Gone too is the belittling of secondary qualities and of emotions. The latter can correspond with the facts, and the former, suitably construed, represent ways things are. We even admitted ineffable subjective states, such as what it is like to be a bat, or to be the persons we are, experiencing time and death as we do, and conscience and mystery. For we were able to reject the view that everything real must be brought under an objective description, just as we rejected any other doctrine of universal reducibility. In this spirit we confirmed metaphor as among the indispensable ways of expressing faces of existence; reduction to literal discourse, let alone to aseptic precision, proved needless. Thus was each domain seen to enjoy its appropriate autonomy; there was no need to commit monopoly in any of its forms.

Gone finally is the idea that "all is becoming, all is changing, all is in passage out of the past and into the future, and so all causes and effects come and go -- and all is mortal -- and nothing else is real" (§3.4.5). Whatever may be the case with the familiar 3-world of common sense, the 4-universe -- the whole of spacetime -- is eternal in the fivefold sense of being (i) not subject to becoming, to change, to process or to corruption and decay; (ii) not in time, and thus not to be dated as past, present or future; (iii) a "tenseless" existent; (iv) with no parts distinguished by being transiently past, present or future, or still to come, relative to one another; and (v) of such "duration" that it neither ceases to be nor comes into existence. In addition, it is the uncreated, explanatorily independent, metaphysically necessary, self-existent First Cause of all that is.

In these and other ways, then, did few things if any prove to be nothing but material things. Yet even though we can thus tighten our belts without narrowing our minds, the fact is that we have tightened our belts. That is, we have tightened our inventory of what there is to include only the mathematical-physical. Not everything is nothing but a physical or even a natural thing, yet everything is either a spatiotemporal sum of basic physical entities, or (if sets exist) a set in some rank of the hierarchy of sets (INV, §4.1). Further, we have endorsed the physical discernibility of all discernibles (MND, §4.2), and the determination of all truth by physical truth (TT, §4.3).

When conjoined with a realism of the sort explained in §1.1, these three principles are the minimal theses characteristic of all versions of physicalism properly so-called. The present philosophy therefore is not physicalist in name only, but in principle. And this means that despite rejecting so much traditionally associated with physicalism, the present view appears to threaten belief in the existence of God; for God seems nowhere to be found in its inventory of what there is. Indeed the present view could well be a greater threat to theism than traditional physicalisms, not despite rejecting so much, but because of rejecting it. For so long as physicalism is reductive, monopolistic, antagonistic to much that is human, and so on, it will seem an obviously inadequate account; theists may well remain correspondingly undisturbed by what such a view would imply. But a nonreductive, non-monopolistic physicalism of the present sort is not so easily dismissed; the usual objections -- typically anti-reductive -- are blocked.

Of course there are nonphysicalist varieties of metaphysics that might also succeed in avoiding reductive and other monopolistic pretensions. But they are likely to prove cold comfort to the theist. The unifying entities posited by such a metaphysics must exist, yet in many instances the claim that they do flunks a prerequisite for truth, typically the objectivity prerequisite; nor is it at all clear that the spirit of such a metaphysics could be retained in any revision that does take full account of contemporary knowledge of spacetime (§§1.5(vii), 3.4, 5.0). Even if the prerequisites for truth happen to be satisfied, there are further hurdles. For one thing, the metaphysics must be able to unify all the domains of discourse by means of its alleged unifying entities, in such a way that truths about the latter determine the truths in the former. A necessary condition of doing so is to unify the sciences. But it is exceedingly difficult, if not impossible, to argue persuasively that truths in the various sciences are determined ultimately by anything other than physical truth. This is especially clear in the case of the natural sciences. What is it that determines, say, the truths of chemistry, if not truths of physics? And as regards the latter, it is even more difficult to argue, in anything approaching convincing detail, that such truths themselves are determined by further truths. What in detail is the connective theory (or happy family of such theories) which would enable one to infer that and how the specifics of quantum physics or of general relativity are determined by particular truths about the alleged unifying nonphysical entity or entities? And what particular truths are these?

