Maitzen and the Problem of Magic

This entry looks at Stephen Maitzen’s essay ‘The Problem of Magic’, which is available from his website here. Said essay is interesting not for the actual arguments it contains but for the implied metaphysical cost of its background. For the sake of charity I will say now that I am only aware of Maitzen’s work through a number of his essays in philosophy of religion and that these may not be an adequate representation of his ability as a philosopher.

In the course of the essay Maitzen sets out to argue several claims, most prominently that that the intelligibility of the universe does not require God and that theism or other forms of supernaturalism in fact threaten its intelligibility. I will focus my attention on his later thesis and only make a few observations on the first part of the discussion.

This starts with largely rhetorical questions on what ‘intelligibility’ here could mean. Maitzen attacks the view, which he accredits to no specific individual, that God’s existence is a prerequisite for the laws of logic holding. Here his responses are predictable enough: that we can attach no sense to the idea of the laws of logic not holding and that said laws would be required to state God’s existence or non-existence. For the purposes of charity let’s assume all talk of ‘the laws of logic’ here refers to narrowly logical necessity. Far be it from me to criticise this conclusion. The problem with the background claim is not the theism but the psychologism, the claim, popular in the nineteenth century, that logic depends on the mental operations of a thinker. Early analytical philosophy put paid to this position but a position it indeed was.

Assuming his readers accept this opening move, he tries to use it show Cartesian views on God as above the laws of logic lead to absurdity, a victory which I think most theists both contemporary and historical will grant him. Funnily enough he misses what would seem to be the biggest non-semantic problem with Descartes’ appeal to Divine goodness to justify even our trust in the laws of logic, which is that those very laws are needed to establish the link between goodness and not allowing continued deception or even the coherence of the cogito. Whether Descartes really upheld such a strong logical voluntarism is a matter for historical scholarship.

It is in the next section where Maitzen’s discussion takes a strange turn. Unsurprisingly enough he mentions Bernard Lonergan, a major proponent of intelligibility arguments¹, however rather than analysing Lonergan’s argument he quotes a criticism of it from R.M. Burns and moves to discuss Burns’ own pietistic sceptical argument (any attempt at a proof of the human capacity for objective knowledge must presuppose that capacity, so to avoid crippling scepticism we should make a transcendental wager that our capacity for truth is guaranteed by a divine being). Maitzen’s response to this is to look further into the Justified True Belief account of propositional knowledge and argue that infallibility is too high a standard for knowledge. Again I have little to quibble with here, though much of the material is really incidental to theism, Burns’ premise being an old argument for global scepticism often treated in the Foundationalism vs Coherentism debate.

Granting that Maitzen’s responses to these quasi-sceptical arguments are successful (it is in fact the best part of his essay) one must still allow that he has not worked very hard in elaborating the intelligibility implies theism thesis. He chooses to formulate it as the fact that the universe is susceptible to empirical, scientific discovery and explanation is evidence for God’s existence. This is different from the much bolder ‘transcendental argument’ claim that theism must be presumed in order for us even to grant the possibility of true knowledge, something even Descartes would blanch at. If we are granted the luxury of the weaker thesis that intelligibility implies theism instead rather than intelligibility tout court requires theism then a lot more can be made out of it. One can parse this as the apparent intelligibility of the universe makes more sense given theism or that God’s mind qua mind can make certain aspects of the world intelligible. Below follow a number of arguments that bear some resemblance to those. My purpose is not to defend one or even any of these arguments, only to highlight the multiple ways that thesis could be developed more effectively and more charitably than in the one anti-sceptical wager taken from Burns.

1. Isomorphism between intellectual activity and reality. There are certain aspects of reality which bear a strong relation to intellectual activity e.g. conception, propositional attitudes et cetera, yet intuitively would still be the case were no contingent beings to exist. Examples of the above line of reasoning would be theistic modal conceptualism, variations on the argument from eternal truths and the omnisciencent knower generated by Fitch's paradox.²

2. Inference from theism to explicability. The existence of an omnibenevolent creator increases the likelihood of the Principle of Sufficient Reason holding and thus all events studied by science having an explanation. If scientific praxis requires us to assume all physical phenomena have an explanation then it also obliges us to assume the truth of theism. This is a riff on Descartes’ moral argument for divine verisimilitude but, unlike in Descartes’ argument in which divine benevolence is used to guarantee the human cognitive capacity to track truths, here it guarantees that there are explanatory truths to be tracked.

