Home » Library » Modern Library » Steven Conifer Ac

Steven Conifer Ac

The Argument from Consciousness Refuted (2001)

Steven J. Conifer



Many Christian philosophers (e.g., Richard Swinburne) advocate a theistic argument which has begun of late to garner considerable popularity in the literature. The argument is one from human consciousness, i.e., it appeals to the phenomenon of human consciousness as its main premise. That phenomenon, so the argument goes, is somehow unlikely or perhaps even impossible on the assumption of atheism (or naturalism). Let us call this argument "the Argument from Human Consciousness for the Existence of God" (AHC).[1]


There are various versions AHC, some more far-reaching than others. For the purposes of the present essay, however, the argument may be formulated quite simply, as follows:

(A) It is a fact that human consciousness exists.

(B) That fact can be adequately explained within a theistic framework (i.e., one which posits God’s existence), whereas it cannot be adequately explained within an atheistic (or naturalistic) framework (i.e., one which denies God’s existence).

(C) Hence, there is a fact which only theism can adequately explain.

(D) Therefore, God must exist.[2]

There are at least four objections which can be raised against AHC. Let us proceed to explore them.


The onus of proving AHC’s premise (B) rests squarely with the advocate of AHC. To simply advance it as a flat matter of fact is question-begging. In other words, since the premise in question is far from obviously true (indeed, it is profoundly controversial), the advocate of AHC must somehow demonstrate that itis true. Until he or she does so, the argument may be reasonably dismissed.


AHC’s step (C) does not follow from its premise (B). Even assuming that theism can adequately explain the fact at hand (i.e., human consciousness) and that atheism (or naturalism) cannot, it by no means follows from that alone that only theism can adequately explain that fact. That is, there are alternatives to theism other than atheism (or naturalism) and nontheism in general (the category to which both agnosticism and noncognitivism[3] belong) which may also be able to adequately explain it. Take, for example, pantheism, deism, and the myriad breeds of polytheism which have abounded (and, to varying degrees, still do) in the dominant religions and philosophies of the East. How might it be that, say, the god of deism or the Hindu god Vishnu (or even the Norse god Thor) could not adequately explain it? I see no reason whatever why none of those deities should be able to sufficiently account for it, yet none of them is the god of (classical) theism. Hence, the inference from AHC’s premise (B) to its step (C) is invalid, thus rendering the argument unsound.


It is not at all clear that the first part of AHC’s premise (B) is true, i.e., that theism can indeed adequately explain human consciousness, especially since most advocates of AHC (as well as the vast majority of theists in general) conceive of God as a transcendent being and transcendent beings are supposed to exist outside space and time. There is a certain conceptual difficulty here, namely, how a being who is outside time might do any thinking or acting at all. Thinking and acting, as they are commonly understood, both necessarily (viz., by definition) involve time. Thus, the claim that God, so defined, performed some action(s) or conceived some thought(s) which resulted in the instantiation of human consciousness seems exceedingly nebulous, if not outright incoherent. But AHC presupposes precisely that claim. Consequently, AHC suffers from a certain kind of obscurity which renders it, at best, highly dubious.

There is also an empirical question which needs to be addressed, a kind of modus operandi problem. That is, AHC fails to supply any information concerning how God, presumably a transcendent being, supposedly brought about the phenomenon at issue, i.e., human consciousness. Hence, it is an explanation which is grossly incomplete (assuming one could even properly regard it to be explanation at all, which is certainly debatable). By failing to elucidate the matter of just how, exactly, God allegedly accomplished the feat in question, AHC implicitly proclaims it to be a "great mystery." But in the words of Theodore Drange, "The purpose of explanation is to dispel mystery, not introduce it."[4]

Due to both its obscurity and inadequacy, AHC ought to be rejected.