At every step of the way, the theist will be haunted by the realization that nonreductive physicalism very likely has already succeeded in unifying the sciences. The theist must therefore argue extra hard for the extra entities, on pain of violating Occam's injunction against multiplying them beyond necessity. One such argument might be that the extras are required for unifying the domains beyond the sciences. But if we have learned anything in this essay, it is that they are not required. Meaning, metaphor, emotions, values, ineffable subjective states and more were all appropriately accounted for. Indeed religious discourse might prove to consist in the expression of such matters. If so, there is no reason why religious truth could not itself be determined by physical truth. The only question would be which religious sentences are true, and in particular whether they include `God exists'.

Suppose, however impossible it may seem, that some nonphysicalist theory clears all these hurdles. There remains another. Unless God can be found somewhere in its inventory of what there is, such a theory would appear to be of little use to theists. Of course this presupposes that the existence of God is at least in part a matter of what there is, and that belief in God's existence is warranted only to the extent that God appears in some warranted inventory. Yet this is a false presupposition, according to certain theists. Their idea is that theism is a matter not of what there is but of how we ought to talk about it, and how we ought to see it. The idea thus has affinities with the point in §§7.0-7.4 that metaphysics often is more a matter of values than of what there is, and that metaphysical disputes often are disputes about what discourse ought to take what sort of priority on what occasion for what purpose. Theists believe their discourse about God often ought to take precedence and in any case is true, though as lately noted its truth need not have implications for what there is, but rather for how to value it. If so, looking for God in any inventory is misguided, as is supposing that Occam's razor has very much to do with ascertaining the truth of God-talk.

We return to this idea in §8.2 and explore it further. Meanwhile let us assume that a metaphysics is of direct use to the theist only if God can be found somewhere in its inventory. The problem then is not only that there is no guarantee the inventory will include God, but that many theists themselves deny that any plausible contemporary metaphysics presents an inventory in which God can be found. Such theists include those who reject any God of the philosophers. Yet they also include those who base their rejection not on blanket condemnation of the relevance of metaphysics to theology, but on particular problems involved in construing the alleged existent as God. It is in this spirit that many theologians emphasize the particular problems involved in construing something in Whitehead's inventory as God, or in so construing Heideggerian Being, or whatever.

On the other hand, suppose God is explicitly mentioned in the inventory. Then we must ask how God got there. What line of reasoning could warrant such a posit? It cannot be any of the traditional arguments for the existence of God. They are too flawed, individually and taken together. The cosmological or first-cause argument presupposes the Principle of Sufficient Reason, among other things. But PSR is fatally vulnerable to the criticisms in §2.4 if not also to others, and the remaining presuppositions too are flawed. The design argument, on behalf of a divine explanation for the order and adaptation we see around us, presumes there is no equally good alternative explanation in purely natural terms, though of course there is; and some of the argument's other assumptions are just as problematic (in particular, the assumption that the alleged designer is identical with God). The ontological argument involves the notion of a being that exists by the necessity of its own nature. The argument falls prey therefore to objections of the sort sketched in §2.2 against this notion and against the kind of necessity it presupposes, as well as to others. If instead God is posited as the ground of morality, then to the usual objections we may add that morality already has an adequate ground, in the sense that moral truth is determined by natural fact. Likewise, if God is wanted for injecting meaning and purpose into an otherwise absurd world, we may add to the usual objections the fact that meaning and purpose were there all along, contained in the world in the sense of being determined by facts about it.

But suppose the line of reasoning for explicitly listing God in the inventory is none of these. Suppose instead that certain aspects of existence have so far found no explanation, and that God is invoked to fill the explanatory gap. Here the trouble is threefold at least. PSR seems once again presupposed, in the presumption that there must always be some explanation to be found. In addition, such a God-of-the-gaps typically has been vulnerable to unexpected advances in the relevant fields of knowledge, when they fill the gaps with some nonsupernatural explanation, as so often they have. Finally, in light of such advances and of the dynamic consilience of certain varieties of naturalism (the present one included), the inductive evidence favors the presumption that where there is indeed an explanation to be found, it will be in natural terms.

