4.1 First, however, it will be as well to accept such a multi-tiered framework of explanation on its own terms. Does it disqualify, not just materialism (yes!), but any form of monism (no, IMO), from prospectively exhausting the furniture of the world?

4.2 Crucial to Chalmers' argument against the possibility of a reductive explanation of our minds is the claim that conscious experience does not logically supervene on the physical. Were mind logically to supervene on low-level facts, or were the notion of "high-" and "low-" level facts untenable, or were our notions of supervenience itself subtly incoherent or misleading, then the argument would fail. And if the equations of physics don't rule out stuff with - or consisting in - intrinsic subjective properties, then Chalmers' project of showing the failure of logical supervenience will depend on there still being an "independent logical space" (p35) for human minds to vary independently of these microphysical facts. Granted both the equations of physics and the intrinsic properties of the stuff they describe, I'm going to argue contra Chalmers that there is no way experiential manifolds can fail to "supervene". Experiential manifolds [a.k.a. minds, a.k.a. multitudes of DNA-driven egocentric virtual worlds churned out in genetic host-vehicles by evolution] support a distinctive mode of experience, nominally insentient matter. This typically gets contrasted with another mode of experience, which is merely a restricted extra-somato-sensory type of consciousness. An explanatory gap opens up when we try to explain one by reducing it to the other. Posed like this, the problem can't be solved. One could no more ontologically reduce a symphony to a landscape.

4.3 If it can't be ruled out except by fiat that complex minds do supervene on "low-level" facts - though they most certainly don't supervene on materialist factoids - then there is a more radically conservative and ontologically parsimonious option than property dualism, epiphenomenalism, pan-informationalism, or any jury-rigged system of further options. The ontology of monistic panpsychism in effect denies any "explanatory gap" between "high-" and "low-" level facts - if such an idiom is to be retained. To be a feasible proposal, the low-level facts will of course need to be essentially different in character from the tenets of classically-inspired materialist orthodoxy. The difference is in their intrinsic nature: the relational properties of micro-qualia will be captured by a type-identical formalism. This is because the inherently phenomenal micro-stuff must still accommodate all the prodigious experimental, explanatory and technological successes of contemporary physical science. More positively, they are far better capable of redeeming the prodigious failures of science; which tend to be far less-touted, particularly by the drug-naïve.

4.4 The existence of a ubiquitous and minimal what-it's-likeness as intrinsic to the stuff of the world, or better, as the stuff of the world itself, is an option which Chalmers does seriously and sympathetically consider. Yet he ultimately rejects it. I want to understand, as best I can, his reasons for doing so; not least because - unsettlingly - Chalmers (see below) gives the best exposition I've come across of the family of positions one of which I'm advocating. For it is quite consistent to agree with the materialist that QM is mathematically (Gödelian complications aside!) complete and yet to argue that what presently pass for physical concepts "can be given topic-neutral analyses that might pick out underlying phenomenal properties" (p135; Chalmers is here expounding the quasi-Russellian position of the late Grover Maxwell).

4.5 This approach, Chalmers acutely notes, delivers a form of idealistic monism which is radically unlike Berkeley's. For on this view, "the world is not supervenient on the mind of an observer, but rather consists in a vast causal network of phenomenal properties underlying the physical laws that science postulates" (p155). Chalmers notes that "Physical laws can be interpreted as laws that connect intrinsic properties (or properties constructed out of these) to their relational profiles (or to complex relational structures). We are led first to the relational structure of the causal network, and only slowly to the underlying intrinsic properties."

4.6 Indeed, "No set of facts about physical structure and dynamics can add up to a fact about phenomenology" (p118). Yet any conclusive inference to the failure of logical supervenience, were it to succeed, would depend on physics additionally granting us an understanding of the noumenal essence of the stuff whose behaviour its equations exhaustively describe. And this is what physics just doesn't do.