While it may be difficult or even impossible to conclusively establish it, a strong case could be made for the contention that, contrary to what the latter half of AHC’s premise (B) states, atheism (or naturalism) can, in fact, adequately explain the fact under scrutiny (i.e., human consciousness). As Drange notes:

[M]any scientists do make contributions to such a project. It is the subject of ongoing research within many sciences, including brain physiology, biology, and chemistry… although it is as yet incomplete, it has not come up against any insuperable difficulty. Scientists show how the chemical origin of life is compatible with certain natural laws. By appeal to mutations and natural selection, they explain the mechanisms by which life developed from simple to complex. They have theories about how "reason" operates and how it might have survival value… formulations of materialist explanations [of consciousness] in such terms [as] "time and chance acting on matter," "random neuron firings," "minds and intelligence arising by chance out of inchoate matter," etc.) [which frequently characterize theistic critiques of such explanations] seem quite misguided and off the track… Science has come a long way towards explaining [consciousness] in materialist terms, and to dismiss all that work [out of hand] is not a piece of reasoning to be taken seriously.[5]

Michael Martin shares this view, writing:

[T]he difficulties that Swinburne finds with scientific nomological explanation[s] of psychophysical correlations either are not as serious as he supposes or else are based on misunderstandings… the most serious problem he finds with such explanations is that because of the very different nature of brain states and mental states, it is difficult to see how simple scientific laws could explain the diverse correlations between them… There is a problem here only if one misunderstands the sort of laws that can be used in such an explanation. [Analogously, it] might be supposed, in order to explain the [relevant] correlations, that there must be particular causal laws explicitly relating falling trees to blocked traffic and disruption of electrical circuits, but this assumption is false. Laws concerning falling bodies can be used to explain the correlation between some of the items covered by the description "disrupted electrical circuit" and some aspect of a tree’s fall. Similarly, laws concerning electricity can be used to explain the correlation between the items and the lack of electric power, and so on. In other words, the complex whole can be broken down into component parts, and these can be explained separately. The same approach can be used to explain mental phenomena… Once we look at the situation in this way the bulk of the difficulties noted by Swinburne dissolve.[6]

As regards Swinburne’s theistic account of consciousness, Martin explicates its defects as follows:

[A]lthough personal explanations are familiar and natural in ordinary life, we know that the way one’s intention brings about some action has a physiological basis… when I intentionally move my finger it may seem that there is a direct connection between the intention and the movement. But we know that this connection is possible only because of a complex physiological causal relation… between my intention and the movement of my finger. In the case of God there is no such relation. According to Swinburne, the relation between God’s intention and the intentional action is direct and unmediated. Given our background knowledge of how personal explanations work in ordinary life, personal explanations of psychophysical correlations in terms of God’s intentions seem improbable. All our evidence indicates that intentions do not directly cause physical events. (Original italics)[7]

In addition to all of this, even if one accepts theistic personal explanations in general, critical questions remain concerning theistic personal explanations of psychophysical correlations in particular. As the late philosopher J.L. Mackie put it:

Has God somehow brought it about that material structures do now generate consciousness? But then is this not almost as hard to understand as that material structures do this of themselves? Or are we to regard each body-mind connection, for example the supervening of each state of perceptual awareness on the appropriate sensory input and neurophysiological disturbance, as the fulfillment of a fresh divine intention, so that sensory perception is, strictly speaking, an indefinitely repeated miracle so we have an endless series of divine interventions in the natural causal order? But further… Could not omnipotency superadd a faculty of thinking as easily to a block of wood as to a brain? If materialism has the difficulty in explaining how even the most elaborate neural system can give rise to consciousness, theism, with its personal explanations and direct intention-fulfillment, has at least as great a difficulty in explaining why consciousness is found only in them. (Original italics)[8]

Furthermore, a supernatural personal explanation is perfectly consistent with the correlations between brain states and mental states’ being occasioned by the deliberate action(s) of either a finite deity or a number of finite deities. Hence, AHC fails to show that consciousness can be explained only be positing the existence of an omnipotent, omniscient, omnibenevolent, and transcendent deity who created and rules the entire universe, viz., the god of classical theism.