At this point some theists concede that no line of reasoning warrants including God explicitly in any list of what there is. They then make a virtue of necessity: here is a matter not of mere reason but of faith. But we saw in §1.5 how one's faith, if it is to be true, must satisfy the logical prerequisites for truth. Defenders of the idea of certification of a belief by faith alone might well be right to the extent there need be no positive line of reasoning sufficient to establish the belief's truth, or even to make it only probable. And they might rightly be encouraged in this view by the difficulties (if not downright failures) in all proposed criteria of truth (or at least in criteria of truth for religious assertions) -- difficulties rehearsed in §1.3. But they are wrong if they think we may be indifferent to the line or lines of reasoning involved in checking to see whether a belief satisfies the prerequisites. One's faith, however fervent, must be well-formed, nonempty, have no false presuppositions, be self-consistent, be consistent with all other truths, have no false implications, and satisfy the objectivity or invariance prerequisite. Otherwise it is not true.

For each of these prerequisites there are powerful arguments to the effect that theism flunks it. Even if theists can counter the arguments one by one, in the course of doing so their view gets revised, often in the direction of emptiness, as we noted in §1.5(iv) in connection with the problem of evil. The result is an increasingly complex, ad hoc set of defensive assumptions. One is reminded of nothing so much as the proliferation of epicycles and eccentrics in the waning days of Ptolemaic astronomy. Of course staggering complexity may be tolerated in a view so long as no better is available. Yet for at least a century theism has been increasingly on the defensive, intellectually and philosophically, mainly because so many reflective people have come to suspect there are better accounts of what there is.

Faced with all these difficulties, how might theists respond? They have four alternatives, logically speaking. Where theism conflicts with naturalism in general or physicalism in particular, they may (i) reject theism. Alternatively, they may (ii) reject naturalism; probably this is the instinctive first choice of most theists. Their next choice usually is to try to show there is no real conflict in the first place, by (iii) showing that naturalism and theism are not logically inconsistent with each other. Often this is attempted, in effect, when theists argue there is no inconsistency between science and religion, and when they (or some of them) insist the question of God is not a question to which Occam's razor could apply. But there is another, more fruitful way of trying to show there is no conflict, not by showing merely that the two views are not inconsistent, but by (iv) attempting a synthesis of the two, in which each enriches the other while correcting the other's occasional extremes.

Obstacles aplenty confront this fourth way, quite possibly insurmountable. All the same, theologians would be well advised to pursue it, for the very good reason that the other alternatives are so bleak. Alternative (i) is the bleakest of all, for theists. But even nontheists perhaps should concede that a wise and tolerant theism, if true, would be invaluable -- or valuable enough, at any rate, that its rejection should be a matter of bitter regret for all concerned, however dull their sense of loss may have become with the passage of time. As regards (ii), we have just been reviewing the strengths of nonreductive physicalism, plus corresponding weaknesses in other views. If these chapters do not establish the truth of some such naturalism, nevertheless they show how costly its rejection would be; few will have read this far only to turn back. As regards (iii), all of us ought ultimately to be dissatisfied with the conclusion merely that there is no logical inconsistency between naturalism and theism. For we would want to know why they are consistent, despite so many appearances to the contrary, and we would want to know the nature of this deeper compatibility. We crave coherence and unity, a craving that can never be satisfied by mere logical consistency, but ultimately only be some sort of synthesis, however uneasy and tentative.