***

We move on to the second, initially more original thesis, that varieties of ‘supernaturalism’, including theism, in fact threaten the intelligibility of the universe. His first argument in favour of this is that the possibility of God should lead to rather than allay sceptical worries, as it opens the door to an omnipotent deceiver similar to Descartes’ demon. Whilst it is true theism might allow for sceptical scenarios many of these scenarios can and in fact have been raised within the framework of non-theistic ontologies—consider how the Cartesian demon thought experiment has been easily reimagined by philosophers uncomfortable with demons as brain in a vat scenarios or Matrix-like virtual reality simulations. Maitzen gives no further indication of whether these sceptical scenarios can be defeated, whether they can be defeated with a preferred naturalistic account of belief, or whether theism raises any additional sceptical worries that cannot be accounted for in said scenarios. Whereas Descartes' demon scenario relied on a per impossibile scenario the truth of theism was intended to reveal as such (the existence, whether necessary or contingent, of an omnipotent deceitful being is not seen as compossible with the existence of God), whether brain in a vat or simulation scenarios are possible appears to be a separate question from the truth of theism or atheism. 

This first foray though is mere preparation for his bolder argument that it is a specific feature of naturalism, as opposed to ‘supernaturalism’, that allows human discovery—one assumes of explanations—to be limitless in depth. In order to back up this claim Maitzen has to opt for a specific definition of naturalism that takes a stance on purpose:

As  I  construe  it,  naturalism  is  the  view  that  purposes  aren’t  fundamental:  every  being,  action,  or  whatever,  that  has  a  purpose  (a  goal,  a  telos)  arises  from  things  that  have  no  purpose. In a phrase, “Purposes don’t go all the way down.”

Most importantly in the present context, only positions that are compatible with naturalism as I’ve defined it allow human discovery to be limitless in depth. For according to supernaturalism—the  contradictory of  naturalism—purposes  do  go  all  the  way  down,  at  least  sometimes  if  not  always. According  to  supernaturalism,  at  least  some  things  have  a  purposive  explanation  that’s  fundamental.

What is meant by the claim that naturalism (and only naturalism) allows ‘human discovery to be limitless in depth’? Apparently it is that only ‘supernaturalist’ accounts will entail that some purposeful explanations are fundamental. If agency is irreducible then situations involving agents' choice cannot be explained in terms of anything more fundamental than ‘X freely wills Y’ and a list of psychological and physiological factors.

Here one faces an ambiguity in Maitzen’s definitions of naturalism and supernaturalism. When he first gave the definition it was in the context of teleological characteristics emerging out of non-teleological mechanistic ones. Is his definition a statement about ontological constitution (i.e. all intentional aspects substances show are emergent) or about the nature of explanation (i.e. that all states-of-affairs reporting purpose can be explained with reference only to more general states-of-affairs reporting mechanistic process)? The former allows for a more robust notion of emergence on which all teleological characteristics depend on more fundamental mechanistic ones for their existence though need not be explainable in terms of them (a form of non-reductive materialism).

Thankfully the next example he discusses ties in with both this and the ‘biggest choice’, which is his using the instance of him writing the above definition of naturalism as something which though genuinely purposeful can be given ‘an explanation which doesn’t bottom out with any purpose’. If this is the case and purposive explanations can be fully explained in terms of non-purposive then such an account is incompatible with truly libertarian free will, on which all the physical factors e.g. neural processes, molecular movement et cetera would only be a necessary and not a sufficient condition for his writing said definition. Maitzen has solely by definition made freedom supernatural.

Maitzen spends the rest of this part of the essay complaining that supernaturalism entails that the world contains ‘genuine magic’ as there will be events which cannot be explained in terms of deeper mechanistic processes. He gives an interesting but completely pretend investigation into how an exorcism could not qualify as a mechanistic process even involving immaterial interactions as how does one determine what factors e.g. the tone of the voice or the language, effect the process (regardless of one’s view on exorcisms what is at stake is communication, namely that certain speech acts have normative force, and that such occurrences are irreducible—as we see this renders most humdrum human and potentially animal communication truly magical).

