[2] In the midst of this activity, a small volume by Nicholas Rescher has appeared, The Riddle of Existence. In it, Rescher echoes approvingly Heidegger's claim that this is `the most fundamental question of metaphysics' and accordingly undertakes (what he has described privately {2} as) a `thought experiment', presumably to see for his own curiosity what an answer along certain lines would look like. What sets his particular effort apart is his mainstream approach. He, distinct from other recent writers on the topic, tries to offer an answer within the tradition and style of major philosophical thinkers during the last three hundred years. Unlike Nozick who believes radical theories are called for {3}, Rescher adopts a far more conservative line.
[3] In seeking an answer, it will not do, of course, to cite earlier states of matter. The Big Bang may do as a theory of cosmology, but it is singularly unsuccessful as a theory of cosmogony. Similarly for all the contemporary variants of that theory advanced by astrophysicists. {4} For all of these theories take as their starting point the existence of something or other, and then proceed to trace the evolution, from that primordial `something', of the successive states of the physical universe. And so `the' metaphysical question remains: "Why should there have been (or Why was there) this primordial `something', rather than nothing?"
[4] If an explanation is to be sought, it cannot, then, be a causal one. It cannot be one couched in terms of an antecedent state of the material world. But if causal explanations -- which presuppose the prior existence of a physical universe -- are inappropriate, on what then shall we rely?
[5] There are, of course, supernatural explanations. But these are out of vogue among secular philosophers. And even supernatural explanations might be thought to be nothing more than a delaying tactic. Why, after all, should a Creator have made something rather than nothing? In any event, few secular philosophers are prepared to regard questions of cosmogony as being questions of natural theology.
[6] Regarding causal explanations as inappropriate, and theological ones as `inherently questionable etiquette' (Rescher 1984: 15), Rescher returns (reverts) to (what he calls) `a teleology of value'. {5} In effect, he sets out -- as have many others -- to try, with suitable adjustments, to derive fact (and existence) from value.
[The existence of things] can be explained in terms of laws. ... To account for the world's law framework itself, we must move to another level of explanation, one that proceeds in terms of a teleology of value. The most promising way to explain the world's laws is that they are constrained to be as they are because only so can certain values be maximized. (Rescher 1984: Intro.)[7] There are two crucial stages to Rescher's argument. In his account, things -- i.e. material objects -- exist because, just as logical necessity implies truth (actuality), natural necessity (nomicity) implies actuality (31). {6} And nomicity, in its turn, is grounded in axiological principles.
[8] Rescher had earlier explored the theme of nomicity in another book, his Scientific Explanation of 1970. There he had argued that nomicity is essentially an epistemic or doxastic category which, in the practice of science, we `project' onto the world; that ultimately, nomicity "lies in the eyes of the beholder" (Rescher 1970: 116).
[9] There are, to be sure, considerable difficulties in any such view which accounts for natural necessity as a projection by us of a psychological conviction onto the world. Such a mind-conferred necessity could not possibly have the effects of a bona fide natural necessity. For example, the impossibility of a material object exceeding the velocity of light is viewed, standardly, among those who plump for nomicity, as being due to a real feature of Nature itself and not to our having done science and having beliefs about Nature. Similarly, the traditional problem of free will is thought to issue from `real' constraints within Nature itself, not from anyone's beliefs, however strong or well-grounded they might be, as to what Nature is ultimately like.
[10] But a sea change in thinking occurs between the earlier and the later book. In The Riddle of Existence, Rescher has so completely abandoned his earlier views about nomicity, that he does not even mention that earlier theory.
[11] In the later work, Rescher embraces wholeheartedly an uncompromising nomological necessity. Gone is any suggestion that nomicity has its ground in us. Nomological necessity is now presented -- as it had been in the writings of generations of prior necessitarians -- as an authentic metaphysical necessity, `out there' in the real (physical) world, owing nothing whatever of its role -- in constraining physical possibility -- to the existence of consciousness.
