Nina Emery’s recent book is grand in its aims and grounded in its approach. The goals of this book are, first, to show that embracing content naturalism commits one to also endorsing methodological naturalism, and second, to demonstrate that methodological naturalism, when applied earnestly to ongoing metaphysical debates, has surprising and significant consequences for philosophers of every stripe.
The keystone is Emery’s argument for what she calls the “content–methodology link,” which asserts that if one is a content naturalist—i.e., if one thinks we ought not accept metaphysical claims that conflict with our best scientific theories (10)—then one ought to be a methodological naturalist too. The rough idea is that if one believes our best scientific theories are true (or approximately true, or pragmatically useful, or whatever desirable epistemic feature one favors), then one must believe the methods which generated those theories are good for arriving at truth (or at something similar, as above). And if one further believes that metaphysics is a truth-guided (or epistemic-feature-X-guided) endeavor, as science is, then one should think the methods of science are appropriate in metaphysics as well.
Emery acknowledges numerous ways for philosophers to escape this hypothetical syllogism: one could deny that the methods of science are what lead our best scientific theories to be true, or withhold assent regarding the claim that metaphysics is truth-guided in the same way as science, or simply reject content naturalism altogether. One could also assert that different domains of science employ different methods which might contradict one another when applied to metaphysics. Though she contends that all sciences do in fact have a shared methodological core, Emery takes the issue of context-dependent methodology quite seriously, devoting the entire penultimate chapter of the book to fleshing out how that might affect her arguments and conclusions. But despite these many off-ramps available to her readers, Emery argues that most philosophers, given their own pre-existing commitments, will find themselves beholden to this chain of reasoning. Most think our metaphysics ought not contradict our best science, that the methods of science are what lead our best scientific theories to be true, and that metaphysics, like science, is in the business of making true claims about the world. If so, then most philosophers ought to take Emery’s argument for the content–methodology link quite seriously. Because if she is right, then we need to radically rethink our approach to deciding our metaphysical commitments. But how, exactly? Emery claims methodological naturalism will have surprising and significant consequences for all sorts of metaphysical topics, even those we commonly think of as having little to do with science. She fleshes out these consequences through three detailed case studies. In each, we recognize some methodological principle that guides science and then apply it decisively to an ongoing metaphysical debate.
These case studies are highly detailed and engaging, collectively occupying over one-third of the book’s length. I lack the space to discuss them all, so I will only mention the subject matter of each, and interested readers can investigate further for themselves.
In Case 1, Emery identifies a principle of scientific methodology she calls the pattern-explanation principle—“choose the theory that is explanatorily adequate even if that theory involves the introduction of some type of entity that is metaphysically weird or novel” (102)—that, when applied to metaphysical debates on laws of nature, encourages naturalists to endorse a governing account of laws over a Humean account of laws. For although the “governing” relation may strike us as weird or metaphysically gratuitous, it nevertheless offers a genuine explanation for why phenomena obey the laws of nature, while Humeanism does not.
In Case 2, Emery identifies a methodological principle in science she calls the principle of minimal divergence—“choose the theory that diverges least from the manifest image” (131)—which recommends that naturalists ought to reject mereological nihilism. That metaphysical view is empirically adequate, but applying the principle of minimal divergence clearly weighs against a view that suggests, contrary to all appearances, that no objects have proper parts.
In Case 3, Emery argues that a methodological principle in science she calls the excess structure principle—‘do not choose a theory that posits excess structure’ (166)—warrants the rejection of actualism in the debate over modal metaphysics. Actualism says that the only existing entities are those which actually exist; endorsing actualism is to assert that our modal reference frame (the actual one) is privileged over others (i.e., other possible worlds), and this assertion is an instance of excess structure. Since actualism requires endorsing excess structure in our modal metaphysics, we ought to reject it.
Emery makes clear that her case study analyses are meant mainly to illustrate how methodological naturalism can be consequential for metaphysical debates; even when the content of our scientific theories does not much matter to a debate, the methodology of science is nevertheless relevant to selecting among competing metaphysical stances. Emery is open to disagreement with her analyses of these particular cases—she does not claim to have fully mapped the methods of science nor their applications to issues in metaphysics. What she is not open to being wrong about is her argument for the content–methodology link. It is that, then, which will be the main focus of my discussion.