More however the ‘magical’ aspect is even more pressing for theism as it makes ‘something magical stand at the foundation of our universe … a non-physical God who created the universe from nothing, without exploiting any laws of nature and without relying on any mechanism’. This leads to the dire problem that ‘if the universe is at bottom magical then our inescapably non-magical ways of figuring it out are to fail eventually’. According to naturalism however ‘there’s no reason in principle why science can’t make our knowledge of the origin and nature of the universe ever deeper’.

Leaving the hidden arts aside for a moment this claim and the above remark about ‘human discovery to be limitless in depth’ are bizarre. Why should naturalism entail that there is always a more fundamental scientific explanation to be sought? By definition that seems to rule out any ‘final theory’ or ultimate scientific explanation, a significant a priori limitation on scientific progress. It also dashes the hopes of his fellow naturalists who might desire that a full scientific explanation of the universe would render theism unnecessary.

Let us grant him that dubious premise that ‘there is no reason in principle why science cannot make our knowledge of the origin and workings of the universe ever deeper’. Would irreducible purpose be incompatible with this? No—in the case of God an explanation stating psychological facts and free choice might be sufficient, but God’s creative act is not an attempt at an explanation within the natural sciences and never pretended to be. In the case of humans and intelligent aliens explanations for their actions would also involve reference to free choice but this would be accompanied by facts about their physical situation, which falling squarely within the natural sciences, could ex hypothesi be limitless in explanatory depth. Since Maitzen’s account rules out ultimate explanations in science he cannot complain that irreducible agency prevents one giving an ultimate explanation in terms of science.

To return those delayed hermetic pursuits, what does it mean to say supernaturalism entails magic? Unlike his careful consideration of naturalism Maitzen nowhere gives the slightest definition of that term. The best one gets is that magical actions cannot be explained mechanistically and involve irreducibly purposive agents, a characterisation which is nearly identical to the denial of naturalism so defined by Maitzen. This means of course that the titular problem of magic resolves into a tautology: if reality depends on an irreducible agent there are actions of irreducible agents. Likewise the claim that theism makes magic fundamental to the universe really amounts to the gripe ‘if mechanistic materialism is false then a mechanistic materialistic explanation for the universe cannot be given’, something the non-materialist will be more than happy to accept.

There are many things wrong with ‘The Problem of Magic’ and I might be accused of belabouring the point by dragging it out at such lengths. Let us put aside the vacuous New Atheist rhetoric complete with lightning being explained by Thor and numerous, borderline bizarre references to existential angst in the Harry Potter series. The real problem this essay presents is not to theists but to naturalists.

Naturalism is a cluster concept. Different naturalists define it differently, accepting philosophical positions which others would take to disqualify them from the naturalist camp. We have naturalists who are Platonists, substance dualists, panpsychists and even—with due respect to Humeans—subjective idealists. If I were to offer an umbrella definition I would classify naturalism as the denial of ultimism, a thesis only slightly stronger than atheism.

What this does mean though is that the naturalist has considerable room in developing their non-theistic world-view; aside from the denial of ultimism there are no Ten Commandments of Naturalism³. Maitzen’s error, or at least disturbing move for the naturalist, is to equate irreducible agency—that is libertarian free will—with the supernatural. Now as a theist I think it nonsense that the sciences alone should explain how libertarian free will is possible but the naturalist need not, and if they are a libertarian should not, hold this view; it is in fact akin to accepting the claim that objective morality requires theism, something many naturalists are loath to admit. The hardnosed naturalist, a figure that inspires grudging respect, may laugh at all this—the first-person perspective, moral facts and free will have no place in their theory and this is made explicit—but they will be flying in the face of intuitions the vast majority of humanity use to structure their lives. The greatest problem of modern naturalism as a philosophical movement—and here we see the heritage of fairly unreconstructed scientism—is that few of its proponents are willing to develop theories which explain (naturalistically of course) as opposed to explain away such phenomena⁴. A comprehensive naturalist project should attempt to account for irreducible purpose within its remit (just as it should for instance the origin of contingent beings). This explanatory dearth is made worse by the fact that many naturalists, Maitzen included, are not ‘hardnosed’; they make frequent gestures to notions of freedom and value, assuring their audience that their theories can account for them, when in fact all they do is ‘redefine away’ by offering a replacement concept under the old semantic label. This is the fate of the reductive naturalist: they are too afraid of betraying their scientism roots to offer full-fledged explanations of such phenomena and too afraid of going against plain experience to deny there is anything there to be explained.