[12] Yet even with his adopting a view which locates natural necessity `in' the world, there remains a considerable problem. For one might object that there is an illegitimate shift in Rescher's argument from "actuality" to "existence". For example, even though the logically necessary truth that all regular chiliagons [thousand-sided polygons] (abbreviated "C") have interior angles of 179 degrees 38 minutes 24 seconds (abbreviated "I") [i.e., Nec(x)(Cx => Ix)] implies the actual truth that all regular chiliagons have interior angles of 179 degrees 38 minutes 24 seconds [i.e., (x)(Cx => Ix)], none of this implies that there are any actual, that is to say, existent, regular chiliagons [i.e., (Ex)Cx]. This logically necessary truth about polygons has, since the nineteenth century, been understood conditionally, more specifically, in just such a manner as not to have existential import. Concomitantly it has become standard practice to regard natural laws analogously (whether or not they are to be regarded as nomological), i.e. as conditionals, often as counterfactual conditionals, which imply their own (actual) truth, but do not imply the existence of their subjects.
[13] Rescher's way around this objection is to invoke a stronger sort of nomicity, not the nomicity of `ordinary' natural laws, but the special -- in effect, world-making -- nomicity of what he calls "protolaws": protolaws "can be conceptualized as representing conditions for existence rather than conditions of existents: They are not so much laws of nature as laws for nature" (1); "it is not by chance that things exist in the world ... but by natural (or, better, protonatural) necessity" (32); and again, "protolaws require the existence of things" (33). (If we let "#" stand for "it is protonaturally necessary that", then it appears that Rescher's theory advances a powerful existence-principle, viz., "#(x)(Px => Qx) -> (Ex)Px" [where "->" stands for "entails", i.e. "logically implies"].)
[14] What are these special super-nomic, or proto, laws: "#(x)(Px => Qx)"? Where are we to seek them? And how shall we recognize them?
[15] At this critical juncture, Rescher is exceedingly brief. The weight of his argument is borne by two paragraphs and three footnotes (34-35), and of these, principally by the first paragraph.
On such an approach, we would accordingly begin by looking to the fundamental field equations that delineate the operation of forces in nature: those which define the structures of the space-time continuum, say the basic laws of quantum mechanics and general relativity, and some fundamental structural principles of physical interaction. Principles of this sort characterizing the electromagnetic, gravitational, and metric fields provide the basic protolaws under whose aegis the drama of natural events will have to play itself out. And the existence of things would then be explained by noting that the fundamental equations themselves admit of no empty solutions -- that any solution that satisfies them must incorporate the sorts of singularities we call "things".... For such an approach to work, it would have to transpire that the only ultimately viable solutions to those cosmic equations are existential solutions. (34)[16] To the last sentence, Rescher adds a footnote:
Or perhaps, even should `empty solutions' exist, they might be highly unstable; the protolaws would then be such that, under their aegis, an existentially empty state of things is inherently liable to undergo a phase transition, having a natural inclination to slip over into an `occupied' condition. (34, note 35)[17] All of this is conspicuously programmatic. Rescher never tells us that these candidates for protolawfulness in fact have the features he attributes to protolaws. Indeed his footnote is frank admission that he does not know whether they do. He entertains the possibility that they do not, and then speculates that empty states might be "unstable" (his word) and be "inherently liable to undergo a phase transition", i.e. to break through from possible existence to real existence. {7}
[18] By the very nature of the case, it is in principle impossible to conduct an empirical test of the proposed theory (just as we cannot conduct empirical tests for other sorts of metaphysical posits: nomicity itself; substance; souls; etc.). The acceptability of the theory must rest on considerations lying within metaphysics.
[19] There is, in Rescher's theory, a familiar metaphysical principle at play, to wit, that essence (structure, lawfulness) `precedes' existence. (English does not serve us very well in this instance. "Precedes" is not of course used here literally in the sense of "being antecedent in time to" but in an ontological sense of "being a sufficient condition of".) According to this account, protolaws `precede' existence, i.e. `precede' the existence of matter, condition that existence, and indeed, ultimately, necessitate that existence. The relationship is not that of causal necessitation, but a kind of natural necessitation.
[20] My own metaphysical intuitions, revealed in The Concept of Physical Law (Swartz 1985), are the exact opposite. There I argued that physical laws (can most profitably be conceived to) `take their truth' from the way the world is: that the world exists (for no reason whatever) and physical laws are nothing more or less than (a certain subclass of) true descriptions of the world. In that contrasting theory, there is no physical nomicity, still less is there proto- nomicity, i.e. a (timelessly?) ready and waiting structure to constrain physical possibility and necessitate actual existence.
