The Blackwell Companion to Naturalism

Russell Blackford

The Blackwell Companion to Naturalism,
edited by Kelly James Clark (Chichester, UK: Wiley-Blackwell, 2016, ISBN 978-1-118-65760-7) 534 pp. Hardcover, $195.

Philosophical naturalism—the idea that there are no supernatural entities, events, places, properties, forces, methods of obtaining knowledge, methods of achieving our desires, and so on—is undoubtedly the majority view among philosophers in the English-speaking world. Philosophical naturalism is, however, difficult to define precisely, since it’s hard to pin down the concept of the supernatural in any precise way. Furthermore, the back-cover blurb of The Blackwell Companion to Naturalism, edited by Kelly James Clark, is correct in stating that “naturalism is more often assumed than defended.” The blurb continues by promising that the Companion “offers a systematic introduction” to the topic, one “that defines, discusses, and defends philosophical naturalism.” For better or worse, though, this turns out not to be true.

First, the book is not especially systematic, reading more like a collection of diverse, thematically linked articles than a systematic effort to define, discuss, and defend naturalism of any kind. That, however, is not a serious problem. The topic is so multifarious that no single volume could cover all of its aspects and implications. The aspects of naturalism selected for discussion are inherently interesting, the contributors are of high caliber, and several chapters on specific implications of naturalist thinking make important contributions to philosophy. There’s much here that’s of value and deserves consideration in future research. From my viewpoint—with my ongoing interest in metaethics—there’s a pleasing focus on the relationship between philosophical naturalism and contemporary approaches to moral philosophy. While the Companion is out of most readers’ price range, it should certainly be acquired by university libraries. Readers hoping for a systematic approach to philosophical naturalism may be disappointed, but perhaps that’s too much to expect.

More surprising—perhaps more alarming—is the book’s overall appearance of hostility to philosophical naturalism. To make this point more clearly, I need to distinguish, as Kelly James Clark does in his introduction, between two kinds of naturalism. Each comes with variants, but the general distinction between the two is crucial. Clark prefers to call the first kind “Naturalism” (i.e., he uses a capital N), but it is also known as metaphysical naturalism, ontological naturalism, and, perhaps most commonly, philosophical naturalism. This is the idea that I mentioned in my opening paragraph: it is a thesis about what sorts of things do and (especially) do not exist. It’s a philosophically considered rejection of belief in the supernatural.

The other kind of naturalism, known as “methodological naturalism,” involves a completely different thesis. Its subject matter is not what sorts of things exist but the proper limits of scientific inquiry. Methodological naturalists do not necessarily deny the existence of supernatural entities (and events and the rest), but they claim that science should not appeal to their existence. It follows that I could be a methodological naturalist without being a philosophical naturalist. I might, for example, think that supernatural entities such as God and Satan exist but that they cannot be discovered or examined by any scientific means. I might think that they’re knowable through nonscientific means such as faith, revelation, or mystical experience.

Conversely, it’s possible to be a philosophical naturalist without embracing the limitations on science required by methodological naturalism. I might think that science is capable of framing and testing hypotheses about gods, ghosts, angels, evil spirits, and (effective, reliable) magical spells—while at the same time thinking that the current state of evidence is heavily against the existence of any of these things or of anything else commonly regarded as supernatural.

The Companion contains useful chapters by philosophical naturalists: among them are those contributed by Philip Kitcher (on how atheists should regard certain kinds of religiosity); Neil Levy (on the centuries-old free will debate); and Richard Joyce (on metaethical issues associated with naturalism and science). Missing, however, is even one chapter that provides a full-scale defense of philosophical naturalism. To say the least, that’s an odd omission. Philosophical naturalism, remember, is not a fringe idea in contemporary philosophy. It is, as Clark acknowledges, pretty clearly the opposite. Since it comes with a variety of emphases and in a range of flavors, we might have expected a number of chapters—and surely a number greater than one!—explaining and defending rival versions.

By contrast, the book is well stocked with contributors—including, it appears, its editor—who are inclined to support methodological naturalism while doubting or rejecting philosophical naturalism. Valerie Gray Hardcastle and Hans Halvorson offer elaborate defenses of the view, popular with U.S.-based science organizations but not necessarily among American philosophers, that science should confine itself to investigating natural phenomena and offering naturalistic explanations of them, with the existence, or otherwise, of supernatural phenomena insulated from scientific inquiry.