In his 1996 book Is There a God?, Swinburne writes the following:

It may well be that certain primitive brain states cause the existence of souls- as the foetal brain reaches a certain state of development it gives rise to a soul connected with it. But what it could not cause is- which soul is connected with it. It could not be the powers of this brain, of the molecules of this foetus arising from these genes, which cause it to be the case that my soul is connected to this brain and yours to that, rather than vice versa. It would be equally compatible with all the regularities between kinds of event (this kind of brain organization and the existence of a kind of thing- a soul) that science could ever discover that you and I should have been connected to brains in the opposite way to the way in which we are connected. There simply is no possible scientific discovery which anyone could ever imagine which would explain why it happened this way rather than that way.[9]

There are several problems with this. First of all, Swinburne seems, to a considerable degree, to just presuppose the falsity of HM, stating simply that "[t]hat view seems obviously false"[10] and that the truth of dualism "stares us in the face."[11] But that HM is indeed false (or even improbable) is no way obvious. As was shown in section 2, above, myriad philosophers and scientists, far from regarding it as absurd, find it extremely compelling (especially in light of its simplicity). The most notable among them is surely Daniel Dennett, who would doubtless deem Swinburne’s markedly cavalier dismissal of HM to be both egregiously unfounded and rather archetypal of many theistic philosophers’ indefensible inclination to write off the outlook as manifestly bogus when it is anything but. Such a dogmatic approach may be effective when addressing an audience whose members share the given bias, but it simply won’t do as an argument which purports to have a broad dialectic value and which aims to persuade a neutral, even marginally scientifically astute individual.[12]

Second, Swinburne’s remarks (particularly his last) fail to acknowledge the significant progress which scientists have made in their efforts to explain consciousness in a strictly materialist context, as discussed by Drange, above. His (Swinburne’s) claim that "[t]here simply is no possible scientific discovery which anyone could ever imagine which would explain why it happened this way rather than that way" is a mere assertion, in no way whatever supported by any scientific data. Nor is there anything inherently contradictory in the notion that there should occur just such a discovery. It would seem, then, that his claim is not only groundless, but flat-out false.

Third, suppose for a moment that science really were incapable (even in principle) of making the discovery at hand (which, of course, it is not). Could it not simply be a brute fact that what Swinburne describes "happened this way rather than that way"? I do not offhand see any reason to disbelieve so, and certainly Swinburne provides no reason to that effect. Not only does it seem intuitively correct to suppose that certain phenomena should ultimately have no explanation at all, but quantum physics has demonstrated already that at least a few such phenomena really do occur (e.g., the moment at which a subatomic particle begins to exist, the decay of a given atomic nucleus at a particular time, etc.). So, even supposing that Swinburne were right about what he says in the above passage (i.e., that it is somehow inconceivable that science should discover an explanation for the phenomenon in question), there would still be no reason to accept his conclusion that only God could adequately account for why certain brain states give rise to certain (corresponding) mental states (assuming, of course, that the latter exist at all).

In his older and much more comprehensive tome, The Existence of God, Swinburne says this:

Let k be the [premise] that there is an orderly (and beautiful) world. Let e be the existence of [consciousness]. Let h be… the hypothesis of theism- that there is a God. P(e/~h.k) is low- it is improbable that the world should just have the kind of complexity with which a dictionary of mind-body connections would set out, without some explanation… regularity in the midst of complexity cries out for explanation… it does not look at all plausible to suppose that there is a scientific explanation of these phenomena… for reasons of simplicity, the most probable personal explanation is one in terms of the agency of God… because such an explanation is in itself the simplest kind of personal explanation…[13]

There are numerous difficulties with this line of argumentation as well. The foremost among them is that it succumbs to analogues of the "Burden-of-proof," "Invalidity," and "Obscurity & Inadequacy" replies, above. An almost equally grave flaw which plagues it is the immense unclarity surrounding the kind of theistic personal explanations of psychophysical correlations to which Mackie alludes in the passage cited above.

