Greene on Nozick on nothing

Brian Greene’s The Hidden Reality surveys the various speculations about parallel universes on offer in contemporary physics.  Toward the end of the book, Greene discusses a proposal put forward by Robert Nozick in chapter 2 of his book Philosophical Explanations.  (Turns out that Greene took a course with Nozick at the time Nozick was writing the book.)  Greene notes that even if any of the multiverse theories currently discussed by physicists -- those inspired by quantum mechanics, string theory, inflationary cosmology, or what have you -- turned out to be correct, one could always ask why the world is as the theory describes it, rather than some other way.  (This is one reason why it is no good to appeal to such theories as a way of blocking arguments for God as an Uncaused Cause of the world.  We had occasion recently to note some other problems with this atheist strategy.)  But Nozick put forward a version that Greene regards as not subject to this question -- what Greene calls the Ultimate Multiverse theory.

On the Ultimate Multiverse theory, all possible universes exist, including a universe consisting of nothing.  To the questions “Why does this universe exist rather than some other?” and “Why does any universe at all exist rather than nothing?”, the Ultimate Multiverse theory responds: “There is no ‘rather than’ about it.  This universe and every other possible universe all exist; indeed, this universe and a universe consisting of nothing both exist.  So there is no special explanation of our universe required, because it isn’t in the first place only one of many possibilities to have been actualized.”  (It should be added that neither Nozick nor Greene actually endorses this theory; they merely float it as a possibility.  David Lewis famously did defend a similar view, though.)

Now, the proposal that every possible universe exists does not, by itself, actually explain anything.  In fact -- again, at least by itself -- it makes things more mysterious rather than less.  Suppose I ask “Why is there a cup on the table?”  It is no good to answer “Actually, there are in fact two cups on the table; hence there is no special reason to ask where the one cup came from!”  This hardly defuses the original question; indeed, there is now more to explain than there was originally.  And the problem would rather obviously only be made worse if it turned out that ten or twenty cups were on the table, and certainly if every possible cup were on the table.  Similarly, that every possible universe exists hardly explains why anything exists at all; it just adds to the explanandum rather than providing an explanans.

But Nozick’s Ultimate Multiverse theory involves more than merely the suggestion that every possible world exists.  Nozick tells us that this “fecundity assumption,” as he calls it (and which he compares to the traditional “principle of plenitude,” which I had reason to discuss in an aesthetic context), follows from a metaphysical “egalitarianism.”  But what does he mean by “egalitarianism,” and why does he think it defensible?  The answers to these questions constitute the heart of Nozicks’ treatment of the question “Why is there something rather than nothing?”  And they reveal, I think, that Nozick misses the point both of the question and of the traditional theistic answer to the question, at least as these are understood within classical (Platonic, Aristotelian, and Scholastic) philosophy.

Recall that in the Platonic/Neo-Platonic tradition, whatever is in any way composite must be explained by reference to what is absolutely simple or non-composite; that in the Thomistic tradition, whatever has an essence distinct from its act of existence must be explained by reference to something whose essence just is subsistent existence; and that these points ultimately reflect the Aristotelian principle that whatever contains potentiality of any sort must ultimately be explained by reference to that which is pure actuality, devoid of potentiality (since that which is a composite of an essence and an act of existence, or indeed composite in any way, is merely potential until the composition of its parts into a whole is actualized by something else).  These points are, of course, very abstract, and as always I presuppose that the reader has some familiarity with the basic concepts given that I’ve spelled them out in detail elsewhere.  (The Aristotelian and Thomistic ideas in question are developed at length in The Last Superstition, Aquinas, and my article “Existential Inertia and the Five Ways” and, more briefly, here and there in various blog posts such as this one or this one; and I had reason to discuss the Neo-Platonic conception of divine simplicity in another earlier post.)  The point for now is to emphasize that it is because the things in the world of our experience are composite, because there is in them a distinction between essence and existence, and because they are mixtures of actuality and potentiality, that they necessarily require a cause.  And it is because God is none of these things -- where what we mean by “God” is that which is none of these things -- that He does not and cannot in principle require a cause.  Precisely because he can actualize without having to be actualized, precisely because He is being or existence itself rather than something which merely participates in existence, and precisely because He is absolutely simple and not in need of composition of any sort, He and He alone can serve as the ultimate terminus of explanation.

