are there any of those things at all?" The phrase "those things" doesn't pick out a new kind whose existence deserves an explanation.
My question to you: How is it different if the list of items is indefinitely longer than 3? One bad reason for thinking it's different is the notion that A can't explain B unless A is self-explanatory. As I say in the article, that's not how our concept of explanation works.
Posted by: Steve Maitzen | Thursday, June 14, 2012 at 05:46 AM
Incidentally, Bill's "pointing" example, in his comment above, may in fact illustrate the problem. He wrote, "Pointing to the same cat, I can also say 'This is a CCB'." Sure, Bill can say "This is a CCB." But without further specification (either explicit or implicit) there's no fact of the matter about what he's pointing to. So when he writes, "I say something about the mode of existence of the referent," he's correct only if that further specification has occurred (explicitly or implicitly). If it hasn't, there's no semantically determinate referent of "This" about whose mode of existence Bill has said anything. (But I don't want this comment to distract anyone from answering the question I asked above!)
Posted by: Steve Maitzen | Thursday, June 14, 2012 at 06:55 AM
Dr. Maitzen,
1. I don't think it's different if the list of terms is indefinitely longer than 3.
2. But let's consider a slightly different scenario. Suppose I mention A's, B's, and C's, and you ask why there are any of those things at all. I then say, "The existence of C's is caused by that of B's, and that of B's by that of A's." Would you be perverse if you demanded, "But why are there any of those things at all?"?
Suppose "ABC's" is a mere covering term for A's, B's, and C's. I ask, "Why are there any ABC's at all?" If you say, "The existence of C's is caused by that of B's, and that of B's by that of A's," haven't you failed to answer the question? Ultimately you must invoke something other than A's, B's, or C's (something that's not an ABC) as an explanans ("The existence of A's is caused by D's), or you must say that ultimately the existence of ABC's is a brute fact, etc. (I'm guessing there are other possibilities).
My main question to you: How is it different if the list of items is indefinitely longer than 3?
Let's suppose that "CCB's" is a mere covering term for pens, plums, penguins, and so on. I ask, "Why are there any CCB's at all?" The imaginary naturalist in your article wants to give answers like this one: Well "penguins exist, and they're contingent. They evolved from--." But "--" are themselves CCB's, so with infinitely many answers like this one, wouldn't the question remain unanswered, even if "CCB's" is just a covering term for pens, plums, penguins, and, of course, "--"?
I hope my questions makes sense.
3. Also, you haven't responded to Dr. Vallicella's point about causation. You say that the existence of penguins, for example, can be empirically explained. How so? I'm guessing the proposed answer will involve some causal relationships.
Anyway, thanks for the intriguing article that started this fascinating discussion! I'm not a philosopher so forgive me if my comments are useless. Just interested in learning from you professionals.
Posted by: BA | Thursday, June 14, 2012 at 11:23 PM
BA,
I'm pleased you found the article intriguing. Bill deserves the credit for turning it into a fascinating discussion, one I've found helpful (if sometimes frustrating). To reply:
1. Glad to hear it!
2a. Your scenario differs from mine in that you haven't explained why there are any A's. In my scenario, each of pens, plums, and penguins is (ex hypothesi) sufficiently explained. A sensible response to your scenario isn't "Why are there any of those things at all?", because you've explained why there are B's and C's; instead, it's "Why are there any As?" That's certainly a fair question from someone who wants to know why there are any A's, B's, and C's. However, if you then explain why there are A's, I've got what I sought: an explanation of why there are any A's, B's, and C's. It doesn't matter if your explanation of A's invokes something else -- D's -- that in turn can be explained by E's. Our concept of explanation doesn't require that the explanans (the "explainer") be self-explanatory. Indeed, I'm not sure it even requires that the explanans must have an explanation, but in any case our concept allows that A can explain B even if something else altogether explains A.
2b. You ask, "[S]o with infinitely many answers like this one, wouldn't the question remain unanswered, even if 'CCB's' is just a covering term for pens, plums, penguins, and, of course, '--'?" Answer: No, not unless the question "Why are there any of those things?" remains unanswered in my original scenario involving pens, plums, and penguins. The fact that I explain penguins by reference to something else that needs explaining doesn't mean I don't explain penguins.