Moreover, as the philosopher Quentin Smith has aptly observed[14], we could just as well let the "h" in Swinburne’s formula (which derives from Bayes’s theorem) represent the hypothesis that "there exists a mostly malevolent deity" as we could theism, i.e., the hypothesis that "there exists an omnibenevolent deity." Thus, by replacing

h = there exists a perfectly good god


h’ = there exists a largely unloving god

and then modifying the given formula so that it states


we would arrive at the conclusion that h’ is more probable than ~h’ (i.e., that the hypothesis that there exists a largely unloving god is more probable than the hypothesis that there exists no such god). Ergo, if we accept Swinburne’s claim that consciousness is improbable barring divine intervention, then we may say of H’ precisely what Swinburne says of H, namely, that P(E/H’.K) > P(E/K). And while certainly there is no reason whatever to suppose H’ better evidentially supported than its negation (since there is no reason whatever to accept Swinburne’s claim), given the vast amount of (ostensibly) gratuitous suffering and premature death in the world and the absence of a satisfactory theodicy,[15] there is every reason to suppose it better evidentially supported than H. Put another way, in light of the available data, H’ is plainly more prima facie likely than is H. (Obviously, a mostly malevolent god would desire a universe with intelligent life no less than would one who is perfectly good, as the actualization of moral evil necessitates the existence of intelligent life no less than does the actualization of moral good; a being cannot inflict its malevolence upon inanimate matter, yet most certainly can do so upon intelligent creatures capable of suffering and premature death.) Hence, we may reasonably assert that P(E/H’.K) > P(E/H.K).

But isn’t this paradoxical? How could the same evidence confirm equally well two incompatible hypotheses? As Quentin Smith explains:

If it appears paradoxical, it is because one is confusing relative confirmation (which I am here using ‘confirmation’ to express) with absolute confirmation (which I shall use ‘makes highly probable’ to express). The same evidence cannot make highly probable each of two incompatible hypotheses, but it can increase the probability of each of two incompatible hypotheses (i.e. make the two hypotheses more probable than they would have been without the evidence).[16]

Needless to say, however, such an implication of his formula was hardly one that Swinburne anticipated; and, to be sure, it shows to be unsuccessful his effort to employ that formula so as to prove h more probable than ~h.

Furthermore, the "h" need not stand for a hypothesis which posits the existence of any deity, whether benevolent or otherwise. Suppose, instead, that it were to designate the following:

h” = there exists a species of exceedingly intellectually and technologically advanced extraterrestrials who somehow (covertly) endowed human beings with consciousness

Again, by performing the appropriate substitution and modification, we could produce the formula P(e/h”.k) and thereby reach the conclusion that P(e/h”.k) > P(e/h.k), viz., that h” is more probable than h (i.e., that the hypothesis that there exists a species of exceedingly intellectually and technologically advanced aliens who somehow [covertly] endowed human beings with consciousness is more probable than the hypothesis that there exists a perfectly good god). Since H” is at least equal in scope, simplicity, and fruitfulness to H, and because the former is undoubtedly less obscure (or conceptually problematic) than the latter, it seems reasonable enough to declare H” explanatorily superior to H and thus (if only slightly) the more plausible of the two hypotheses. It follows that P(E/H”.K) > P(E/H.K), or, at least, that P(E/H”.K) = P(E/H.K). Clearly, then, Swinburne’s formula does absolutely nothing to bolster the case even for supernaturalism, let alone (classical) theism. And since it is by no means clear that there is any better reason to suppose h than to suppose h”, the latter is just as prima facie plausible as the former. Therefore, it is it just as legitimate to conclude that P(e/h”.k) > P(e/h.k) as it is to conclude the converse. Thus, it is obvious that Swinburne’s formula lends not even the most negligible shred of plausibility to theism.