Now at the beginning of his discussion, Nozick tells us, plausibly enough, that the alternative to an infinite regress of explanations is that there is some truth to which no further truth stands as an explanation.  But he regards the latter possibility as entailing that there is some truth or truths “without any explanation.”  And this claim, in turn, is one he says can be interpreted in either of two ways:

About such truths p lacking further explanation, there also appear to be two possibilities.  First, that such truths are necessarily true, and could not have been otherwise.  (Aristotle, as standardly interpreted, maintained this.)  But it is difficult to see how this would be true.  It is not enough merely for it to be of the essence of the things which exist (and so necessarily true of them) that p.  There would remain the question of why those and only those sorts of things (subject to p) exist; only if p must be true of everything possible would this question be avoided.

The second possibility is that p is a brute fact.  It just happens that things are that way.  There is no explanation (or reason) why they are that way rather than another way, no (hint of) necessity to remove the arbitrariness. (p. 117)

Here, it seems to me, Nozick’s account already goes awry in several respects.  For one thing, the notion of what is “necessary” needs to be more carefully unpacked than he (and many other contemporary philosophers) unpack it.  For instance, there is in Aristotelian-Thomistic (A-T) metaphysics a sense in which a necessary being might derive its necessity from some other necessary being.  That a derivatively necessary being exists (if it does exist) is thus a truth which will have a “further explanation,” in terms of a further necessary being from which the one in question derives its own necessity.  (See my discussion of the Third Way in Aquinas for more details.)  Moreover, even with respect to an absolutely necessary being -- one which does not in any way derive its necessity from another, but has it in itself -- it will not be correct to say that its existence is “without any explanation.”  Rather, its existence is self-explanatory.  For that which is absolutely necessary is (for A-T) absolutely necessary precisely because it is pure actuality or subsistent being itself.  Hence it doesn’t “have” or merely “participate in” being or existence in the way contingent and derivatively necessary things do, and it doesn’t have a potential for contingent or derivatively necessary existence which needs in some way to be made actual.  Again, it “already” is being or existence itself; it “already” is pure actuality.  That does not make its existence less intelligible than that of other things, but more intelligible.  Contingent and derivatively necessary things are contingent or derivatively necessary precisely because their existence is merely participated existence, and is in one way or another merely potential until actualized.  Their explanation must accordingly lie in something outside them.  The being or actuality of an absolutely necessary being, by contrast, is like the goodness of Plato’s Form of the Good -- it is intrinsic, intelligible in itself and the source and standard of the intelligibility of all other things.

(Note that the notion of being self-explanatory is not to be confused with the notion of being self-caused, which is incoherent.  Causation is a metaphysical notion, having to do with the source from which a thing derives some aspect of its being.  But explanation is a logical notion, having to do with the way in which we understand or make sense of some aspect of a thing’s being.  We cannot coherently say that a thing derives its existence from itself or its nature, for that would entail, absurdly, that the thing or its nature exists prior to itself, in an ontological sense even if not a temporal sense.  But we can coherently say that a thing’s existence can be made sense of in terms of its nature, for that has to do, not with where a thing “gets” its existence from -- an absolutely necessary being doesn’t get it from anywhere -- but rather with how we can make intelligible or understand its existence.)

This brings us to the second part of the passage from Nozick just quoted, wherein (to repeat) he says:

It is not enough merely for it to be of the essence of the things [lacking further explanation] which exist (and so necessarily true of them) that p.  There would remain the question of why those and only those sorts of things (subject to p) exist…

Nozick seems to mean by this that we would have to ask, with respect to any purportedly necessary terminus of explanation, whence it derived its necessity and why only the thing or things that constitute the terminus have derived it, and that we would have to ask this even if this necessity were “of the essence” of these things -- as if the necessity of the terminus of all explanation were merely a “participated” or “instantiated” necessity, in the way that the four-leggedness of a dog, though “of the essence” of being a dog, is still a “participated” or “instantiated” four-leggedness.  But this misses the whole point of the idea of God as an absolutely necessary being, at least as that is understood in classical metaphysics (i.e. in terms of pure actuality, subsistent being itself, and so forth).

Nozick may have been misled here by the way modern philosophers of religion often speak of God, under the influence of what Brian Davies has called a “theistic personalist” conception of God that is very different from the classical theism of Augustine, Anselm, Aquinas, and other ancient and medieval thinkers.  (I have addressed the difference between classical theism and theistic personalism in a number of posts, such as this one, this one, and this one.)  Proponents of Leibnizian cosmological arguments and Cartesian ontological arguments often say that God’s essence “includes” existence -- as if existence (or perhaps “necessary existence”) was merely one among a number of “great-making properties” that God “instantiates.”  But that is not at all what Aquinas (and, I would say, Anselm) are saying.  God doesn’t “instantiate” properties.  That would make of God merely “a being” among other beings, and the God of classical theism is not that.  Aquinas, Anselm, and other classical theists are saying something far more radical.  For them (and to repeat) God does not “have” existence but is existence itself; and He doesn’t “instantiate” or “participate in” anything, but is rather that in which everything else participates.  