Importantly, my opponents say that my explanation can't hope to answer LQ even if every CCB is noncircularly explainable in terms of other CCBs: not because my explanation goes on forever -- after all, maybe there's no first CCB -- but because (to quote them) "It doesn't explain why there are any CCBs at all." In using the quoted language, they're treating 'CCB' as if it denoted a kind of thing whose existence needs explaining, which Bill concedes it doesn't.
3. I didn't respond publicly to Bill's point about causation, because I think talk of causation obscures matters rather than clarifying them. Here's what I wrote him by email regarding his specific claim about causation: "[N]othing in my critique of the Leibnizian question depends on any claims about causation. In my article, I never use the words 'cause' or 'causation' in propria persona. Instead, I talk about explanation, a pragmatic rather than an ontological notion. (Actually, I think causation is also a pragmatic notion.) If there are ontologically objective causes, then they're
instantaneous states of the whole universe. Since spacetime is continuous, no 'cause' is ever contiguous with its 'effect', just as no two points in the spacetime continuum are adjacent."
To be fair, there are questions of mine to which Bill never replied, such as my question about the reference of 'superhero'.
Posted by: Steve Maitzen | Friday, June 15, 2012 at 11:09 AM
Steve M,
I promise not to ignore your first comment. But let me start with the second one. 'This is a CCB' is short for 'This cat is a CCB.' That's semantically determinate. Suppose there is only one cat in the vicinity.
I take it you agree that I have said something about the mode of existence of the referent: that it is concrete, and that it is contingent.
Now I am assuming that one can predicate existence of an individual such as a cat. Suppose the cat's name is 'Max.' I am assuming that 'Max exists' as it stands is meaningful, something denied by Frege and Russell. Russell famously said that 'Socrates exists' as a first-level predication of existence) is as meaningless as 'Socrates is numerous.'
One question I have is whether our disagreement about the LQ is at bottom a disagreement about the nature of existence.
Posted by: Bill Vallicella | Friday, June 15, 2012 at 11:13 AM
Steve M writes, >>(LQ) Why are there any CCBs?
I've said that LQ can be answered by empirically explaining why there are any pens, plums, penguins, and so on through every sort of CCB; admittedly this explanation will (I think it must) go on indefinitely. The issue between us is this: You guys say that this explanation, no matter how long it goes on, must in principle fail to answer LQ. Indeed, its alleged failure is the basis of the Leibnizian cosmological argument.<<
I think we should leave the cosmological argument out of the discussion because I am willing, for the purposes of this discussion, to countenance the 'brute fact' response: CCBs just exist without case, reason, explanation.
It seems to me that any explanation of the existence of CCBs must make reference to other CCBs. And to explain these other CCBs one must make reference to still other ones, and so on. But there might not have been any CCBs at all. If you grant that possibility, then it seems evident to me that you cannot explain why there are any CCBs at all (as opposed to no CCBs at all) by explaining some in terms of others -- even if there is an actually infinite regress of explanations.
I am beginning to wonder whether our dispute at bottom concerns the nature of explanation.
Why are there pens? Because there are geese, and there are humans, and some humans pulled some quills out of some geese, and used them to write with. If the only pens are goose quill pens, then that explains why there are goose quill pens. I grant that one could stop right there, whtout explaining why there are geese and why there are humans just as, if I want to know why the crops failed it suffices to say that there was a drought: I don't have to explain why there was a drought to have an explanation of why the crops failed.
But this won't work if I want an explanation of why there is anything concrete and contingent at all -- for the reason given three paragraphs back.
Posted by: Bill Vallicella | Friday, June 15, 2012 at 12:10 PM
Steve: Did King Arthur actually exist or is he a fictional being? Not knowing the answer to that, I did not know what you were getting at in your earier comment.
You don't want to talk about causation. But isn't the only sort of explanation that is relevant to this discussion causal explanation? To understand causal explanation, however, we need to to understand causation, and how causation and existence are related.
On a regularity theory of causation, causation is just regular succession: x causes y iff the x-y sequence instantiates a regularity. But on such a theory, if x causes y, then it is not the case that x causes y to exist.
You need a more robust theory of causation. So it seems appropriate for me to ask you what your theory of causation is.