Is the exclusion of libertarian free will such a sacrifice for the naturalist? I say it is. Unlike God or immaterial souls, neither of which we have strong immediate intuitions about, most of the activities which constitute normal human life, within which falls philosophising, depend on explicit assumptions about choice and responsibility. Maitzen’s definition of naturalism makes any argument for irreducible agency an argument for the ‘supernatural’, which grants the theist several keen new arrows for their quiver. For instance, take Peter van Inwagen’s consequence argument that states agents cannot be responsible for their actions if the reasons for said actions are reducible to mechanistic laws of nature and events which took place before we came into existence⁵. Or Michael Huemer’s far less well known but I think compelling argument that the reasoning process itself presupposes significant free choice⁶. Neither of those nor the host of other arguments for libertarian free will make reference to God, spirits or anything that would be traditionally understood as ‘supernatural’.

To close, the great irony is that Maitzen attempts to disqualify naturalist explanation of certain ideas simply through stipulation, that is the pejorative label ‘magical’, an example of the very arbitrariness which is said to constitute the ‘problem of magic’.

Footnotes

¹. Lonergan’s intelligibility argument is presented in his magnum opus Insight. Since I am unfamiliar with his work, I will not comment upon it at any length.

². Theistic modal conceptualism is discussed (though not endorsed) by Nicholas Rescher in the essay ‘The ontology of the possible’ in Michael Loux (eds) The Possible and the Actual. The proof from eternal truths is most famously associated with Leibniz but has been discussed by many contemporary philosophers from Edward Feser to Quentin Smith. The theistic implications of Fitch’s Paradox are noted with alarm by John Bigelow in his article ‘Omnificence’ available here.

³. For a definition of ultimism and its relation to theism see J.L. Schellenberg’s ‘God for All Time: From Theism to Ultimism’ available online here.

⁴. For an interesting take on the explanatory laziness of many contemporary naturalists see Quentin Smith’s ‘The Metaphilosophy of Naturalism’ available online here.

⁵. Inwagen’s argument was originally presented in his 1983 An Essay on Free Will, but a short restatement of it in the form of a teaching aid is to be found here. Robert Kaine’s A Contemporary Introduction to Free Will presents a good summary of the argument and the controversies surrounding it.

⁶. Michael Huemer’s argument appears in the essay ‘A Proof of Free Will’ which is available online here.

Bibliography 

Bigelow, John (2005) ‘Omnificence’, Analysis, Vol. 65, No. 3
Feser, Edward (2017). Five Proofs of the Existence of God. Ignatius Press.
Huemer, Michael ‘A Proof of Free Will’. Unpublished.
Inwagen, Peter Van (1983) An Essay on Free Will. Oxford University Press.
Kane, Robert (2005) A Contemporary Introduction to Free Will. Oxford University Press.
Lonergan, Bernard (1957) Insight: A Study of Human Understanding.  University of Toronto Press.
Loux, Michael (1979) eds. The Possible and the Actual: Readings on the Metaphysics of Modality. Cornell University Press.
Maitzen, Stephen (2017) ‘The Problem of Magic’, in Kraay, Klaas eds. Does God Matter?: Essays on the Axiological Consequences of Theism. Routledge.
Schellenberg’ J. L. ‘God for All Time: From Theism to Ultimism. ’ Unpublished draft.
Smith, Quentin (2010) ‘The Metaphilosophy of Naturalism’, Philo: A Journal of Philosophy, Volume 4, No 2
Smith, Quentin (1984) ‘The Conceptualist Argument for God’s Existence’, Faith and Philosophy, vol. 11, no. 1