Of course, one might object on various grounds to classical theism and its classical (Platonic, Aristotelian, and Scholastic) metaphysical underpinnings.  The point is just that Nozick’s discussion of the question of why there is something rather than nothing simply fails even to take account of this entire classical tradition -- no small lacuna given that it is, historically, the mainstream approach to the question.  And it is his failure to take account of it that leads him quickly to jump to the conclusion that any answer to the question is to some extent going to have to appeal to an inexplicable “brute fact” -- a position the classical tradition vehemently rejects.  One can intelligibly deny that the God of classical theism exists.  But one cannot intelligibly say that even if He did exist, His existence would be a “brute fact,” or that “there would remain the question of why that and only that sort of thing” does not require further explanation.   As I have noted before, suggestions of this kind completely miss the point of classical theism.

This brings us to Nozick’s metaphysical “egalitarianism,” which undergirds his Ultimate Multiverse scenario and which is best understood in contrast with what he calls “inegalitarian” views:

An inegalitarian theory partitions states into two classes: those requiring explanation, and those neither needing nor admitting of explanation.  Inegalitarian theories are especially well geared to answer questions of the form “why is there X rather than Y?”  There is a non-N state rather than an N state because of the forces F that acted to bring the system away from N.  When there is an N state, this is because there were no unbalanced forces acting to bring the system away from N.

Inegalitarian theories unavoidably leave two questions unanswered.  First, why is it N that is the natural state which occurs in the absence of unbalanced external forces, rather than some other (type of) state N’?  Second, given that N is a natural or privileged state, why is it forces of type F, not of some other type F’, that produce deviations from N?  If our fundamental theory has an inegalitarian structure, it will leave as brute and unexplained the fact that N rather than something else is a natural state, and that F rather than something else is the deviation force. (p. 121)

Perhaps it is obvious from the foregoing what is wrong with all this, at least if intended as an analysis of classical metaphysical approaches to the question of why anything exists rather than nothing.  As the classical tradition understands it, the N from which there are “deviations” would be (say) a potency or potential, such as the potential redness of skin (before it has become sunburned), the potential squishiness of an ice cream cone (before it has been left out to melt), or indeed the potential existence of a universe.  And there is nothing “brute,” “unexplained,” or “unanswered” about why these or any other potentials need some “external factor” or “force” to bring about a “deviation from N” -- being merely potential rather than actual, the Ns in question quite obviously cannot do anything at all.  In particular, potential redness, potential squishiness, potential universes, etc. cannot actualize themselves.  The relevant F which does actualize them is something which is itself already actual (such as the sun in the case of the sunburn and the melted ice cream cone), and there is nothing remotely “brute” or “unexplained” about why F has to be something already actual -- nothing non-actual could be F, because being non-actual, it (obviously) couldn’t do anything at all.  

Nozick makes the whole issue sound more mysterious than it really is precisely because of his use of formalisms of the sort fetishized by analytic philosophers, and he thereby provides a good illustration of how this method can generate obfuscation rather than rigor.  For the formalisms simply ignore the actual content of the principles classical writers appeal to when addressing questions of ultimate explanation, and thereby miss the entire point.  When, for example, a Scholastic writer says that no potential can actualize itself, but has to be actualized by something already actual, he is not postulating some mysterious “force of type F” whose power to “produce deviations from N” is left “brute” and “unexplained.”  For he is not talking about “a” force among other forces in the first place, not even a special kind of force; he is rather making the extremely obvious point -- indeed, one almost wants to say the trivial point, except that some decidedly non-trivial consequences follow from it -- that only what is actual can serve as any kind of force at all.  And neither is the Scholastic postulating some mysterious “natural state N which occurs in the absence of unbalanced external forces.”  He is rather making another obvious point, viz. that what is merely potential -- whether it is “natural” or not is not relevant -- cannot do anything, precisely because it is merely potential and not actual.

Thus, when Nozick asks “Why is it forces of type F, not of some other type F’, that produce deviations from N?” he is not asking a profound question, but a very silly question, at least if he intends to raise problems for an account like that of Aquinas or some other Scholastic.  You might as well ask: “Why is it only actual forces that act as forces?”  To ask the question is to answer it.  Nor is it profound to ask “Why is it N that is the natural state which occurs in the absence of unbalanced external forces?”, again, at least not if this is intended to raise problems for views of the sort defended by ancient and medieval philosophers.  You might as well ask: “Why do potentials remain potential until actualized?”  When we put things the way traditional writers actually put them, instead of in terms of Nozick’s pseudo-rigorous formalisms, the answers are obvious, and obviously hard to deny.  