Posted by: Bill Vallicella | Friday, June 15, 2012 at 12:55 PM
Bill,
1. I don't think our dispute arises because of a disagreement about the nature of existence or about the nature of explanation. I agree that your assertion "This is a CCB" makes sense, and can be true, provided that an answer to "This what?" -- in terms of a sortal -- is already supplied at least implicity by your intentions, or by the context, or by a combination of both. Nor do I have a problem with "Max exists" (which I'd render logically as "There exists an x such that x = Max"); it's meaningful and true.
2. I grant that there might have been no CCBs at all, even though that doesn't follow from the fact that each CCB is contingent. But your replies increase my confidence that you're using 'CCB' as if it denoted a genuine kind (or sort), something you agree it doesn't denote. I'm convinced that the error I've alleged occurs in this latest claim of yours: "If you grant that possibility, then it seems evident to me that you cannot explain why there are any CCBs at all (as opposed to no CCBs at all) by explaining some in terms of others -- even if there is an actually infinite regress of explanations." You've said that 'CCB' isn't a covering term or a sortal. What is it, then?
3. Historians dispute whether King Arthur existed, just as ontologists dispute whether arbitrary undetached parts exist. So the examples are parallel. If we agree that (as seems plausible) the extension of 'superhero' is semantically indeterminate in that it neither includes nor excludes King Arthur, it would be bizarre for either of us to say "King Arthur was a superhero if he existed." Likewise, because you agree the extension of 'CCB' is semantically indeterminate in that it neither includes nor excludes arbitrary undetached parts, it's bizarre for you to write "Tibbles-minus-his tail, which is an arbitrary undetached part of Tibbles, is a CCB if there is such an item as Tibbles-minus-his tail." You seem thereby to deny the semantic indeterminacy of 'CCB' in favor of saying that the extension of 'CCB' does include arbitrary undetached parts if any exist.
4. We don't need a "theory of causation" in order to agree that A explains B. You wrote, "[I]f I want to know why the crops failed it suffices to say that there was a drought." The drought explains the crop-failure -- if you like, the drought "caused" the crop failure, if "caused" means something pragmatic, like "explains." Surely someone can know that the drought caused the crop-failure (in the sense of "explains it") without having a theory of causation. Good thing, too, since the metaphysics of causation is a mess.
Posted by: Steve Maitzen | Friday, June 15, 2012 at 01:40 PM
Bill,
Maybe this will convince you? The general principle you seem to be assuming is this:
(GP) You can't explain why there are any A's at all by invoking only items of kind (or sort) A, even if your explanation goes on forever.
GP is a correct principle, as far as I can see. GP implies that you can't explain why there are any penguins at all just by talking about more penguins, even if you talk forever. Right! But then you seem to move, on the basis of GP, to this conclusion: You can't explain why there are any CCBs at all by invoking only CCBs. If that's your reasoning, then we both know why it's mistaken: 'CCB' doesn't denote a kind (or sort), so you can't use GP to deduce your conclusion about CCBs.
Yes?
Posted by: Steve Maitzen | Friday, June 15, 2012 at 02:16 PM
Steve:
I hope to respond to your latest comments later, but for now, I have a little argument for you.
There are possible worlds in which there are no CCBs. Our world, however, is a world of CCBs. Let 'A' rigidly designate our world. A is actual, but might not have been. Why is A actual rather than some world in which there are no CCBs?
It seems evident to me that the explanation why A is actual cannot be in terms of anything going on in A.
Why not? Well, consider world B which is like ours in that it contains CCBs, but different sorts of CCBs than A. And in B, the instantiation of every sort has an explanation that satisfies your naturalist scruples. B, however, is merely possible. (I am assuming, of course, that only one world can be actual, and thus that actuality is absolute, not world-relative.) Since the fact that the instantiation of every sort in B has an explanation is consistent with B's being merely possible, the same goes for A.
Therefore, the explanation why A is actual cannot be in terms of anything going on in A, or the holding of explanations in A.
Therefore, even after you have explained every sort in A, you can still legitimately ask: Why are there any CCBs at all? For this is equivalent to the question why A is actual.
Posted by: Bill Vallicella | Friday, June 15, 2012 at 04:22 PM
Bill: I address something like this issue in the last paragraph of p. 58 of my article. First, if 'A' rigidly designates the actual world (as you want it to), then it's metaphysically necessary that A contains exactly the CCBs it contains. Second, as I think you suspect, the question "Why is A actual?" (where 'A' designates rigidly) boils down to the question "Why are the particular CCBs in A actual?" [really, it boils down to "Why are the particular contingent truths of A actually true?"]. "Why are the particular CCBs in A actual?" admits of a naturalistic answer just as much as does (LQA) "Why are any CCBs actual?"