What if the theologians believe no variety of naturalism is true, including the present nonreductive, tolerant sort? Even then they ought to explore how far theism might incorporate such a metaphysics. The reason is that true or not, naturalism is so widespread and deeply entrenched among philosophers and reflective laymen. For better or worse, it is the dominant view in some of the most influential parts of our culture. If theists hope even to gain a hearing for their view in an increasingly secular world, let alone persuade others, what better course than to explain and justify themselves in terms consistent with the naturalism they too assume (if only for the sake of argument)? Whatever the spiritual, emotional and cultural sources of their beliefs -- and whatever their justification in theists' eyes -- theists need to present a rational apologetic that incorporates an austere contemporary naturalism. For it is late in the day, and theists may have no other hope of restoring their views to the influence which, if true, they so richly deserve. To decline the attempt would be as though Aquinas had proceeded without the Aristotelian-Ptolemaic system of the world.

Indeed our situation in many ways resembles that of the 13th century, when translation of the Aristotelian corpus presented theists with a science and philosophy evidently superior to anything yet at hand, and threatened to shatter their picture of the world. In effect they too confronted the four alternatives above. Averroists adopted an analogue of the first: where scripture conflicts with Aristotelian-Ptolemaic science and philosophy -- that is, with the naturalism of the day -- scripture must yield. Bonaventure adopted an analogue of the second: in case of conflict the naturalism must go. Aquinas adopted what amounted to a middle path: evidently dissatisfied with such extremes and also with declarations of mere consistency, he strove to show how his theism could incorporate the seemingly hostile newly translated world view. The result was the powerful synthesis we call Thomism.

Our own picture of the world, descended from the medieval, has been shattered not by recovery of ancient text but by discovery in our own sciences from Copernicus to Darwin and beyond, and by related blows from philosophy. One by one religious claims have had to be abandoned or qualified, until it is unclear whether such claims have any content left at all. The impression often is of a series of rear-guard actions in a losing battle, delaying the inevitable, perhaps, but little more. And beyond the matter of particular advances in particular sciences which conflict with particular religious claims, there looms the matter of the general explanatory power of scientific naturalism, as outlined and extended in these chapters -- a power which, in view of Occam's principle, seems to leave no room for God in any plausible inventory of what there is. The sciences may well have no peer in ascertaining what there is, because of the organized and systematic way they hunt for the invariants in our perspectives, in line with the objectivity prerequisite. If so, then those who care about what there is and about the implications of objectivity have little choice, in their "apprenticeship in the school of the transcendent," but to absorb the teachings of an austere naturalism.

The moral seems clear. What the theologian requires above all is a simple, compelling theory of God, world and us, which does full justice to scientific naturalism -- full justice in the sense of assuming it is true, and of living with its full implications. Far easier said than done, of course, but the obligation merely to try, and, failing, to try again and again, has been neglected for so long that it is no longer seen as an obligation, only as a Quixotic ideal to which one might reasonably have aspired in the middle ages. Yet there is no more powerful way for the theologian to proceed, and a faith that seeks understanding would be well advised to seek its understanding, in this day and age, within some such framework of science and philosophy. Even the true believer probably should be satisfied with no less. The reason is that the most illuminating and long-lasting theologies of the past -- Augustine's, say, or Aquinas's -- were achieved not by denouncing or ignoring the best science and philosophy of their day, or even by presenting a theology merely not inconsistent with it, but by incorporating it into their rational theology as a framework within whose confines they could interpret and support the central theistic claims.

It would be absurd to suggest that the waning pages of this or any other book might rival the great theologies. What §§8.2-8.3 present are mere possibilities, though a true theologian might see fit some day to exploit them in full. The possibilities are connective theories, or rather sketches of them, presented in the same spirit as are theories in physics, a field in which it is not at all surprising to learn for example that "Bohr...doubted the [Heisenberg-Pauli] theory would be the great new revolution in physics because it was not suffciently `crazy."' Or consider the spirit in which Georgi and Glashow advance SU(5) as "the guage group of the world" in an effort to unify all elementary-particle forces: "Our hypotheses may be wrong and our speculations idle, but the uniqueness and simplicity of our scheme are reasons enough that it be taken seriously." The speculations of §§8.2-8.3 are presented in the same spirit, and are likewise revisionary rather than descriptive. In any case they have as their aim to beg no question against physicalists, other naturalists, or nominalists, but to accept their terms of the debate, if only for the sake of argument, and to show how, strictly within such limits, one might conceivably build a sound theism. And what better basis is there for this sort of apologetics than physicalism? It can hardly be accused of multiplying entities beyond necessity, or of any other tender-minded loading of the dice in favor of theism. In this respect its minimal theses are uncompromising.