Thomas M. Crisp goes further: his chapter defends a watered-down version of Alvin Plantinga’s famous argument against philosophical naturalism. The idea is that if our evolution from earlier species occurred entirely through natural processes, with no input from a supernatural being such as God, we would not be able to trust our own cognitive faculties sufficiently to draw scientific conclusions. On this approach, we cannot even consider whether a purely naturalistic process of evolution took place without tacitly presuming that it did not take place. God, or something a bit like him, must have been involved in the process at some stage. To be fair, Hardcastle’s chapter offers mild criticism of such theistic arguments, but only in the context of her larger critique of philosophical naturalism and her defense of methodological naturalism.

Viewing all this from another angle, the Companion is equipped with numerous contributors who take an accommodationist approach to the relationship between religion and science. That is, they suggest that there is room for both in a plausible worldview. Some contributors may have anti-accommodationist tendencies: they may embrace the idea (which has its own variety of emphases and flavors) that science causes serious difficulties for religious belief. But even if some contributors are anti-accommodationists, none of them defend anti-accommodationism. The nearest thing to such a defense can be found in some halfhearted suggestions in the introductory chapter and in a few brief remarks by Kitcher—the latter in a chapter devoted to a very different topic.

I don’t object to publication of the chapters by Crisp, Halvorson, Hard­castle, and other accommodationists, and it’s useful to see these authors lay out their respective cases. Alas, Halvorson’s conception of science is so narrow and specific that it would exclude much of the activity of working scientists outside his own field of mathematical physics. But at least he provides detailed reasoning in favor of his position. What’s surprising, then, is not so much the content of particular chapters. Rather, it’s what’s missing from the book: any chapters that offer well-focused defenses of philosophical naturalism and/or anti-accommodationism.


The impression that the Companion is hostile to philosophical naturalism is strengthened by the abusive tone of a chapter by Nicholas Wolsterstorff. We can contrast this with the sympathy that Kitcher asks his fellow atheists and philosophical naturalists to grant to religious belief (at least in some of its guises).

Wolsterstorff argues that natural human rights cannot be supported on purely naturalistic grounds, though he offers an argument that they can be supported on supernatural grounds involving God’s special love for human beings. While I find this theological justification for human rights unconvincing, we can let that pass. It’s unfortunate, though, that Wolterstorff spends much of his chapter sniping at Daniel Dennett over some remarks that the latter apparently made at a workshop in 2012—remarks that were subsequently reported in the Weekly Standard, a popular magazine of conservative opinion.

As a result, the chapter reads more like the work of someone continuing a personal feud than that of an academic contributing to a scholarly volume such as one of the respected Blackwell Companions. Wolterstorff’s material would have benefited from considerable redaction, either by the author (perhaps on reflection after a good night’s sleep) or by Clark in his role as book editor. This is not just a matter of tone policing, although tone conveys meaning and is important. There’s also a degree of substantive unfairness in Wolterstorff’s approach. In effect, he accuses Dennett and other philosophical naturalists of intellectual dishonesty.

Wolsterstorff’s claim is that philosophical naturalists cannot justify the existence of human rights, understood in a very strong sense as legitimate, institution-transcendent demands to be treated in certain ways that can be made on behalf of human beings at all stages of their growth and development, and simply on the basis of their membership of the human species. He does, however, concede that it’s possible to justify the rights of fully-fledged human persons on entirely naturalistic grounds, and he thinks that it’s the latter sort of rights—rights for fully-fledged human persons—that are proclaimed in international human rights law, such as in United Nations (UN) conventions. Wolterstorff states that he “would not be surprised” if Dennett sometimes talks about (human) rights, but so what if he does? Even if Dennett has no naturalistic basis to believe in the strong sort of human rights whose existence Wolterstorff intuits, he could well have a naturalistic basis to approve of the sorts of rights posited in, say, UN conventions.

In insisting on such a strong conception of human rights, Wolterstorff evidently wants to ascribe natural rights to permanently comatose or severely mentally disabled individuals—and presumably to human embryos, fetuses, and anencephalic babies. He does not conceive of these rights to be the product merely of legal instruments, such as statutes, constitutions, and international conventions but, if I follow his account, as inhering somehow in the nature of things, transcending human laws and institutions. In a sense, that’s fine. It’s all very well for Wolsterstorff to intuit that such strong rights exist, but I frankly see no reason to share the intuition. Of course, even if there are no such rights, there may still be good reasons to treat severely mentally disabled individuals (for example) with as much kindness as possible. There may, likewise, be good reasons to protect them from unkind treatment through instruments that give them legal rights.

