Much of contemporary philosophy engages in some way or other with science. But what kind of engagement with science is important, and what is the scope and value of that kind of engagement? Philosophers don’t spend as much time thinking about these questions as they should. And here, I do mean philosophers, not just philosophers of science. Philosophers of all subfields can and should consider the extent to which their work engages with science. Or, I suppose another way to put the point is that philosophers of all subfields can and should be philosophers of science in at least the minimal sense of having a clear view about the extent to which their work engages with science.

In what follows, I will set out three varieties of scientific engagement or, alternatively, three ways of thinking about what naturalism commits us to. In broad strokes, to be a naturalist is to be guided by science in some way. But this guidance can come in various forms. For instance, different types of naturalism may vary depending on the degree to which philosophers are constrained by this guidance: Are they required to follow it? Or is the guidance in some way optional? Different types of naturalism may also vary depending on which sciences are supposed to be doing the guiding. Just physics? Or do we include biology and chemistry and neuroscience, too? And if so, maybe we should go all the way up to economics? And, as I will focus on below, different types of naturalism can vary depending on the aspect of philosophical inquiry that is potentially impacted by the relevant guidance from science. Science can provide guidance with respect to the topics of philosophical inquiry, with respect to the content of philosophical theses, and with respect to the methodology that philosophers use. 

And as we’ll see in what follows, the project of exploring the space of naturalistic approaches to philosophy is not purely an exercise in intellectual curiosity. Once we attend to the varieties of scientific engagement and the relationships between them, we see significant consequences for how one thinks about the scope and importance of scientifically-engaged philosophy as a whole.

1. Science as a Source of Philosophical Questions

The first variety of scientific engagement involves taking science as a source of philosophical questions. In this approach, the role of the scientifically-engaged philosopher is to investigate the philosophical topics and debates that are generated by scientific inquiry.

Here’s an example: scientists spend quite a bit of time trying to figure out what the laws are. Is Schrödinger’s equation a law? Is it the only law? If not, what other laws are there? These are central questions in quantum theory. But these questions, in turn, give rise to an additional set of questions that apply far more broadly. In addition to asking what the laws are, we can ask what laws are. Do laws have to have a universal scope? Do they have to be deterministic? Can they depend on features of us humans and the particular interests we bring to inquiry? These latter questions are ones that philosophers have taken up, refined, and hotly debated. Something similar happens with concepts like chance, space, time, field, force, cause, explanation, species, life, and many, many more.

This kind of example–in which a particular concept shows up in scientific theorizing and then can be subject to philosophical scrutiny–is most obviously impactful in metaphysics, philosophy of science, philosophy of mind, and similar subfields. But science can also be a source of philosophical inspiration in epistemology and even in ethics or political philosophy. For instance, some aspects of scientific practice suggest an adherence to Occam’s Razor–in other words, a preference for simpler theories over more complex ones. What should we make of this kind of practice? Is it in good epistemological standing? What kind of simplicity is relevant? And, of course, the technological advancements that are being continually driven by scientific inquiry themselves give rise to all sorts of ethical and political questions of the schematic form: “So now we can _____, but should we?”

The above examples are very far from exhaustive. But they are sufficient to illustrate the big picture idea, which is that one way to be a naturalist is to take science as a source of inspiration. In this approach, the core commitment is that philosophers can and should engage with topics that are connected with or inspired by science or technological advancement in some way. 

As with all of the varieties of scientific engagement that we will discuss, this approach to naturalism comes in various degrees depending on whether it is a view about something that philosophers can do or a view about something that philosophers should or must do. Here, for instance, is an especially strong version of the view that science should be a source of philosophical questions:

Topic Naturalism. Philosophers should only engage with topics that are connected with or inspired by science or technological advancement in some way.

Topic naturalism might sound attractive enough at first, but further scrutiny should make you skeptical of the view, for the simple reason that the topics that are connected with or inspired by science evolve dramatically and unexpectedly over time.  

If you want an example of this kind of surprising evolution, consider the trolley problem. For quite a long time, the trolley problem was a poster child for arcane, unproductive philosophical debates. This attitude perhaps reached its apogee in the mid-2010s when, within a two-year span, both McSweeney’s and The New Yorker ran (quite funny) pieces lampooning philosophical discussion of the problem. But these days, when I teach the trolley problem, it is in the context of a course on ethical issues raised by advances in science and technology. After all, one of the many decisions we need to make when programming a self-driving car is how to think about cases like the trolley problem. If your self-driving car is going to hit five people, but can swerve to avoid them at the cost of only hitting one other person, should it swerve? (Notice that to refuse to program the car as to what to do in this situation still involves taking a stand: the car won’t swerve.)

In retrospect, it seems pretty obvious that there’s a connection between this old-school philosophical question and ongoing technological advancement. But I can attest that it really has come as a surprise. When I first taught the trolley problem, just fifteen years ago, as a graduate student at MIT, I would have loved an easy way of making the real-world implications of the topic salient to my science and technology-obsessed undergraduates. But this connection hadn’t occurred to me–nor was it part of the standard way of teaching or thinking about the trolley problem at the time. 