4.7 So why does Chalmers reject such an account which he expounds so incisively? He acknowledges that "Physics tells us nothing about what mass is, or what charge is; it simply tells us the range of different values that these features can take on, and tells us their effects on other features"(302). He grants that "Physical theory only characterises its basic entities relationally, in terms of their causal and other relations to other entities" (p153). He's prepared to canvass an interpretation that avoids both epiphenomenalism and Cartesian dualism. He's willing to speculate that phenomenal properties "implement" the dispositions that physical theory deals in. Yet he still doesn't buy monistic idealism. If our ignorance qua physicists of what "breathes fire into the equations" really is unqualified, however, then recourse to, at the very least, a topic-neutral construal of the QM formalism would seem inescapable. The equations, as before, exhaustively describe the behaviour of the pro tem topic-neutrally construed stuff of the world. Our enforced agnosticism about the stuff's intrinsic nature means one might just as well allege that insentience as much as sentience fails logically to supervene on its properties. We simply don't know whether it does or doesn't; the panpsychist is betting, he fancies with inside information, that it does.

4.8 I'd guess Chalmers would argue that even if it naturally supervenes, it doesn't logically supervene. Unfortunately, any serious discussion of the issues here would demand at least a treatise on modality and doubtless a better cocktail of nutritional supplements. In this piece, I shall just for once have to dogmatise.

4.9 The notion of non-epistemic possibility, or any substantive distinction between x and (redundantly IMO) necessarily x, is empirically ungrounded and philosophically ill-motivated. In practice of course it is indispensable. I'm just not convinced it can do all the philosophical work Chalmers wants it to do. For unfortunately contingency is a deeply mysterious and ill-defined notion. Any bid to show that anything could really be different from how it is founders in assuming what it seeks to prove. There are no real grounds for believing that what exists could naturally or logically be otherwise. If sometimes we don't understand why something couldn't be otherwise relative to certain background conditions, then no deep ontological consequences flow from our epistemic frailties. Logical possibility is often taken to be possibility in the broadest sense ["It is useful to think of a logically possible world as a world that it would have been in God's power (hypothetically!) to create, had he so chosen." (p35)] Yet the fact that we find the need to resort to such conceits is revealing. It shows the problems our intuitive conceptions bring when made to bear independent ontological weight.

4.10 Surely, the counter-argument might run, it needn't logically be the case that everything is the way it is? Well, it depends on your conception of logic and your choice of primitives, axioms and operators. With the right primitives and operators, perhaps everything else logically supervenes, albeit non-provably. Intuitively, one is liable to find the notion of anything existing logically absurd. The empirical evidence suggests one is massively in error. So one's concepts are wrong. Wholly at sea, one can't simply junk one's entire categorical raft altogether [Neurath]. So one tries to rework parts of it while guessing it is all rotten. Any sort of existence strikes me as daft; but I've had to learn to live with it. So assuming that something somehow analogous to 0 ought to obtain as the default condition, I look rigorously to define what precisely would have had to obtain for 0 to be the case as decency suggests it should. If it does seems nonsensical and absurd that anything exists - though something-or-other manifestly does - then one seeks to explicate one's intuitive concept of 'nothingness' or complete absence of properties so it becomes formally equivalent to zero. On this analysis, our pretheoretic idea of "no properties at all existing" as commonly understood is muddled. But something analogous to this notion i.e. zero net properties - is not merely intelligible, but true; and what we call the world is the expression, mathematically describable by physics, of this timeless logical principle. Zero is a surprisingly rich concept of astonishing fertility in maths, however queasy one may feel about using anything which smacks of reifying "Nothing". Given the properties of this ill-understood primitive, however, then an attempt can be made to extract everything - logically, physically, phenomenologically, and mathematically - from its unique status.

4.11 No, as it happens, I wasn't on acid at the time I was delivered of this little epiphany; just feeling intellectually desperate. For at times even much of physics can seem remarkably akin to stamp-collecting. More to our purposes here, however, Chalmers still has work to do to show that our minds fail logically to supervene on the micro-facts.

4.12 Substantive possibility aside, even epistemic possibility is more problematic than one would wish. Again relative to certain background conditions, one can apparently conceive of all sorts of things as being otherwise than how they are. This is a far from straightforward idea. Inevitably, simply to get off the ground, such a notion already presupposes that one's mental episodes can possess - as distinct from simulate - an otherwise ineffable feature of the Multiverse known in the trade as "broad" content. This is the marvellous sort of relational content that "ain't in the head". All rational debate must assume something like it, even if no one quite knows what 'it' is. Calling it 'naturalistic' demystifies without illuminating its nature. In the case of contingent propositions, the problem is far worse. For one can't even invoke causal co-variation (etc) to legitimate their force. Yet unless realism about subjunctive conditionals can be reconciled with scientific naturalism, then one hasn't even really conceived of the purported referents of one's thought-episodes as being different from how they are. Scientifically legitimating indicatives is hard enough; subjunctives pose difficulties which are orders of magnitude worse.