We have adequate secular, non-supernatural reasons to support UN conventions in particular—though we should give them only our critical support where they appear to cover too much ground, declare rights that are overly specific, or include language that would actually restrict the legal rights of individual citizens in unjustifiable ways. Instruments drafted by UN officials are not perfect or beyond criticism, but they act as something of a bulwark against tyranny, oppression, and genocide. At a general level, therefore, there is much to be said in their favor. If philosophical naturalists sometimes speak with concern about breaches of human rights, they are likely to have in mind rights that they think are justifiably enshrined in international human rights law. In that case, no intellectual dishonesty is involved. So what’s the problem?


To their credit, many contributors to the Companion recognize the vagueness of the term supernatural and consequently an element of vagueness in naturalism itself. This affects both philosophical naturalism and methodological naturalism. Thus, we might wonder exactly what kinds of entities, events, and so on should be rejected as imaginary by consistent philosophical naturalists and as out of bounds for scientific consideration by consistent methodological naturalists.

We can, of course, provide many examples of supernatural things that have been described in mythologies, religions, and folklore: the God of the Abrahamic monotheisms; other gods worshipped in various places and at various historical times; angels, ghosts, and vampires; demons and evil spirits; immortal souls capable of surviving the death of the human body; heaven, hell, and (if it comes to that) Hades and Valhalla; astral influences; large-scale miraculous events such as Noah’s flood; faith healings; magical spells and objects; and perhaps such forces or principles as karma, the Tao, and (at least on some conceptions) fate.

But it’s not immediately clear what all these have in common apart from being supernatural, whatever that means. Moreover, what should we say about the existence of numbers, irreducibly mental properties, and objective moral requirements (such as demands or prohibitions that inhere in the nature of things)? Are they natural, supernatural, or something else? Should we think of dragons, unicorns, and Bigfoot as supernatural? Perhaps they are natural if imagined to be merely animal species that we have not yet (reliably) encountered, but they are supernatural if imagined to possess extraordinary and anomalous powers.

The answer to these questions cannot be merely that things are supernatural if they are unseen by us. Nobody has seen a non-avian dinosaur or a trilobite, although we’ve seen and examined what we interpret scientifically as their fossilized remains. Yet these creatures once existed, and there was nothing supernatural about them. No one has seen the moons of Jupiter with his or her naked eyes, although the largest of them have been observed through telescopes for over four hundred years. More generally, science postulates many theoretical entities, properties, forces, etc., often corroborating their existence by effects that we can, indeed, observe (though again, often only via instruments rather than with our naked senses).

I’m inclined to doubt that there is a single straightforward difference between natural things and allegedly supernatural things. Our current West­ern concept of supernatural is likely to be a family resemblance concept grounded in our cultural history. Still, we don’t usually have too much trouble sorting out whether truth claims are, intuitively, naturalistic or supernatural ones. In that case, philosophical naturalism need not be hopelessly vague as a label for someone’s overall image of the world. But how can that be so?

The entities, events, and so on, we regard as supernatural are ordinarily unseen and (from the viewpoint of everyday common sense) anomalous. Then again, so are many of the entities and events theorized by science. However, those that we tend to classify as supernatural were originally described—or if they have a recent origin, they resemble those originally described—in prescientific, often narrative, sources such as myths, holy books, and folktales. Sometimes supernatural entities take the form of powerful disembodied intelligences such as gods and demons, but there are numerous other categories of supernatural things (embodied beings with magical powers;
astral influences, which are imagined as a kind of invisible force that acts upon us; heaven and Valhalla, which are portrayed as otherworldly places; and so on).

By contrast, the unseen or theoretical entities postulated by science, such as quarks, quasars, and Quetzalcoatlus, have come to be accepted through an incremental process of investigation involving observations, attempts at explanation, testing of hypotheses, and the integration of hypotheses into theories. Some of the entities and events discussed in scientific literature may seem less intuitively understandable than the supernatural beings whose activities are portrayed in prescientific narratives (at least superficially, gods and ghosts are easier concepts to grasp than quantum events), but their existence has been theorized and corroborated one step at a time—often including backward steps—with much checking along the way.

Thus, we can readily categorize the specimens of Quetzalcoatlus that existed and flew in the skies millions of years ago as natural though scientifically theorized entities whose remains we can now observe as fossils, whereas we categorize the Mesoamerican god Quetzalcoatl as a supernatural entity. More generally, we can be confident that science will discover many new, as-yet-unseen kinds of entities and events. But we can also be confident that these will not resemble the unseen and (commonsensically) anomalous entities portrayed in many prescientific narratives. Though it can’t be ruled out with absolute certainty, there is no reasonable prospect that a more perfected science will reinstate gods, ghosts, or goblins within its expanding and thickening web of knowledge.