[21] I can (just barely) understand the idea of a proposition's `taking its truth' from the way the world is. (My difficulty stems from the obscurity of the notion of proposition, but I'll let this pass.) To use an example I have used elsewhere, I can understand that my wearing a blue shirt `makes' (in the semantic sense of "makes") it true that I am wearing a blue shirt. But I cannot conceive of the remotest possibility of this relationship working conversely: that somehow its being true that I am wearing a blue shirt `makes' (or accounts for, or explains) my wearing a blue shirt. Yet it is just this latter sort of model, or one very like it, which Rescher and all other espousers of nomicity would have us adopt. According to Rescher's account, protolaws are `already' or `antecedently' (in some timeless sense) true, {8} and it is because these laws are true that the world is the way it is, including, in his latest theory, that the world includes material objects.
[22] Perhaps this latter objection is not so much a refutation but rather an argument to the effect that protolaws are not, after all, propositions. If propositions do not constrain nature, and protolaws do, then (by cesare) protolaws are not propositions. But if not propositions, what, then, are protolaws? Clearly they are not physical things. They are abstract, or subsistent, things. Like `ordinary', garden-variety, nomological laws, protolaws are subsistent things whose `nature' it is to constrain reality. However, unlike ordinary nomologicals, protolaws have further, extraordinary, powers: the latter possess the power to be able to `drive' their subjects into existence (to account for what Rescher calls a "phase transition"). In short, they are sui generis.
[23] The question, at this point, should not be taken to be "Is Rescher right?" There cannot, so far as I can see, be any determinate resolution of such a debate. The question should be regarded, rather, as being "How much machinery will we accept to drive the cogs of our metaphysical apparatus?" It is clear that Rescher is not only happy with, but positively demanding of, several strata of metaphysical overlay. He wants this machinery because he refuses to allow that the question "Why does something exist?" cannot rationally be answered.
[24] But even with the positing of protolaws, Rescher is only halfway through his announced program. For he has yet to account for the truth of the protolaws.
[25] Protolaws do not `take their truth' from the way the world is; quite the contrary, it is their role, as we have just seen, to constrain (or condition) the existential truths of the world. Protolaws are not the dictates of a god. And protolaws do not bear their truth as inexplicable `brute fact'; for if so, we would have simply sloughed off the original problem onto a merely invented device having the requisite explanatory properties by fiat.
[26] The protolaws which actually obtain are, according to Rescher, those which maximize certain values. But why should protolaws (/the world /nature) be value-maximizing? Rescher offers this answer:
It is the great advantage of a principle of axiology to be in the position to provide materials of its own explanation. Principles of economy, simplicity, etc., are literally self-explanatory by virtue of being optimal on their own footing.... A value principle... must validate itself. (53)Thus the prospect of an infinite regress of potential explanatory steps is terminated. Existence is explained in terms of proto- nomicity; proto-nomicity, in its turn by being value-maximizing; and the particular values concerned, it is asserted, need no further explanatory justification: they are self-validating.
[27] (We cannot read much of this and not detect the ghost of Leibniz stalking, repeating sotto voce, "the best of all possible worlds, the best of all possible worlds". The difference between the two is that Rescher has substituted "maximally value-enhancing" (43) for "best".)
[28] Is such a theory plausible? Which exactly are these cosmic values that protolaws maximize and which, therefore, ultimately drove (drive?) this world from mere possibility into full-blown existence?
[29] They are emphatically not theological values (45); nor are they ethical values (45 and 48); nor are they aesthetic values (48); nor, finally, are they pragmatic values (48-49). Indeed these `values' have nothing whatever `to do with' human interests. They are, rather, ontological and "cosmic" [Rescher's word and quotation marks (48)]. Such values "represent objective and impersonally characteristic features of systemic order" (49).
[30] Rescher mentions such values as simplicity, harmony, systemic elegance, uniformity (49), and economy (53). And he explains that it is an empirical question -- to be decided by a `remote induction' (50) from observable features of the world -- what values protolaws in fact maximize.{9} This is a tall order. So tall in fact that I think it beyond realizability.