At any rate, what all this serves to demonstrate is that we ought to be quite humble about our ability to predict which topics are connected with or inspired by science or technological advancement in some way. So even if we are attracted to topic naturalism in principle, we should be very, very careful about using it to guide our philosophical practice.

2. Science as a Source of Philosophical Content

Here’s another way in which philosophy might engage with science: it might use the content of scientific theories to shape or constrain the content of philosophical theories. Whereas the first variety of engagement was about what questions are worthy of investigation, the second variety is about what kinds of views we should adopt in light of those questions.

Consider, for instance, the long tradition of asking philosophical questions about time. Since antiquity, philosophers across Western and non-Western traditions have wondered whether time passes, whether the future is fixed, and many other such questions. But time is also, of course, a topic of interest in physics. Indeed some of our best scientific theories, including most obviously relativity theory, include claims about the nature of time. On the variety of philosophical engagement that I have in mind here, philosophers can and should attend to the content of scientific theories of time when they are answering philosophical questions about the same topic.

Here is a relatively natural way of spelling this variety of philosophical engagement out as a type of naturalism:

Content Naturalism. Philosophers ought not put forward theories that conflict with the content of our best scientific theories.

In my recent book, Naturalism Beyond the Limits of Science, I argued that content naturalism is a widespread commitment of contemporary philosophers. One way to illustrate this is by looking at the history of philosophy, where one can find a whole range of theories about what the world is like that were once taken seriously but now aren’t discussed at all. Thales, for instance, thought that everything was made of water. We don’t consider this a view worthy of investigation in our metaphysics classes. Why not? Chemistry. Some traditional Buddhists, meanwhile, thought that the heart was the seat of consciousness. We don’t ever see this possibility discussed in articles on the philosophy of mind, however. Why not? Because it is so obviously in conflict with our current biological and anatomical understanding of the heart. The reader can no doubt also generate plenty of examples of her own. And what all of these examples illustrate, of course, is a commitment to content naturalism. We don’t take these views seriously because they so obviously conflict with science.

A commitment to content naturalism also plays a role in structuring live philosophical debates—those actively under discussion in contemporary literature. In these debates, objections that stem from our best scientific theories are taken to be the most serious kinds of objections that a theory can face. Consider, for instance, the debate over whether presentism is still a viable view in light of special relativity. Presentism says that only presently existing things exist. Special relativity says that what is present depends on your reference frame. And it’s very, very hard to get on board with thinking that what exists depends on your reference frame. The debate over what follows continues to unfold, and committed presentists have come up with all sorts of clever ways of tweaking the definition of presentism, or reinterpreting special relativity to try to avoid the conflict. What no one ever does is just say: too bad for science; I’m a presentist, and I’m sticking to it. 

There are plenty of other debates that exhibit the same structure. Biological essentialists about race go to great lengths to contend with results from genetics that seem to conflict with their view. Naive realists about perception provide surprising reinterpretations of vision science. Again, readers will no doubt have their own favorite examples. And what these kinds of examples show is that there’s a widespread commitment to content naturalism. The objection that your theory conflicts with science cannot simply be ignored.

Of course, there’s a difference between the descriptive claim that most philosophers, in fact, do adhere to content naturalism and the normative claim that they should. And arguments for the latter, normative claim are both interesting and, in my view, underexplored. Indeed, one of the most surprising aspects of the many lectures I have given on the views in Naturalism Beyond the Limits of Science is the extent to which members of the audience are willing to question the widespread commitment to content naturalism. 

I won’t attempt anything like a full-blown argument for content naturalism here. Such an attempt would need to go into such complications as whether and how often we can clearly discern the content of a scientific theory, the relationship between content naturalism and scientific realism, the distinction between science and philosophy, and whether that distinction is vague and much, much more. So I will just make the following two points, all of which should be kept in mind in any discussion of whether we ought to be content naturalists. 

First, the success of science and the robust phenomenon of scientific progress will loom large in the background of any debate over content naturalism. These ideas–that science is successful and that it makes clear progress–are themselves controversial, but they are far less controversial than the corresponding ideas as applied to philosophy. Insofar as one is willing to reject content naturalism, one will need to give reasons for thinking that despite the relative success of scientific inquiry and the relative lack of success in philosophy, we can eschew the results of the former when pursuing the latter. 

Second, it is worth pausing for a minute on what would follow if we were to reject content naturalism (or take it to be trivially satisfied). Yes, it would be easier to be, for instance, a presentist. But suppose you come across someone, a modern-day Thales, who is convinced that everything is made of water. It’s natural to think that you don’t need to engage seriously with this person’s views about what the world is like. And if you are really pressed, dismissing them is as easy as pointing to the periodic table. But if you’re not a content naturalist then things are not, in fact, so simple. Maybe everything is made of water, after all.