If we view the problem of what counts as the supernatural in this way, we can also understand why mathematical entities such as numbers are, though not material objects or straightforwardly part of “the natural world,” not usually classified as supernatural. Why not? Perhaps because they are not among the many unseen and anomalous things that were portrayed in prescientific narratives, only to prove of no use in scientific explanations of the world’s phenomena. Unlike gods, ghosts, and magical monsters, mathematics has played an indispensable role in the advance of science. The ontological status of numbers may be puzzling—and it is grist to the mill of philosophers of mathematics and philosophically inclined mathematicians—but few of us are inclined to classify numbers as supernatural entities.

Different issues again arise in relation to objective moral demands and prohibitions, certain kinds of irreducibly mental properties, and paranormal abilities insofar as these receive pseudoscientific rationalizations (rather than being viewed as magical). Some philosophical naturalists may be inclined toward disbelief in all of these, but they fall within a gray area rather than being intuitively and clearly supernatural. Thus, objective moral requirements, theories of property dualism (in philosophy of mind), and even paranormal abilities may not be ruled out all by people who identify as philosophical naturalists. Some mysterious things are, it seems, more intuitively supernatural and more clearly regarded as imaginary within philosophical naturalism than others.

For the purposes of this review, however, my point is not to insist on my own pet understanding of the term supernatural or of how it is used in English. Rather, it’s to complain that the term, and hence the concept behind it, needs more, deeper, and more systematic attention than the contributors to the Companion provide—and especially from contributors who see themselves as philosophical naturalists or view philosophical naturalism with sympathy. Despite the Companion’s real virtues, this dimension is largely missing from it. Contributors are aware of problems with the term supernatural, as I’ve mentioned, but they don’t do much work to solve the problems.

As for methodological naturalism, I don’t see how it can ever be more than a rule of thumb for scientists, since the concept of the supernatural is vague. In any event, science often can investigate entities and events widely regarded as supernatural. If these entities and events are said to leave traces on the observable world, science can look for them. By now, however, scientists have good pragmatic reasons to shy away from framing hypotheses that involve gods, ghosts, miracles, magic, and various other familiar categories. When scientists do investigate these things, it should not be because they are actually likely to exist but because there is sometimes social utility in scrutinizing extraordinary claims.


To recap briefly, philosophical naturalism is not an unusual view among philosophers. At least in the Anglophone world, with its analytic tradition of inquiry, philosophical naturalism is a majority viewpoint that almost approaches orthodoxy. At the same time, it comes in different forms, and it is more often assumed than explicitly defended. There’s merit in publishing critiques of orthodoxies and near-orthodoxies, but a reference work such as the The Blackwell Companion to Naturalism should at least make clear why the near-orthodoxy of philosophical naturalism has seemed so intellectually compelling to so many well-informed people.

Instead, the Companion contains little in philosophical naturalism’s defense and nothing in the way of a sustained defense. It does, however, contain sustained attacks on philosophical naturalism.

There’s an appearance, therefore, of imbalance, or even of editorial bias, but we should hesitate before drawing sinister inferences about this. Almost any volume of this kind will show some sort of imbalance. That’s partly the result of editors’ conscious and unconscious biases, but these combine with other contingencies. Book editors inevitably depend to a large extent on their personal networks of contacts, and they are also at the mercy of who is prepared to contribute to a scholarly book like this (for little or no payment) and who actually delivers a manuscript within the original deadline or whatever deadlines are renegotiated during the process. In this case, it should have been possible to find competent authors who were keen to argue for the truth of philosophical naturalism, but getting a perfect balance is more easily said than done. (Speaking for myself, I would have welcomed an invitation to contribute … but I don’t expect to be on every editor’s radar.)

Despite my objections to some positions and arguments—not to mention my annoyance with Wolsterstorff’s chapter in particular—all of the contributions to the Companion have something worthwhile to offer. Even Wolsterstorff presents cogent arguments about the difficulties for philosophical naturalists in finding a plausible rationale for strong human rights (though supernaturalists also find this difficult). With its high-quality chapters and their extensive reference lists, The Blackwell Companion to Naturalism is a valuable resource—as long as readers look elsewhere for defenses of philosophical naturalism in its various forms.

Russell Blackford

Russell Blackford is a conjoint senior lecturer in philosophy at the University of Newcastle (Australia) and a regular columnist for Free Inquiry. His latest book, The Tyranny of Opinion: Conformity and the Future of Liberalism (2019), is published by Bloomsbury Academic.

“… The book is not especially systematic, reading more like a
collection of diverse, thematically linked articles than a systematic effort to define,
iscuss, and defend naturalism of any kind.”

This article is available to subscribers only.
Subscribe now or log in to read this article.