In addition, while perhaps somewhat plausible two decades ago, Swinburne’s claim that scientists cannot link mental events with neurophysiological states can no longer be sustained. Admittedly, due to the complexity of the brain and the difficulty in observing live brain processes, it is no easy task. Nevertheless, in recent years it has become possible to observe and measure certain thought processes and the reciprocal chemical and electrical events in the brain. Memory locations of certain thoughts have been ascertained, and progress in this field is rapidly accelerating (especially with the aid of such machines as magnetic resonance devices), with technological breakthroughs occurring regularly. Thus, although the correlations between some mental events and neurophysiological states cannot as yet be clearly determined, it is nonetheless virtually indisputable that they do exist. By way of analogy, while we are not as yet able to fully understand all the complex interactions which occur in global weather patterns, it would be quite absurd to declare that those interactions do not really take place. For example, merely because we cannot predict with much (if any) accuracy the course of a cyclone, we should hardly assert that there exist no meteorological regularities which govern that course, or that it is impossible that we should one day discover them. Similarly, we cannot assume that simply because scientists cannot as yet identify all the correlations between mental events and neurophysiological states, those correlations do not exist. To be sure, the scope of mind-brain correlations which has already been established is more than sufficient to support the contention that such correlations exist even in those cases where they are not readily discernible.

Furthermore, Swinburne’s claim that "it is improbable that the world should just have the kind of complexity with which a dictionary of mind-body connections would set out, without some explanation" suffers from both a certain vagueness of its own and an entirely unfounded (implicit) assumption. As regards the vagueness, it arises from his use of the word "complexity," which certainly admits of considerable subjectivity: what one person takes to be complex another may very well regard to be quite simple (and vice versa), and there seems to be no objective procedure whereby either of them might prove the other mistaken. With respect to the assumption, it is one to which I made reference earlier, namely, that the phenomenon in question (i.e., the peculiar manner in which the mind and brain interact, assuming they do at all) must have some explanation. However, as was explained above, that simply isn’t so, for it might very well be that that phenomenon is just a brute fact in the same way that the acausal behavior of subatomic particles and atomic nuclei appears to be such. At the very least, Swinburne does not acknowledge this possibility, which he most assuredly ought to.

Yet another objection to Swinburne’s argument, above, is that many philosophers, including some eminent theists, outright reject the notion that any hypothesis has any a priori probability whatever. Others simply deny that theism and naturalism in particular have intrinsic, a priori probabilities. An example of the former is the theistic philosopher Alvin Plantinga, who rejects the view that hypotheses can have purely a priori probabilities.[17] An example of the latter is the Christian theologian and philosopher John Hick, who regards theism and naturalism to be exhaustive, all-encompassing worldviews which, by virtue of their very comprehensiveness, preclude the existence of a neutral scale upon which might be weighed their relative probabilities.[18] Swinburne fails to adequately address such objections.

Something also needs to be said about Swinburne’s assertion that "for reasons of simplicity, the most probable personal explanation is one in terms of the agency of God… because such an explanation is in itself the simplest kind of personal explanation." First, as indicated in section 7.65, above, due to the colossal murkiness which afflicts the "God Hypothesis" itself, it is not at all clear, despite all that Swinburne says, that that hypothesis actually constitutes any explanation whatever. Furthermore, even if it were the case that that hypothesis is "in itself the simplest kind of personal explanation," I fail to see that it is the simplest explanation altogether. To be certain, postulating the existence of a seemingly superfluous entity (e.g., God) violates parsimony irrespective of how well that entity might fit into some allegedly explanatory framework within which one happens to mistakenly operate. And clearly, in view of all that has been said, it may be reasonably submitted that no such additional entity need be postulated in order to adequately account for human consciousness. Hence, postulating any such entity would indeed require that one disregard parsimony, from which it follows that Swinburne’s assertion is erroneous.