3. Science as a Source of Philosophical Methodology 

Here is a third variety of scientific engagement, and one that I think has the potential to be especially impactful once properly understood. Philosophers might take themselves to be constrained not (or not only) by the content of our best scientific theories but also by the methodology that produces those theories. This type of scientific engagement could, in turn, inspire the following form of naturalism

Methodological Naturalism. Philosophers should, whenever possible, make use of the same methodology that scientists use. 

Methodological naturalism might strike those who haven’t spent much time thinking about scientific methodology as highly surprising. Doesn’t the methodology of science involve laboratories, particle colliders, and telescopes? What bearing would any of that kind of empirical methodology have on philosophical debates? It’s not as though we can train a telescope out there and figure out if there are merely possible worlds, abstract objects, or objective standards of beauty.

There is quite a lot that could be said about the distinction between empirical methodology and other methodology, and about the role of straightforward empirical methodology in philosophy. But for our purposes, let’s just focus on the fact that the sort of empirical methodology described above does not exhaust scientific methodology. It is often (if not always) the case that there is more than one available theory that is empirically adequate—that is, there is often (if not always) more than one theory that correctly predicts all of the data that we have collected. Consider the choice between Ptolemaic and Copernican astronomy, or between Special Relativity as set out by Einstein and the nearby view that preserves a privileged (if undetectable) reference frame, or between any scientific theory and a theory that makes all the same claims not as claims about an objective external world but instead as claims about the programming of a computer that is stimulating your envatted brain.

When scientists choose between multiple empirically adequate theories, as they often do, they make use of extra-empirical methodology. This methodology can be understood in different ways, but the way I like best involves thinking of extra-empirical methodology as consisting of a set of extra-empirical principles. Remember Occam’s Razor, mentioned above? Here it is, as an extra-empirical principle:

Occam’s Razor. When faced with two or more empirically adequate theories, choose the one that is overall simplest. 

Now, there are all sorts of questions that we can and should ask about this extra-empirical principle, including: What the heck does “overall simplest” mean? And: Is this really a principle of scientific methodology? Let’s set all those questions aside and focus on the following fact: this kind of extra-empirical principle has the potential to play a significant role in philosophical theorizing. Maybe no telescope can determine whether merely possible worlds or abstract objects or objective standards of beauty exist, but a principle like Occam’s Razor could very well settle those debates. Indeed, this is the sort of methodology that very plausibly may turn out to impact philosophical debates from any subfield.

In this way, methodological naturalism has the potential to be the most widely impactful version of naturalism discussed here. Whereas content naturalism suggests that scientific engagement is primarily required of those philosophers who work on topics on which scientific theories directly bear (time, laws, chance, etc). Methodological naturalism suggests that any philosophical debate has the potential to be impacted by scientific engagement. Indeed methodological naturalism has the potential to dissolve the standard division between naturalistic metaphysics on the one hand and a priori metaphysics on the other. It’s all potentially naturalistic metaphysics, as far as the methodological naturalist is concerned.

In Naturalism Beyond the Limits of Science, I drew attention to the following tension: on the one hand, as I said above, there is a widespread commitment to content naturalism. On the other hand, few philosophers proceed as though they are committed to methodological naturalism–maybe they simply aren’t committed to the view, or maybe they don’t realize its potential impacts, but they certainly don’t act as though regardless of which area of philosophy they are working in, they need to consider how their views are impacted by the methodology of science.

This tension is especially interesting in part because there is a link between content naturalism and methodological naturalism that makes the status quo that the tension describes unstable. Why would you adhere to content naturalism if you don’t also adhere to methodological naturalism. If you aren’t a methodological naturalist, then you don’t think the methodology of science leads to good philosophical theories. But then why would you take yourself to be constrained by the content of our best scientific theories–those theories were produced by a methodology that, by your lights, is not good at producing philosophy! 

I won’t say any more about this tension or how to resolve it here, other than to note that either of the obvious routes—fully integrating methodological naturalism into philosophical practice (across all areas of philosophy) or giving up content naturalism—would involve a significant upsetting of the status quo. The key point for our purposes is just that the tension between the widespread acceptance of content naturalism and the lack of engagement with methodological naturalism provides a good example of a way in which a more careful investigation of what it means to do scientifically-engaged philosophy has the potential to generate real philosophical progress.

Importantly, however, this is only one example. As a closing thought, let me note that the discussion above as a whole (and even the more robust discussion in Naturalism Beyond the Limits of Science) is only the very beginning steps of what it would mean to subject the idea of scientific engagement to robust philosophical scrutiny. The three varieties of scientific engagement discussed here seem to me to be of particular interest. But I gave no reason whatsoever for thinking that these are the only varieties of scientific engagement worth investigating or adhering to, or for thinking that the particular form in which I presented them is the only or best way of understanding the potential naturalistic commitments inspired by these varieties of scientific engagement. What I do hope to have demonstrated is that further exploration of the space of naturalistic approaches to philosophy can generate interesting consequences. But this has only been an initial foray in the overall scope of that work.