4.13 Ascribing to some of one's real, spatio-temporally located thought-episodes the capacity to access real contingency is a practical necessity. Yet once again it's worryingly like a (sometimes) heuristically useful fiction - and just the sort of snow-job at which natural selection, played out over the aeons, excels. Even Everett's Multiverse, a false friend to the possible world enthusiast, simply is a superposition, not a set of alternate, absolutely non-interfering realties. [Actually, our little multiverse is arguably only one vacuum fluctuation among many, just "one of those things that happen from time to time". If so, then cornucopian googolplexes of multi-verses may proliferate all over the place. Scary] Yet in no case does this entail that there is unactualised potentiality (or whatever) for anything being otherwise than it actually is. The notion of real contingency in any form is adaptive. In default of any substantive grounds for swallowing it, this is grounds for harbour ing greater suspicion of its credentials, not less.

4.14 If contingency isn't taken to rest on some psychologistic criterion of actual conceivability, but instead on a notion(!) of possible worlds, then it might(!) seem to enjoy greater objective warrant. Yet the very concept of a possible world again presupposes the existence of the contingency which it purports to explain. [In all honesty, these one-liners should be flagged to warn the unwary that there is rather more to be said]. And granting God any say in the matter, even as a philosophical facon de parler, opens up a different can of worms altogether.

4.15 How does this review's scepticism about the force of (obscurely) literal contingency tie in with the alleged failure of logical supervenience of the conscious mind on low-level micro-facts? Well, in reality, Chalmers, like most of us, inevitably does entertain intuitions in a general sort of way about what the basic stuff of the world broadly can and can't be. These intuitions sit uncomfortably with any professed ignorance of noumenal essences. "Physics tells us nothing about what those [intrinsic] properties might be. We have some vague intuitions about these properties based on our experience of their macroscopic analogs - intuitions about the "massiveness of mass, for example -..."

4.16 Insofar as these "macroscopic analogues" really are inherently similar - and not just functional counterparts - then perhaps our intuitions could be justified; and the failure of supervenience vindicated. Yet the concept of a macroscopic analogue needs exploring carefully. It rests, in practice, on the notion that we are presented with, or granted direct epistemic access to, or apprehend transparently [or whatever], a mind-independent macroscopic world. The virtual realist favours myriad data-driven simulations of local real-world macro-patterns which our individual minds are running. A folk-sanctioned everyday macroworlds, on the other hand, is supposed to be an extra-mental reality on the other side of one's skull (and not simply on the other side of a causally co-varying skull-surrogate of a somato-sensory homunculus) which perception mysteriously allows us communally to access. By contrast again, inferential realism, although committed to a vast mind-independent Multiverse as the best explanation of our autobiographical virtual worlds, is not committed to the existence of a distinctively macroscopic stratum of reality whose properties supervene on layers "beneath" it. 'Virtual' strata, yes, real strata, no. Instead, macroscopic neo-cortically-generated material objects, located in the coherently awake mind, are our reified counterparts of recurrent patterns in fields of what-it's-likeness in the inferred Multiverse beyond. Perhaps!

4.17 Moreover, even if we did have such innocent access to Reality, in just what respects are "macro-" properties analogues of the intrinsic properties of their "underlying" micro-constituents? If tepid grey porridge is supposedly the extrinsic analogue of intrinsic phenomenological mind [ pah!], then what faith can we have that our alleged "extrinsic" and relational handle on the properties of particles/fields/superstrings(!) captures their intrinsic nature any more faithfully?

4.18 For reasons to be elaborated, if panpsychism is to seem remotely viable and a Chalmers-inspired dualism averted, two controversial conditions must be fulfilled. The first is a concession of the feasibility - if not inescapability, given our present ignorance - of a topic-neutral reconstrual of the formalism of physics. The second is the falsity of orthodox perceptual direct realism.