Finally, quite apart from AHC (whether Swinburne’s particular version of it or any other), there is no good objective evidence that God exists. None of the theistic arguments which has ever been put forward in the public arena is any good.[19] Conversely, there are literally dozens of atheological arguments which have yet to be clearly refuted, most notably the evidential Argument from Evil, Drange’s Arguments from Nonbelief and Confusion, and the sundry Arguments from Incoherence.[20]

As should be overwhelmingly evident by this point, AHC’s premise (B) is utterly without merit. And, in any case, AHC is invalid, as was shown in section 7.64, above. Accordingly, I conclude that AHC is a failure.[21]


[1] I employ this abbreviation so as to clearly distinguish this argument from Theodore Drange’s atheological "Argument from Confusion" (which is abbreviated AC).

[2] It is not my claim that this is a precise formulation of Swinburne’s argument from consciousness per se. Rather, it constitutes merely my attempt to construct an argument which accurately represents a more general form of Swinburne’s, viz., any argument which proceeds from the premise that, given atheism (or naturalism), human consciousness is either unlikely or else impossible.

[3] See Theodore M. Drange’s "Atheism, Agnosticism, Noncognitivism," (<URL:https://infidels.org/library/modern/theodore_drange/definition.html>).

[4] Drange, Writings Regarding the Bible (1998), p. 103.

[5] Drange, "The Drange-Wilson Debate: Dr. Drange’s First Rebuttal," (<URL:https://infidels.org/library/modern/theodore_drange/drange1.html>).

[6] Michael Martin, Atheism: A Philosophical Justification (Philadelphia: Temple University Press, 1990), pp. 218-219.

[7] Ibid., p. 220.

[8] J.L. Mackie, The Miracle of Theism (Oxford: Clarendon Press, 1982), pp. 130-131.

[9] Richard Swinburne, Is There a God? (New York, N.Y.: Oxford University Press, 1996), p. 89.

[10] Ibid., p. 39.

[11] Swinburne, The Existence of God (New York, N.Y.: Oxford University Press, 1979), p. 166.

[12] While later in his 1996 book (pp. 71-77) Swinburne does make a rather flimsy attempt to defend two common dualistic arguments (namely, those which appeal to the so-called "privileged access" which a person has to his mental states and theoretical brain transplants), both of those arguments are highly controversial and their chief premises are enormously vulnerable to challenges from hard materialists such as Dennett. To all but assume them to be true (as Swinburne does to a large extent, arguing for them in only a very sketchy fashion) is question-begging. This is also so in the case of his argument from phenomenal properties (e.g., "blueness, painfulness, smelling of roses"), to which he appeals on pp. 164-166 of his larger book, The Existence of God.

[13] Swinburne, The Existence of God (New York, N.Y.: Oxford University Press, 1979), pp. 173-174.

[14] Quentin Smith, "The Anthropic Coincidence, Evil, and the Disconfirmation of Theism," (<URL:https://infidels.org/library/modern/quentin_smith/anthropic.html>)

[15] See, especially, Drange’s Nonbelief & Evil: Two Arguments for the Nonexistence of God (Amherst, N.Y.: Prometheus Books, 1998).

[16] Smith, "The Anthropic Coincidence, Evil, and the Disconfirmation of Theism"

[17] Alvin Plantinga, "The Probabilistic Argument from Evil," Philosophical Studies, Vol. 35, #1 (1979).

[18] John Hick, Arguments for the Existence of God (London: Macmillan, 1970), pp. 32-33.

[19] For thorough refutations of the theistic arguments in question, see (<URL:https://infidels.org/library/modern/theism/arguments.html>)

[20] To access these and other atheological arguments, see (<URL:https://infidels.org/library/modern/nontheism/atheism/arguments.html>)

[21] At the risk of beating a dead horse, I would be remiss if I failed to mention Michael Martin’s potent and rather clever parody of TAG (an argument rather akin to AHC), which he calls TANG ("the Transcendental Argument for the Nonexistence of God"). It is located on the Secular Web at (<URL:https://infidels.org/library/modern/michael_martin/martin-frame/tang.html>)

Steven J. Conifer is the president of R.U.S.H. (Rationalists United for Secular Humanism) at Marshall University in Huntington, WV and can be reached via e-mail at email removed.

all rights reserved