I agree that "the explanation why A is actual cannot be in terms of...the [mere] holding of explanations in A." I disagree, for reasons just given, that "the explanation why A is actual cannot be in terms of anything going on in A."
I'll await your reaction to GP.
Posted by: Steve Maitzen | Friday, June 15, 2012 at 05:07 PM
Dr. Maitzen,
Thanks for the response! I don't want to distract you from your more sophisticated discussion with Dr. Vallicella, but I do have a a question.
Do you agree with this paragraph from above (with minor changes for clarity)?
Suppose "ABC's" is a mere covering term for A's, B's, and C's. I ask, "Why are there any ABC's at all?" If you say, "The existence of C's is caused by that of B's, and that of B's by that of A's," then you have failed to answer the question. Ultimately you must invoke something other than A's, B's, or C's--something that's not an ABC--as an explanans ("The existence of A's is caused by D's," for example); or you must say that ultimately the existence of ABC's is a brute fact.
Posted by: BA | Friday, June 15, 2012 at 09:59 PM
BA: Yes, I agree. But keep in mind in 2b of my earlier reply. More important, keep in mind the limited scope of the correct principle I labeled "GP" above, which I think gets to the heart of the matter.
Posted by: Steve Maitzen | Saturday, June 16, 2012 at 05:56 AM
Scenario #1
Questioner: Why are there X's, where "X's" is a covering term for A's, B's, and C's?
Answerer: The existence of C's is explained by B's, and that of B's by A's.
Dr. Maitzen, we agree that 1) the answerer has explained the existence of C's and B's, and 2) he has failed to answer the question. And we agree that in order to answer the question, he must ultimately invoke an item (such as D's) that is not an X.
Increasing the number of items that are covered by the covering term "X's" wouldn't make a difference. Ultimately, the answerer must invoke at least one non-X as an explanans, or he must say that the existence of X's is a brute fact.
Scenario #2
Questioner: Why are there X's, where "X's" is a covering term for A's, B's, and C's?
Answerer: The existence of A's is caused by D's, that of B's by E's, and that of C's by F's.
We agree that the question has been answered, even if D's, E's, and F's are not self-explanatory. The question has been answered because each of the items (each of the sorts) covered by the covering term "X's" has been explained.
Again, increasing the number of items covered by "X's" wouldn't make a difference. If each item is explained, the question is answered. Notice, though, that the question is answered by invoking items (D's, E's, and F's) that aren't X's. Scenario #2 is not a counterexample to the idea that the question can only be answered by invoking at least one non-X at some point in the explanation.
Scenario #3 (your scenario, as I understand it)
Questioner: Why are there X's, where "X's" is a covering term for penguins, plums, and pens?
Answerer: [gives an empirical explanation for the existence of penguins, then for that of plums, and then for that of pens.]
Scenario #3 fits the "form" of Scenario #2. The question is answered in both. But notice again that the question is answered (I hope) by invoking items (geese, humans, etc.) that aren't X's.
Scenario #4
Questioner: Why are there X's, where "X's" is a covering term for penguins, plums, pens, people, and so on?
Answerer: [begins to explain each X he knows of in terms of other X's, a project to be continued indefinitely by future generations of answerers.]
You say that answerer #4 fails only if answerers #2 and 3 fail. I say that answerer #4 succeeds only if answerer #1 does. If answerer #1 (and, of course, answerers #2 and 3) must invoke at least one non-X, then why mustn't answerer #4?
I know you asked me to keep in mind the limited scope of GP, but I don't see why it doesn't apply in scenario #4, in which "X's" just happens to be a covering term for a much larger number of items. If GP applies when explaining the existence of three sorts of things, why not when explaining the existence of many, many more sorts?
Posted by: BA | Saturday, June 16, 2012 at 11:39 PM
If GP applies when explaining the existence of three sorts of things, why not when explaining the existence of many, many more sorts?