Despite the efforts of the chapters to date, to say nothing of what a whole pantheon of naturalists might add, many theists will remain doubtful not only about the truth of naturalism in general and its severe physicalist versions in particular, but also about the truth of any "scientific image" of the world. Further, some will doubt not only the correctness of these views, but whether it is advisable even to contemplate expressing theism within such a framework, for purposes of rational apologetics or any other purpose, however widespread or influential naturalism might be, and however responsible for the intellectual component of rampant secularism. Let those who share these doubts read on. For they can still enjoy the spectacle of seeing an (alleged) arch-enemy having to open the rear door to what he threw out the front. Or, to change the figure, little is so entertaining as seeing the devil quoted to scripture's purpose.

Atheists will voice the opposite extreme of reaction to the hypotheses and speculations of §§8.2-8.3. One of them, a distinguished professional philosopher, writes, "What happened to the sound judgment that has pervaded the book till now? Is there a crowd of theologians somewhere who need to be placated?" Another regards my treatment of theism as a reduction to absurdity of the idea that we currently understand either the relation of determination or the physicalism that presupposes it. Still another insists it can only be illusory to believe that physicalism might allow for objective values, let alone for God, and doubts that such values and God have anything more to be said for them than does phlogiston.

When one encounters this sort of reflex hostility, which is fairly widespread among professional philosophers and other intellectuals, it sometimes helps to be candid about one's motives. Mine are mixed, perhaps beyond unraveling. But one motive I do not have is the desire to defend theism at all costs. I have no such irrevisable commitment. Instead I have a detached stance that horrifies many of my theist friends. For among my motives the overriding one is, as I like to think, a dispassionate curiosity about what the best evidence, argument and theory-construction might allow us to believe. This is a philosopher's motive, not a theologian's (perhaps because "curiosity is insubordination in its purest form.") Further, I believe that if the enterprise sketched in §8.2 does not succeed, or if no related attempt at a synthesis of theism and an austere naturalism succeeds, then the only alternative is atheism. The reason is that the other attempts to articulate and support theism, ranging from the classical arguments to various contemporary theologies, cannot survive the withering scrutiny of what I take to be the best of today's science and philosophy, a best that includes a minimal physicalist version of naturalism. Thus the failure, if it comes to that, of the contemplated synthesis of theism and physicalism, so far from being a reduction to absurdity of the latter, I would regard as a reduction to absurdity of the former.

Note in this connection that the theism entertained in §8.2, even if it were itself patently absurd, would be a reductio of the determination relation, and of the physicalism that presupposes the relation, only if the theism followed from the minimal physicalist theses. But it does not. In this respect there is a parallel with objectivism as regards values. The objectivism, we saw, does not follow from the minimal physicalist theses (unless it follows indirectly from realism as regards truth-value -- §§6.2-6.3). Instead it follows from them in conjunction with the assumption of Meta-Ethical Anti-relativism -- an assumption which in context is non-question-begging and for which there is sufficient independent evidence, as noted in §6.2.

Similarly, the theism entertained below follows from the minimal theses only in conjunction with a further assumption. But unlike MEA, this further assumption is problematic. It is a normative assumption to the effect that the objectively correct values include certain theistic values, as they are called in §8.2, together with the theistic ways of seeing and experiencing they give rise to. For some forms of theism -- some non-traditional ones -- the assumption is perhaps not too difficult to defend (as we see in effect toward the end of §8.2). For traditional forms, including the Judeo-Christian, the assumption is very difficult to defend, and §8.2 lists some of the possibly insuperable objections to it.