My central concern with Emery’s project is a foundational one. It is not clear what normative force is supposed to be enjoyed by certain methodological principles by virtue of the fact that they play a role in scientific reasoning. I am not convinced that the methods used to produce good science—especially the “extra-empirical principles” that Emery’s case studies focus on—are distinct from good epistemic methods more generally.
Emery briefly contends with a worry similar to mine. She notes that some philosophers see no clear difference between the content of metaphysics and the content of science, which they think undermines content naturalism since there can be no cases of metaphysical claims conflicting with scientific theories if it’s all the same damn thing anyway. Emery circumvents this worry by saying that just because the distinction between science and metaphysics is vague, “doesn’t mean that there aren’t clear cases of scientific theories—cases where there is widespread consensus that the theory in question is indeed an instance of scientific theorizing” (50). As a defense of content naturalism, this rejoinder is effective. Some theories are uncontroversially scientific, and in such cases we genuinely do think their claims ought to constrain our metaphysical theorizing. Emery provides a beautiful example of this: presentism was a live metaphysical stance that was effectively killed off by the realization that it conflicts with the content of relativity theory.
But the worry here is about the lack of distinction between the methods, not the content, of science and of metaphysics. In each case where Emery identifies some methodological practice that is prevalent in the history of science, distills it, and restates it as a general principle, it seems by the end that we have arrived at what is just a general epistemological principle, rather than a distinctly scientific one. I, for instance, implicitly draw on such principles when I decide that the barn before me really is red, and does not just appear red to me due to some improbable trick of the light. The pattern-explanation principle, the principle of minimal divergence, the excess structure principle, etc., seem to me to be core aspects not just of scientific method, but of metaphysical theorizing and even just good everyday common sense. Of course, some philosophers do not feel the need to use good common sense methods when engaging in metaphysical theorizing, whether because they are committed to a particular school of thought or for some other reason. But I’d reckon most philosophers do. The question for those readers will be: what value is added to these epistemological principles by identifying them as principles of scientific methodology? Emery’s answer is that since these principles guide scientific theorizing, and science has such an impressive track record with producing true (or useful, or whatever) theories, these epistemological principles are recommended by their history of success. That is indeed a solid reason to endorse them, but the scope of inquiry where these principles lead to success is wider than just science, and to consult only scientific evidence is to engage in an oblique form of scientism where such a thing is wholly unnecessary.
This might initially appear to be just a semantic gripe, but my concern is more than just a protest over language choice. Emery could not easily avoid my concern simply by replacing “scientific methodology” with “good epistemic practice” throughout the book—not, at least, without significantly altering her own philosophical stance. A methodological naturalism that urges us to employ scientific methods, whenever applicable, when doing metaphysics is a substantively different form of methodological naturalism than one which urges us to employ good epistemic methods, some of which can be gleaned from their successful operation in scientific inquiry, when doing metaphysics. The latter does not give scientific methods any normative force in virtue of their role in science; it just uses scientific inquiry as a wellspring of evidence concerning which epistemic methods are generally good and which are not. The former, however, does give certain methods normative force just in virtue of them being part of scientific methodology.
I will close with one further concern. Emery suggests that methodological naturalism will have surprising and significant consequences for metaphysics of all sorts, but what the impact of methodological naturalism amounts to will depend entirely on the results of its application. However, the specific conclusions Emery draws from her case studies—how she thinks the application of scientific methodology to these particular metaphysical debates settles them—are exactly those about which she is open to disagreement. There is no issue with intellectual humility, of course, and Emery emphasizes that these case studies are just meant as demonstration, just a plea for more people to take methodological naturalism seriously and to get more people working on these questions of application. But the fact remains that Emery’s claims about methodological naturalism having surprising and significant consequences are merely promissory if we cannot, as of yet, apply it in a way that avoids the risk of generating the same sort of recalcitrant disagreement that already characterizes metaphysical debates over mereology and modality. I suspect that with enough work, methodological naturalism really can be formulated in a way that allows us to settle certain metaphysical debates (for naturalists at least), but it remains to be seen whether Emery’s claims concerning the significance of methodological naturalism to metaphysics will prove themselves true. For now, what we get is a novel argument about what committing to content naturalism entails: a further commitment to methodological naturalism. Emery’s claim that that further commitment will have a significant impact on our metaphysics will require additional work—not just by Emery, but by all of us—in order to prove it.