GP does apply, no matter how many sorts are involved. By "the limited scope of GP," I meant the fact that GP applies only to sorts (or kinds) of things, as I mentioned above. Bill agrees that 'CCB' doesn't denote a sort (or kind) in the first place, and so GP, although true, doesn't preclude an answer to "Why are there any CCBs at all?" that invokes only CCBs. Principles stronger than GP (principles that aren't restricted to sorts) fall prey to clear counterexamples.
Posted by: Steve Maitzen | Monday, June 18, 2012 at 04:48 AM
Thanks again for your time, Dr. Maitzen!
I think the question you answered above was a bad one on my part. I know the discussion has moved away from here, but I'd like to try one last time to be clear, even if I have to be wordy again.
Do you agree that the success of Answerers #2 and 3 doesn't entail the success of Answerer #4? That's been my main claim all along.
In your first comment, you suggested that if you can give a naturalistic, genuine answer to Question #3
(Why are there X's or "those things," where "X's" or "those things" is a mere covering term for penguins, plums, and pens?)
then you can give a naturalistic, genuine answer to Question #4
(Why are there CCB's, where "CCB's" is a mere covering term for penguins, plums, pens, and so on?)
because the only important difference between the questions is the number of items for which an explanation is demanded. In the latter question the list of items is "indefinitely longer than 3," as you said; but each question is about a list of items, a short list and an extremely long list.
Now, I'm assuming, with you (right?), that "CCB's" is a mere covering term. I'm also assuming (also with you?) that the only difference between the "X's" of Question #3 and the "CCB's" of Question #4 is that the latter is a covering term for a much longer list of items than 3. If you explain every item, then you answer the question. If the questioner isn't satisfied with the answer, that's his fault: to explain the existence of X's is just to explain the existence of penguins, the existence of plums, and the existence of pens; likewise, to explain the existence of CCB's is just to explain the existence of penguins, the existence plums, the existence of pens, and so on through all the different sorts there are.
But if that's the only difference between CCB's and the X's of Question #3, then it's also the only difference between CCB's and the X's of Question #1. And just as Answerer #1 must invoke a non-X (as must Answerers #2 and 3, for that matter), so too must Answerer #4 invoke a non-CCB (where "CCB's" is a covering term like "X's" and "those things" in your original scenario). There's just no way he's going to explain every item covered by the covering term without invoking an item that's not covered. (And your original scenario (Scenario #3) doesn't show why he would be able to, because Answerer #3 does invoke items that aren't covered by "X's" or "those things.") The explanation would be circular, or the list of items covered by the covering term would have to be infinitely long in the "backwards" direction: there couldn't be a time t such that at t there were no X's or CCB's or "those things."
If there are no non-CCB's, then the question can't be answered. All we can say in that case is that the existence of CCB's is a brute fact. Likewise if there are no non-CCB's that can explain the existence of CCB's.
If there is no first CCB (where "CCB's" is a mere covering term like "those things" in your scenario), then naturalism might be able to answer the question--but only if there is no first CCB and if "CCB's" is a mere covering term. Even so, I think an explanation that invoked non-CCB's would somehow be a better, more intellectually satisfying explanation.
What do you think? Am I making a mistake somewhere?
Posted by: BA | Tuesday, June 19, 2012 at 12:21 AM
BA: I don't have the time it would take to answer your 600-word comment point by point, so let me cut to what I think is the chase.
You wrote, "The explanation would be circular, or the list of items covered by the covering term would have to be infinitely long in the 'backwards' direction: there couldn't be a time t such that at t there were no X's or CCB's or 'those things'."
Bill and I have been debating whether the existence of CCBs can be adequately explained by an explanation that invokes only CCBs. Bill says it can't be, even if the explanation goes on forever. I've asked him "Why not?" more than once, and so far he hasn't said why not, but he hasn't challenged the adequacy, in principle, of explanations that go on forever.
I think that (noncircular) explanations entailing the existence of CCBs must go on forever, or else we arrive at either (1) a brute contingent fact or (2) a self-explanatory (hence necessary) fact entailing the existence of CCBs, which fact therefore makes the existence of CCBs necessary rather than contingent. I don't accept (1) or (2), and I take it neither does Bill.
You also wrote, "Even so, I think an explanation that invoked non-CCB's would somehow be a better, more intellectually satisfying explanation." Explanations invoking non-CCBs invite worry (2); how would you steer clear of that?
Posted by: Steve Maitzen | Tuesday, June 19, 2012 at 08:57 AM