4.19 I'd argue that Chalmers is pushed into a semi-sanitised and non-interactionist dualism rather than the panpsychist monism to which he might otherwise lean mainly because he implicitly relies on an untenable background theory of perception. It's this theory, I think, which leads to his 'inside'/'outside' dichotomy and his proposed ontology of ubiquitous information states linking 'internal' micro-phenomenology and 'external' physics. The Conscious Mind doesn't offer a full-blooded account of its author's philosophy of (what passes as) sensory experience and its relationship to the rest of the world. Yet the assumptions of an implicit theory creep into his story in a number of places. In a crucial passage, Chalmers alleges that the panpsychist position entails a "dualism of 'accessible' and 'hidden' physical properties" (p136). The intrinsic properties of the world-stuff posited by the panpsychist aren't the "properties physics deals with directly"(153).

4.20 Yet what are the properties physics (physicists?) deal with directly? In what sense are they "revealed"(p136)? Is this an allusion to [our notional] shared access to a supposedly token-identical [and not merely type-similar] macroscopic world? This presumably self-subsistent macroworld would be where public experiments are performed. Some of its notional high-level properties would serve as analogues to microphysical primitives. Or, on the other hand, is the revelatory metaphor an allusion to explicitly represented properties of basic microphysical entities whose mathematical description the theorists play around with? If the latter, then the argument is undercut by the need for topic-neutralism given our present absolute ignorance of "what breathes fire into the equations and makes there a world for us to describe." [Or do we know something Hawking doesn't? Maybe. But not this...].

4.21 If the 'directness' and 'accessibility' of physical properties is indeed an allusion to a notional shared macro-world, then the position slides into a descendant of folk-style perceptual direct realism. And as the phenomena of e.g. phantom limbs, near-death experiences, Penfield's electrode studies, ketamine-induced out-of-the-body episodes etc corroborate, one doesn't even have direct access to the body of one's extra-psycho-cerebral host organism external to the mind's somato-sensory-homunculus, let alone exotic pastures further afield. As Kant once quipped, it's a small world. More anecdotally, let's suppose that, for instance, one is gazing down at the visual body-image one normally intimately consorts with ['my' body(-image)] when not out of one's head on the dissociative anaesthetic ketamine. In this condition, then the prospect of abandoning science for the intellectual equivalent of tea-leaf reading seems more tempting than it probably does right now. For the perceptual realist, undergoing a ketamine-induced OBE must seem like a clear refutation of scientific materialism; though I'd argue that it's materialism and perception that ought to be retired, not quantum mechanics. For what it's worth, I find "in-the-body" experiences are intellectually at least as odd as "out of it" modes of existence. This is because organic mind seems too warm to allow coherent, multi-modally-bound states; and coherent states would seem to offer the best hope for superposing (pre-fronto-cortical-] self-concepts and the different [visual, kinaesthetic etc] modes of body-image.

4.22 Chalmers own commitment to the IMO fatally misleading presuppositions of perceptual direct realism is revealed in such phrases as "How could [consciousness] possibly arise from lumpy gray matter?"(p3) "Whoever would have thought that this hunk of gray matter would be the sort of thing that could produce vivid subjective experiences? And yet it does." (p251). Yet does it?

4.23 When one is dreaming, for instance, and one inspects the properties of a brain, then that lump of porridge which one's body-image manipulates isn't generating the sentience of another mind. It is simply a fleeting coalition of psychons of one's own mind-dust as described by the canonical equations. When one is in that peculiar state of consciousness self-flatteringly known as being "awake", then a more-or-less type-identical brainy experience would be activated only under the tightest selection by peripheral psychoneural impulses. These are distally caused via the activation of surface transducers in the host vehicle by patterns in the external world. Possibly, in that external world, an experimentalist or neurosurgeon is working with electrode probes in a laboratory or operating theatre. In this setting, the brain which the neurosurgeon's etc own visual-cum-somato-sensory body-image is manipulating in his experiential manifold will be serving as the causally co-varying analogue of distinctive macro-patterns of sentience in his [alert, locally anaesthetised] patient's mind/virtual world. Unreassuringly, the reality must be vastly more complicated than the prose used to express it. Yet what sounds natural on the ear would only be a telling advantage if our concepts were the gift of God; and even then, absentee landlords are not to be trusted.