It is also obvious why there is nothing the least bit “brute” or “unexplained” about the classical metaphysician’s “inegalitarian” “privileging” of actuality over non-actuality.  Nozick’s own discussion presupposes that actuality is “privileged” in this way, insofar as he is keen to explore various possible answers to the question of what “states” and “forces” might account for this or that aspect of reality.  For to raise the question of whether this or that “state” or “force” is the correct explanation of anything is precisely to ask whether this or that “state” or “force” actually obtains or is operative, and is thus available to serve as an explanans.  (This is true even of Nozick’s rather farcical discussion of the idea of a “nothingness force” which “nothings” other things, and even “nothings itself” -- see pp. 122-24, complete with the obligatory variables, and even a graph for extra “rigor.”  To ask whether there is such a force is precisely to ask whether it is actual and thus an available explanans.  It is, by the way, standard Nozick shtick to devote many pages to exploring ideas that are obviously non-starters and which he does not even believe himself.  Some of Nozick’s fans seem to find this kind of mental onanism entertaining.  I find that it gets tiresome pretty fast, especially in a book of over 700 pages.)

Now, Nozick’s Ultimate Multiverse proposal crucially depends on the suggestion that there is something fishy about “inegalitarian” theories.  The idea is that if no state of affairs is special or “privileged” in the way “inegalitarian” theories suppose, then there is no reason not to regard all possible worlds as equally actual.  In this way, Nozick’s “egalitarianism” -- which is just a rejection of “inegalitarian” theories -- underwrites the “fecundity assumption.”  But what Nozick does not realize is that what the classical, “inegalitarian” metaphysician regards as “privileged” is not this or that state of affairs or possible world, and not even the actual world per se, but rather actuality itself -- something that, as I have said, Nozick too implicitly “privileges” no less than the classical metaphysician does.  The classical metaphysician simply pushes this “privileging” out to its logical conclusion: Since actuality is more fundamental than potentiality, the ultimate explanation of things must be purely actual, without potentiality; for anything less than that would itself require actualization in some respect, and thus not be the ultimate explanation.

To be sure, Nozick suggests that his “principle of fecundity… den[ies] special status to actuality” insofar as it makes every possible world as real as the actual world (p. 131).  But what it really denies special status to is the actual world, not actuality itself.  That is to say, it denies that there is anything special about a universe which (say) includes human beings, began with a Big Bang 13.7 billion years ago, is governed by the laws of quantum mechanics and relativity, and so forth.  It does not deny that being actual is privileged over being non-actual; on the contrary, rather than denying this privilege to any possible universe, it extends it to all of them.  Indeed, its attribution of actuality to all of them is precisely what is supposed to be doing the theory’s explanatory work vis-à-vis answering the question “Why is there something rather than nothing?”

That Nozick does not understand the way ancient and medieval philosophers would approach that question is especially evident from the following passage:

To ask ‘why is there something rather than nothing?’ assumes that nothing(ness) is the natural state that does not need to be explained, while deviations or divergences from nothingness have to be explained by the introduction of special causal factors.  There is, so to speak, a presumption in favor of nothingness.  The problem is so intractable because any special causal factor that could explain a deviation from nothingness is itself a divergence from nothingness, and so the question seeks its explanation also. (p. 122)

This is, in fact, the reverse of what the classical metaphysician thinks.  For the classical metaphysician, God -- understood as pure actuality, subsistent being itself, and absolute simplicity -- could not have failed to exist, precisely because He is pure actuality, etc.  Hence His existence -- the existence of that which is the opposite of “nothingness” -- is the “natural state” of things, in the relevant sense.  When the classical metaphysician claims to explain why there is something rather than nothing, then, he doesn’t mean that sheer nothingness is the natural state of things and that we need to find out why it doesn’t obtain.  He means that the world of our experience, since it is a mixture of actuality and potentiality, composite rather than simple, etc., could have failed to exist, so that its explanation must lie in something distinct from it, something which actualizes its potentials, which composes its parts, and so forth.  And when we arrive at that explanation, we find that it lay in something whose existence is self-explanatory, precisely because it is pure actuality without any admixture of potentiality, absolutely simple, and subsistent being itself.  Rightly understood, then, the question “Why is there something rather than nothing?” leads us to conclude that nothingness is not the natural state of things and that there is nothing without an explanation -- precisely the opposite of where Nozick seems to think the question leads.  And we are led to these conclusions however many possible universes -- one, two, or all of them -- exist.  For that they could have failed to exist -- that they are mixtures of actuality and potentiality -- is what leads to those conclusions.