[31] If axiological principles are to explain protolaws there must be some strong relation between them. Either protolaws are deducible from axiological principles or they are reducible to such principles.
[32] On the proof-theoretic model of deducibility, it seems unlikely (or impossible) that axiological principles -- such as (the cosmic version of) "Maximize simplicity", "Maximize harmony", etc. -- could have the requisite logical features to serve as premises in proofs whose conclusions were protolaws such as (to cite Rescher's own best guesses as to what the protolaws will turn out to be): general relativity, quantum mechanics, the metric field equations, etc.
[33] The laws of general relativity, quantum mechanics, etc. are expressed by equations of considerable complexity involving mathematical esoterica such as tensors and eigenvalues, etc.; quantitative descriptive predicates such as "m" (mass), "v" (velocity), and "t" (time), etc.; and physical constants such as "c" (the speed of light) and "h" (Planck's constant), etc. If these sorts of propositions are to be deducible (in a proof-theoretic manner) from axiological principles, then those principles themselves, as a matter of logic, would have to (1) contain the descriptive (i.e. nonlogical) terms of these latter protolaws and (2) be logically more powerful than their implications. {10} But no principles, of the sort "Maximize harmony", which are themselves devoid of such descriptive terms as "m" and "h" could possibly serve as premises which validated a proof-theoretic inference to propositions (or protolaws, if protolaws are not propositions) which non-vacuously contain such descriptive terms.
[34] Invoking reducibility seems no more promising for Rescher's program. In this latter instance, one might imagine the claim being made that although the vocabulary (i.e. the descriptive predicates) of the protolaws is not literally to be found within that of the axiological principles, the former might, even so, be `reduced to' and hence explained by, the latter. The model presupposed would be that already familiar in the reduction of, e.g., phenomenological thermodynamics to statistical mechanics, or of the (anticipated) reduction of mental states to states of the central nervous system. In the former of these cases we have been told, for example, that -- even in the absence of synonymy of terms (there is no term in statistical mechanics that is synonymous with "temperature") -- temperature nonetheless literally (and metaphysically) just is a particular statistical feature of the micro structure of a substance. Or, again, we are told that intentional states just are (assuming certain metaphysical principles, e.g., materialistic monism) certain states of the central nervous system.
[35] The trouble with trying to adapt such an account in support of Rescher's theory is that for a reductive account to be epistemically effective, i.e. for it to provide grounds for our believing it to be true, it must satisfy certain methodological desiderata. Such accounts must in some sense, however weak, be empirically testable. {11} But even before we can begin to seek empirical evidence for the `linkage' between the two, that linkage must be formulated in explicit statements. It is easy to overlook the essential prose `accompaniment' to reductive accounts. Outside the formal aspects of the `reduced' theory and the `reducing' theory, one must say (i.e. explain) how the two link up. Without that gloss -- occasionally downplayed in some philosophical accounts -- what one would have, would be, simply, two disjoint theories. Somewhere, almost always outside the formalism of the theories, in an extensive prose (i.e. natural language) gloss, scientists must `link up' the two theories, the `reduced' theory and the `reducing' theory. {12} For example, if the vocabulary (e.g. "temperature" or "mass" or "velocity" or "h" [Planck's constant]) does not appear in the grounding theory, then the linkage between the terms (or laws) of the two theories must be explained, just as is done in all physics texts, in prose outside the formalism of the theories. There is no escaping that requirement: if it is not met formally, it must be met informally, otherwise there is no explanation. {13}
[36] Rescher's announced program is to try to explain why the universe exists. To say that unknown axiological principles necessitate protolaws, without explicitly showing the connections between specific instances, is at best to provide the form of an explanation. But it is virtually impossible to see how that form could ever be satisfied by an actual case. For it is extremely implausible that axiological principles, such propositions as (to repeat) the cosmic version of "Maximize simplicity" could ever serve -- even in the sort of account favored in explications of reduction -- to explain such (proto)laws as those of general relativity or quantum mechanics. It is not just that the axiological principles are devoid of the relevant vocabulary (i.e. lack the unique descriptive terms of scientific principles), they also lack the requisite specificity. The protolaws are highly specific, indeed their information content is diminishingly close to maximum. To explain why such laws obtain, rather than any of an infinity of alternatives, the explanatory `base' (Rescher's cosmic axiological principles) would have, also, to be highly specific, i.e. to have very high information content. But to date, we have not been given even a single instance of such a principle, let alone any good grounds for believing that such principles have the requisite degree of specificity.