4.24 By way of illustration again, it is easy to slip into thinking of consciousness as being by its nature ethereal and nebulous. Chalmers remarks on how it is "diaphanous" and "intangible"(p3). Yet tangibility and opaque solidity are highly distinctive properties of certain modes of consciousness. In one's dreams, for instance [invoking dreams is convenient when one wishes temporarily to avoid using question-begging accounts of perception], one is not simply conscious of a rock, a sunset, a table, or a brain. These rocks, sunsets, tables and brains are fleetingly-bound experiential manifolds of sentience themselves. [I think this self-referential thought qua a fleetingly bound experiential manifold of sentience myself]. Rocks, even dream-rocks, are thoroughly opaque and tangible. One can ham it up playing Dr Johnson all one wants. One can even feel the pain. One just won't enjoy any greater degree of success at refuting idealism, whether Berkeleian or naturalistic, than the robust-minded Doctor. Real philosophy is hard work.

4.25 Now when one "wakes up", the weighted connections which mediate such dream-phenomena will, just as yesterday, be activated in the normal way only by peripheral, environmentally-triggered stimulation. [In dreamless sleep there's no experiential manifold; just an incoherent quasi-punctate psychic sludge]. These weighted psychochemical patterns of one's waking hours don't inexplicably cease to manifest the same type of solid, abrasive etc modes of consciousness which dream-rocks and tables do at night. [One's life may regain a certain narrative drive; but that's another story.] Nor do these patterns of excitation take on instead the role of mere resonant vehicles [or whatever] for putative real world access. If one doubts their daytime status, then allowing electrode-stimulation of one's awake mind/brain is a splendid way of getting disabused. Another is to dim the selecting peripheral inputs, drop a lot of acid, and grant one's virtual worlds a greater measure of daytime independence than they usually get the freedom to express. My advice would be: don't do it; not yet at any rate.

4.26 To recap. If there were any such spectacle as single, mind-independent macroscopic world - rather than selfish DNA-driven egocentric travesties of salient fitness-relevant macro-patterns in the superposition of a unitary Multiverse - then the conceptual division of reality into ontological levels could do real explanatory work.

In fact, folk-inspired theories of perception are multiply misleading in other respects too.

4.27 First, perceptual realism (as distinct from inferential realism or some sort of VR-inspired data-driven simulationism) commits us to the existence of a non-branching, quantum-theoretically anomalous macroscopic world. This invites physically ill-grounded talk of dynamically collapsing wavefunctions.

4.28 Second, perceptual realism invites a division of the world into a hierarchy of levels on which the whole notion of reductive explanation and its inferred panoply of sub- and super- venience relations depends. For analytic philosophy's spatial metaphors of "high-" and "low-" level facts, and the vexed notion of supervenience with which they are so often associated, are an artefact. They turn on our tendency to divide reality into one tangible and manifest macroscopic world, and an invisible microphysical counterpart which underlies it. Invoking supervenience (once pithily dubbed "epiphenomenalism without causation") is admittedly preferable to a tendency to make "low-level" events literally cause "high-level" events (cf. Searle) with which they are supposedly identical. For brain states can't cause the very same mental states with which they are alleged to be one and the same. They can cause subsequent states, which confusingly may then be known under a mental as well as physical description. Unfortunately the usual philosophical vocabulary serves as a perennial temptation to extract ontological rabbits from an epistemological hat.

4.29 In either case, it's worth recalling that if QM is complete then the Multiverse exists as a single awesomely big superposition. For sure, given the evolution of a hot, simple and symmetric universe via spontaneous symmetry-breaking phase transitions into the complex universe of this particular era, the world from our perspective simulates in varying degrees the evolution of different levels. Non-equilibrium thermodynamics throws up an increasing number of [what may be more-or-less systematically interpreted as] virtual layers. It is sometimes very useful to treat these persisting or recurrent patterns as though they were ontologically autonomous echelons of reality with their own laws - just as one might divide up the different "virtual" machines running on a computer. Yet the world itself doesn't literally instantiate, as distinct from progressively simulate, different layers. The fundamental fields/strings/n-branes/psychons or whatever just wiggle energetically in different ways taking on different values. They contrive, I think, to bewitch some of their throwaway quick-and-dirty patterns into taking an extended family of stratigraphical metaphors more literally than it deserves.