[37] Two properties in Rescher's theory that are essential to cosmic axiological principles -- viz. (1) their being self-validating and (2) their being very specific -- are in direct conflict. Unless the protolaws -- presumably, general relativity, quantum mechanics, etc. -- are themselves self-validating, then it is beyond plausibility that an axiological foundation specific enough to `account' for them should be self-validating. And no one, so far as I know, has ever suggested that quantum mechanics or general relativity is self-validating.
Bradley, Raymond D. and Norman Swartz (1979) Possible Worlds: An Introduction to Logic and its Philosophy. Indianapolis: Hackett.
Hawking, Stephen W. (1988) A Brief History of Time. New York: Bantam Books.
Jennings, Raymond E. (ed.) (1990) Being and Somethingness: Essays in Honour of John Tietz. Burnaby, B.C.: Simon Fraser University.
Nagel, Ernest (1961) The Structure of Science: Problems in the Logic of Scientific Explanation. New York: Harcourt, Brace & World.
Nozick, Robert (1981) Philosophical Explanations. Cambridge, MA: Harvard University Press.
Rescher, Nicholas (1970) Scientific Explanation. New York: Free Press.
---- (1984) The Riddle of Existence: An Essay in Idealistic Metaphysics. Lanham, MD: University Press of America.
Swartz, Norman (1985) The Concept of Physical Law. New York: Cambridge University Press.
---- (1991) Beyond Experience: Metaphysical Theories and Philosophical Constraints. Toronto: University of Toronto Press.
{2} Personal correspondence, May 21, 1990. Return
{3} Nozick expresses pessimism about all `traditional' approaches:
The question cuts so deep, however, that any approach that stands a chance of yielding an answer will look extremely weird. Someone who proposes a non-strange answer shows he didn't understand this question. Since the question is not to be rejected, though, we must be prepared to accept strangeness or apparent craziness in a theory which answers it (Nozick 1981: 116).Return
{4} Anthropic theories, even when advanced by astrophysicists, are not themselves cosmological theories but are separable metaphysical complements to such theories. Return
{5} Rescher also rejects a few other approaches to the problem, ones (including Nozick's) which we cannot -- for lack of space -- review here. Return
{6} Hereinafter, all page citations unaccompanied by a cross-reference to the References section are drawn from Rescher 1984. Return
{7} Stephen Hawking is far more tentative than Rescher:
Even if there is only one possible unified theory, it is just a set of rules and equations. What is that breathes fire into the equations and makes a universe for them to describe? The usual approach of science of constructing a mathematical model cannot answer the questions of why there should be a universe for the model to describe. Why does the universe go to the bother of existing? Is the unified theory so compelling that it brings about its own existence? (Hawking 1988: 174).Return
{8} "... protophysical laws ... `precede' nature" (25-26). Return
{9} That such values could be knowable only experientially, i.e. empirically, does not count against those values being necessary. There is no good argument to the effect that all knowable necessary truths can be known a priori (see Bradley et al. 1979: 168-170). Return
{10} The precise logical strictures involved -- for a proof-theoretic account -- were laid out more than thirty years ago in The Structure of Science (Nagel 1961: 366-380) where Nagel argued persuasively that it is logically impossible to derive (deduce) the existence of so-called `emergent' properties from a set of propositions which do not themselves contain descriptive predicates naming those properties. Whether the properties are emergent or not, the logical requirements -- for a proof- theoretic account -- remain the same. Return
{11} I have treated this issue in Beyond Experience (Swartz 1991: 311-27). Return
{12} Again, see Swartz 1991, pp. 24-38, esp. 34-38. Return
{13} I am, obviously, tacitly arguing that explanations are not, after all, logical relations, but are epistemic successes. The arguments, pro and con, are well-known. Return